1.48

fix https://github.com/chrislusf/seaweedfs/issues/1176

.github/ISSUE_TEMPLATE/bug_report.md

@@ -1,16 +1,27 @@
 ---
 name: Bug report
 about: Create a report to help us improve
+title: ''
+labels: ''
+assignees: ''
 
 ---
 
 Sponsors SeaweedFS via Patreon https://www.patreon.com/seaweedfs
 
+example of a good issue report:
+https://github.com/chrislusf/seaweedfs/issues/1005
+example of a bad issue report:
+https://github.com/chrislusf/seaweedfs/issues/1008
+
 **Describe the bug**
 A clear and concise description of what the bug is.
 
 **System Setup**
-List the command line to start "weed master", "weed volume", "weed filer", "weed s3", "weed mount".
+- List the command line to start "weed master", "weed volume", "weed filer", "weed s3", "weed mount".
+- OS version
+- output of `weed version`
+- if using filer, show the content of `filer.toml`
 
 **Expected behavior**
 A clear and concise description of what you expected to happen.

.gitignore

@@ -77,4 +77,9 @@ crashlytics-build.properties
 
 test_data
 build
+target
 *.class
+other/java/hdfs/dependency-reduced-pom.xml
+
+# binary file
+weed/weed

.project

@@ -1,13 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-<projectDescription>
-	<name>seaweedfs</name>
-	<comment></comment>
-	<projects>
-	</projects>
-	<buildSpec>
-	</buildSpec>
-	<natures>
-		<nature>goclipse.goNature</nature>
-		<nature>org.eclipse.wst.common.project.facet.core.nature</nature>
-	</natures>
-</projectDescription>

.travis.yml

@@ -1,18 +1,19 @@
 sudo: false
 language: go
 go:
-- 1.10.x
-- 1.11.x
-- tip
+  - 1.11.x
+  - 1.12.x
+  - 1.13.x
 
 before_install:
-- export PATH=/home/travis/gopath/bin:$PATH
+  - export PATH=/home/travis/gopath/bin:$PATH
 
 install:
-- go get ./weed/...
+  - export CGO_ENABLED="0"
+  - go env
 
 script:
-- go test ./weed/...
+  - env GO111MODULE=on go test ./weed/...
 
 before_deploy:
   - make release
@@ -22,23 +23,26 @@ deploy:
   api_key:
     secure: ERL986+ncQ8lwAJUYDrQ8s2/FxF/cyNIwJIFCqspnWxQgGNNyokET9HapmlPSxjpFRF0q6L2WCg9OY3mSVRq4oI6hg1igOQ12KlLyN71XSJ3c8w0Ay5ho48TQ9l3f3Iu97mntBCe9l0R9pnT8wj1VI8YJxloXwUMG2yeTjA9aBI=
   file:
-  - build/linux_arm.tar.gz
-  - build/linux_arm64.tar.gz
-  - build/linux_386.tar.gz
-  - build/linux_amd64.tar.gz
-  - build/darwin_amd64.tar.gz
-  - build/windows_386.zip
-  - build/windows_amd64.zip
-  - build/freebsd_arm.tar.gz
-  - build/freebsd_amd64.tar.gz
-  - build/freebsd_386.tar.gz
-  - build/netbsd_arm.tar.gz
-  - build/netbsd_amd64.tar.gz
-  - build/netbsd_386.tar.gz
-  - build/openbsd_arm.tar.gz
-  - build/openbsd_amd64.tar.gz
-  - build/openbsd_386.tar.gz
+    - build/linux_arm.tar.gz
+    - build/linux_arm64.tar.gz
+    - build/linux_386.tar.gz
+    - build/linux_amd64.tar.gz
+    - build/linux_amd64_large_disk.tar.gz
+    - build/darwin_amd64.tar.gz
+    - build/darwin_amd64_large_disk.tar.gz
+    - build/windows_386.zip
+    - build/windows_amd64.zip
+    - build/windows_amd64_large_disk.zip
+    - build/freebsd_arm.tar.gz
+    - build/freebsd_amd64.tar.gz
+    - build/freebsd_386.tar.gz
+    - build/netbsd_arm.tar.gz
+    - build/netbsd_amd64.tar.gz
+    - build/netbsd_386.tar.gz
+    - build/openbsd_arm.tar.gz
+    - build/openbsd_amd64.tar.gz
+    - build/openbsd_386.tar.gz
   on:
     tags: true
     repo: chrislusf/seaweedfs
-    go: tip
+    go: 1.13.x

Dockerfile.go_build

@@ -1,6 +0,0 @@
-FROM cydev/go
-RUN go get github.com/chrislusf/seaweedfs/weed
-EXPOSE 8080
-EXPOSE 9333
-VOLUME /data
-ENTRYPOINT ["weed"]

Makefile

@@ -8,10 +8,13 @@ appname := weed
 
 sources := $(wildcard *.go)
 
-build = GOOS=$(1) GOARCH=$(2) go build -o build/$(appname)$(3) $(SOURCE_DIR)
+build = CGO_ENABLED=0 GOOS=$(1) GOARCH=$(2) go build -ldflags "-extldflags -static" -o build/$(appname)$(3) $(SOURCE_DIR)
 tar = cd build && tar -cvzf $(1)_$(2).tar.gz $(appname)$(3) && rm $(appname)$(3)
 zip = cd build && zip $(1)_$(2).zip $(appname)$(3) && rm $(appname)$(3)
 
+build_large = CGO_ENABLED=0 GOOS=$(1) GOARCH=$(2) go build -tags 5BytesOffset -ldflags "-extldflags -static" -o build/$(appname)$(3) $(SOURCE_DIR)
+tar_large = cd build && tar -cvzf $(1)_$(2)_large_disk.tar.gz $(appname)$(3) && rm $(appname)$(3)
+zip_large = cd build && zip $(1)_$(2)_large_disk.zip $(appname)$(3) && rm $(appname)$(3)
 
 all: build
 
@@ -24,6 +27,8 @@ clean:
 
 deps:
 	go get $(GO_FLAGS) -d $(SOURCE_DIR)
+	rm -rf /home/travis/gopath/src/github.com/coreos/etcd/vendor/golang.org/x/net/trace
+	rm -rf /home/travis/gopath/src/go.etcd.io/etcd/vendor/golang.org/x/net/trace
 
 build: deps
 	go build $(GO_FLAGS) -o $(BINARY) $(SOURCE_DIR)
@@ -32,9 +37,21 @@ linux: deps
 	mkdir -p linux
 	GOOS=linux GOARCH=amd64 go build $(GO_FLAGS) -o linux/$(BINARY) $(SOURCE_DIR)
 
-release: deps windows_build darwin_build linux_build bsd_build
+release: deps windows_build darwin_build linux_build bsd_build 5_byte_linux_build 5_byte_darwin_build 5_byte_windows_build
 
 ##### LINUX BUILDS #####
+5_byte_linux_build:
+	$(call build_large,linux,amd64,)
+	$(call tar_large,linux,amd64)
+
+5_byte_darwin_build:
+	$(call build_large,darwin,amd64,)
+	$(call tar_large,darwin,amd64)
+
+5_byte_windows_build:
+	$(call build_large,windows,amd64,.exe)
+	$(call zip_large,windows,amd64,.exe)
+
 linux_build: build/linux_arm.tar.gz build/linux_arm64.tar.gz build/linux_386.tar.gz build/linux_amd64.tar.gz
 
 build/linux_386.tar.gz: $(sources)

README.md

@@ -1,25 +1,25 @@
 # SeaweedFS
 
+
 [![Build Status](https://travis-ci.org/chrislusf/seaweedfs.svg?branch=master)](https://travis-ci.org/chrislusf/seaweedfs)
 [![GoDoc](https://godoc.org/github.com/chrislusf/seaweedfs/weed?status.svg)](https://godoc.org/github.com/chrislusf/seaweedfs/weed)
 [![Wiki](https://img.shields.io/badge/docs-wiki-blue.svg)](https://github.com/chrislusf/seaweedfs/wiki)
+[![Docker Pulls](https://img.shields.io/docker/pulls/chrislusf/seaweedfs.svg?maxAge=604800)](https://hub.docker.com/r/chrislusf/seaweedfs/)
 
 ![SeaweedFS Logo](https://raw.githubusercontent.com/chrislusf/seaweedfs/master/note/seaweedfs.png)
 
 <h2 align="center">Supporting SeaweedFS</h2>
 
-SeaweedFS is Apache-licensed open source project, independent project with its ongoing development made 
-possible entirely thanks to the support by these awesome [backers](https://github.com/chrislusf/seaweedfs/blob/master/backers.md). 
-If you'd like to grow SeaweedFS even stronger, please consider to 
-<a href="https://www.patreon.com/seaweedfs">Sponsor SeaweedFS via Patreon</a>.
-
-Platinum ($2500/month), Gold ($500/month): put your company logo on the SeaweedFS github page
-Generous Backer($50/month), Backer($10/month): put your name on the SeaweedFS backer page.
+SeaweedFS is an independent Apache-licensed open source project with its ongoing development made
+possible entirely thanks to the support of these awesome [backers](https://github.com/chrislusf/seaweedfs/blob/master/backers.md).
+If you'd like to grow SeaweedFS even stronger, please consider joining our
+<a href="https://www.patreon.com/seaweedfs">sponsors on Patreon</a>.
 
 Your support will be really appreciated by me and other supporters!
 
-<h3 align="center"><a href="https://www.patreon.com/seaweedfs">Sponsors SeaweedFS via Patreon</a></h3>
+<h3 align="center"><a href="https://www.patreon.com/seaweedfs">Sponsor SeaweedFS via Patreon</a></h3>
 
+<!--
 <h4 align="center">Platinum</h4>
 
 <p align="center">
@@ -43,16 +43,38 @@ Your support will be really appreciated by me and other supporters!
   </tbody>
 </table>
 
+-->
+
 ---
 
 
 - [Download Binaries for different platforms](https://github.com/chrislusf/seaweedfs/releases/latest)
-- [SeaweedFS on Slack](https://join.slack.com/t/seaweedfs/shared_invite/enQtMzI4MTMwMjU2MzA3LTc4MmVlYmFlNjBmZTgzZmJlYmI1MDE1YzkyNWYyZjkwZDFiM2RlMDdjNjVlNjdjYzc4NGFhZGIyYzEyMzJkYTA)
+- [SeaweedFS on Slack](https://join.slack.com/t/seaweedfs/shared_invite/enQtMzI4MTMwMjU2MzA3LTEyYzZmZWYzOGQ3MDJlZWMzYmI0OTE4OTJiZjJjODBmMzUxNmYwODg0YjY3MTNlMjBmZDQ1NzQ5NDJhZWI2ZmY)
 - [SeaweedFS Mailing List](https://groups.google.com/d/forum/seaweedfs)
 - [Wiki Documentation](https://github.com/chrislusf/seaweedfs/wiki)
+- [SeaweedFS Introduction Slides](https://www.slideshare.net/chrislusf/seaweedfs-introduction)
 
+Table of Contents
+=================
 
-## Introduction
+* [Introduction](#introduction)
+* [Features](#features)
+    * [Additional Features](#additional-features)
+    * [Filer Features](#filer-features)
+* [Example Usage](#example-usage)
+* [Architecture](#architecture)
+* [Compared to Other File Systems](#compared-to-other-file-systems)
+    * [Compared to HDFS](#compared-to-hdfs)
+    * [Compared to GlusterFS, Ceph](#compared-to-glusterfs-ceph)
+    * [Compared to GlusterFS](#compared-to-glusterfs)
+    * [Compared to Ceph](#compared-to-ceph)
+* [Dev Plan](#dev-plan)
+* [Installation Guide](#installation-guide)
+* [Disk Related Topics](#disk-related-topics)
+* [Benchmark](#Benchmark)
+* [License](#license)
+
+## Introduction ##
 
 SeaweedFS is a simple and highly scalable distributed file system. There are two objectives:
 
@@ -61,40 +83,59 @@ SeaweedFS is a simple and highly scalable distributed file system. There are two
 
 SeaweedFS started as an Object Store to handle small files efficiently. Instead of managing all file metadata in a central master, the central master only manages file volumes, and it lets these volume servers manage files and their metadata. This relieves concurrency pressure from the central master and spreads file metadata into volume servers, allowing faster file access (just one disk read operation).
 
-There is only a 40 bytes disk storage overhead for each file's metadata. It is so simple with O(1) disk read that you are welcome to challenge the performance with your actual use cases.
+There is only 40 bytes of disk storage overhead for each file's metadata. It is so simple with O(1) disk reads that you are welcome to challenge the performance with your actual use cases.
 
-SeaweedFS started by implementing [Facebook's Haystack design paper](http://www.usenix.org/event/osdi10/tech/full_papers/Beaver.pdf).
+SeaweedFS started by implementing [Facebook's Haystack design paper](http://www.usenix.org/event/osdi10/tech/full_papers/Beaver.pdf). Also, SeaweedFS implements erasure coding with ideas from [f4: Facebook’s Warm BLOB Storage System](https://www.usenix.org/system/files/conference/osdi14/osdi14-paper-muralidhar.pdf)
 
-SeaweedFS can work very well with just the object store. [[Filer]] is added later to support directories and POSIX attributes. Filer is a separate linearly-scalable stateless server with customizable metadata stores, e.g., MySql/Postgres/Redis/Cassandra/LevelDB.
+SeaweedFS can work very well with just the object store. [[Filer]] can then be added later to support directories and POSIX attributes. Filer is a separate linearly-scalable stateless server with customizable metadata stores, e.g., MySql/Postgres/Redis/Etcd/Cassandra/LevelDB/MemSql/TiDB/CockroachDB.
 
-## Additional Features
-* Can choose no replication or different replication level, rack and data center aware
-* Automatic master servers failover - no single point of failure (SPOF)
-* Automatic Gzip compression depending on file mime type
-* Automatic compaction to reclaimed disk spaces after deletion or update
+[Back to TOC](#table-of-contents)
+
+## Features ##
+
+[Back to TOC](#table-of-contents)
+
+## Additional Features ##
+* Can choose no replication or different replication levels, rack and data center aware.
+* Automatic master servers failover - no single point of failure (SPOF).
+* Automatic Gzip compression depending on file mime type.
+* Automatic compaction to reclaim disk space after deletion or update.
 * Servers in the same cluster can have different disk spaces, file systems, OS etc.
-* Adding/Removing servers does **not** cause any data re-balancing
-* Optionally fix the orientation for jpeg pictures
-* Support Etag, Accept-Range, Last-Modified, etc.
+* Adding/Removing servers does **not** cause any data re-balancing.
+* Optionally fix the orientation for jpeg pictures.
+* Support ETag, Accept-Range, Last-Modified, etc.
 * Support in-memory/leveldb/boltdb/btree mode tuning for memory/performance balance.
-## Filer Features
+* Support rebalancing the writable and readonly volumes.
+
+[Back to TOC](#table-of-contents)
+
+## Filer Features ##
 * [filer server][Filer] provide "normal" directories and files via http.
 * [mount filer][Mount] to read and write files directly as a local directory via FUSE.
 * [Amazon S3 compatible API][AmazonS3API] to access files with S3 tooling.
-* [Async Backup To Cloud][BackupToCloud] can enjoy extreme fast local access and backup to Amazon S3, Google Cloud Storage, Azure, BackBlaze.
+* [Erasure Coding for warm storage][ErasureCoding]  Rack-Aware 10.4 erasure coding reduces storage cost and increases availability.
+* [Hadoop Compatible File System][Hadoop] to access files from Hadoop/Spark/Flink/etc jobs.
+* [Async Backup To Cloud][BackupToCloud] has extremely fast local access and backups to Amazon S3, Google Cloud Storage, Azure, BackBlaze.
+* [WebDAV] access as a mapped drive on Mac and Windows, or from mobile devices.
 
 [Filer]: https://github.com/chrislusf/seaweedfs/wiki/Directories-and-Files
 [Mount]: https://github.com/chrislusf/seaweedfs/wiki/Mount
 [AmazonS3API]: https://github.com/chrislusf/seaweedfs/wiki/Amazon-S3-API
 [BackupToCloud]: https://github.com/chrislusf/seaweedfs/wiki/Backup-to-Cloud
+[Hadoop]: https://github.com/chrislusf/seaweedfs/wiki/Hadoop-Compatible-File-System
+[WebDAV]: https://github.com/chrislusf/seaweedfs/wiki/WebDAV
+[ErasureCoding]: https://github.com/chrislusf/seaweedfs/wiki/Erasure-coding-for-warm-storage
+
+[Back to TOC](#table-of-contents)
+
+## Example Usage ##
 
-## Example Usage
 By default, the master node runs on port 9333, and the volume nodes run on port 8080.
-Here I will start one master node, and two volume nodes on port 8080 and 8081. Ideally, they should be started from different machines. I just use localhost as an example.
+Let's start one master node, and two volume nodes on port 8080 and 8081. Ideally, they should be started from different machines. We'll use localhost as an example.
 
-SeaweedFS uses HTTP REST operations to write, read, delete. The responses are in JSON or JSONP format.
+SeaweedFS uses HTTP REST operations to read, write, and delete. The responses are in JSON or JSONP format.
 
-### Start Master Server
+### Start Master Server ###
 
 ```
 > ./weed master
@@ -107,7 +148,6 @@ SeaweedFS uses HTTP REST operations to write, read, delete. The responses are in
 > weed volume -dir="/tmp/data2" -max=10 -mserver="localhost:9333" -port=8081 &
 ```
 
-
 ### Write File ###
 
 To upload a file: first, send a HTTP POST, PUT, or GET request to `/dir/assign` to get an `fid` and a volume server url:
@@ -121,7 +161,7 @@ Second, to store the file content, send a HTTP multi-part POST request to `url +
 
 ```
 > curl -F file=@/home/chris/myphoto.jpg http://127.0.0.1:8080/3,01637037d6
-{"size": 43234}
+{"name":"myphoto.jpg","size":43234,"eTag":"1cc0118e"}
 ```
 
 To update, send another POST request with updated file content.
@@ -131,19 +171,20 @@ For deletion, send an HTTP DELETE request to the same `url + '/' + fid` URL:
 ```
 > curl -X DELETE http://127.0.0.1:8080/3,01637037d6
 ```
+
 ### Save File Id ###
 
-Now you can save the `fid`, 3,01637037d6 in this case, to a database field.
+Now, you can save the `fid`, 3,01637037d6 in this case, to a database field.
 
 The number 3 at the start represents a volume id. After the comma, it's one file key, 01, and a file cookie, 637037d6.
 
 The volume id is an unsigned 32-bit integer. The file key is an unsigned 64-bit integer. The file cookie is an unsigned 32-bit integer, used to prevent URL guessing.
 
-The file key and file cookie are both coded in hex. You can store the <volume id, file key, file cookie> tuple in your own format, or simply store the `fid` as string.
+The file key and file cookie are both coded in hex. You can store the <volume id, file key, file cookie> tuple in your own format, or simply store the `fid` as a string.
 
-If stored as a string, in theory, you would need 8+1+16+8=33 bytes. A char(33) would be enough, if not more than enough, since most usage would not need 2^32 volumes.
+If stored as a string, in theory, you would need 8+1+16+8=33 bytes. A char(33) would be enough, if not more than enough, since most uses will not need 2^32 volumes.
 
-If space is really a concern, you can store the file id in your own format. You would need one 4-byte integer for volume id, 8-byte long number for file key, 4-byte integer for file cookie. So 16 bytes are more than enough.
+If space is really a concern, you can store the file id in your own format. You would need one 4-byte integer for volume id, 8-byte long number for file key, and a 4-byte integer for the file cookie. So 16 bytes are more than enough.
 
 ### Read File ###
 
@@ -153,7 +194,7 @@ First look up the volume server's URLs by the file's volumeId:
 
 ```
 > curl http://localhost:9333/dir/lookup?volumeId=3
-{"locations":[{"publicUrl":"localhost:8080","url":"localhost:8080"}]}
+{"volumeId":"3","locations":[{"publicUrl":"localhost:8080","url":"localhost:8080"}]}
 ```
 
 Since (usually) there are not too many volume servers, and volumes don't move often, you can cache the results most of the time. Depending on the replication type, one volume can have multiple replica locations. Just randomly pick one location to read.
@@ -209,9 +250,9 @@ More details about replication can be found [on the wiki][Replication].
 
 You can also set the default replication strategy when starting the master server.
 
-### Allocate File Key on specific data center ###
+### Allocate File Key on Specific Data Center ###
 
-Volume servers can start with a specific data center name:
+Volume servers can be started with a specific data center name:
 
 ```
  weed volume -dir=/tmp/1 -port=8080 -dataCenter=dc1
@@ -235,13 +276,15 @@ When requesting a file key, an optional "dataCenter" parameter can limit the ass
 [feat-3]: https://github.com/chrislusf/seaweedfs/wiki/Optimization#upload-large-files
 [feat-4]: https://github.com/chrislusf/seaweedfs/wiki/Optimization#collection-as-a-simple-name-space
 
+[Back to TOC](#table-of-contents)
+
 ## Architecture ##
 
 Usually distributed file systems split each file into chunks, a central master keeps a mapping of filenames, chunk indices to chunk handles, and also which chunks each chunk server has.
 
-The main drawback is that the central master can't handle many small files efficiently, and since all read requests need to go through the chunk master, might not scale well for many concurrent users.
+The main drawback is that the central master can't handle many small files efficiently, and since all read requests need to go through the chunk master, so it might not scale well for many concurrent users.
 
-Instead of managing chunks, SeaweedFS manages data volumes in the master server. Each data volume is size 32GB, and can hold a lot of files. And each storage node can have many data volumes. So the master node only needs to store the metadata about the volumes, which is fairly small amount of data and is generally stable.
+Instead of managing chunks, SeaweedFS manages data volumes in the master server. Each data volume is 32GB in size, and can hold a lot of files. And each storage node can have many data volumes. So the master node only needs to store the metadata about the volumes, which is a fairly small amount of data and is generally stable.
 
 The actual file metadata is stored in each volume on volume servers. Since each volume server only manages metadata of files on its own disk, with only 16 bytes for each file, all file access can read file metadata just from memory and only needs one disk operation to actually read file data.
 
@@ -251,7 +294,7 @@ For comparison, consider that an xfs inode structure in Linux is 536 bytes.
 
 The architecture is fairly simple. The actual data is stored in volumes on storage nodes. One volume server can have multiple volumes, and can both support read and write access with basic authentication.
 
-All volumes are managed by a master server. The master server contains volume id to volume server mapping. This is fairly static information, and could be cached easily.
+All volumes are managed by a master server. The master server contains the volume id to volume server mapping. This is fairly static information, and can be easily cached.
 
 On each write request, the master server also generates a file key, which is a growing 64-bit unsigned integer. Since write requests are not generally as frequent as read requests, one master server should be able to handle the concurrency well.
 
@@ -259,28 +302,31 @@ On each write request, the master server also generates a file key, which is a g
 
 When a client sends a write request, the master server returns (volume id, file key, file cookie, volume node url) for the file. The client then contacts the volume node and POSTs the file content.
 
-When a client needs to read a file based on (volume id, file key, file cookie), it can ask the master server by the volume id for the (volume node url, volume node public url), or retrieve this from a cache. Then the client can GET the content, or just render the URL on web pages and let browsers fetch the content.
+When a client needs to read a file based on (volume id, file key, file cookie), it asks the master server by the volume id for the (volume node url, volume node public url), or retrieves this from a cache. Then the client can GET the content, or just render the URL on web pages and let browsers fetch the content.
 
 Please see the example for details on the write-read process.
 
 ### Storage Size ###
 
-In the current implementation, each volume can be 8x2^32 bytes (32GiB). This is because of we align content to 8 bytes. We can easily increase this to 64G, or 128G, or more, by changing 2 lines of code, at the cost of some wasted padding space due to alignment.
+In the current implementation, each volume can hold 32 gibibytes (32GiB or 8x2^32 bytes). This is because we align content to 8 bytes. We can easily increase this to 64GiB, or 128GiB, or more, by changing 2 lines of code, at the cost of some wasted padding space due to alignment.
 
-There can be 2^32 volumes. So total system size is 8 x 2^32 bytes x 2^32 = 8 x 4GiB x 4Gi = 128EiB (2^67 bytes, or 128 exbibytes).
+There can be 4 gibibytes (4GiB or 2^32 bytes) of volumes. So the total system size is 8 x 4GiB x 4GiB which is 128 exbibytes (128EiB or 2^67 bytes).
 
 Each individual file size is limited to the volume size.
 
 ### Saving memory ###
 
-All file meta information on volume server is readable from memory without disk access. Each file just takes an 16-byte map entry of <64bit key, 32bit offset, 32bit size>. Of course, each map entry has its own the space cost for the map. But usually the disk runs out before the memory does.
+All file meta information stored on an volume server is readable from memory without disk access. Each file takes just a 16-byte map entry of <64bit key, 32bit offset, 32bit size>. Of course, each map entry has its own space cost for the map. But usually the disk space runs out before the memory does.
 
+[Back to TOC](#table-of-contents)
 
 ## Compared to Other File Systems ##
 
 Most other distributed file systems seem more complicated than necessary.
 
-SeaweedFS is meant to be fast and simple, both during usage and during setup. If you do not understand how it works when you reach here, we failed! Please raise an issue with any questions or update this file with clarifications.
+SeaweedFS is meant to be fast and simple, in both setup and operation. If you do not understand how it works when you reach here, we've failed! Please raise an issue with any questions or update this file with clarifications.
+
+[Back to TOC](#table-of-contents)
 
 ### Compared to HDFS ###
 
@@ -290,6 +336,7 @@ SeaweedFS is ideal for serving relatively smaller files quickly and concurrently
 
 SeaweedFS can also store extra large files by splitting them into manageable data chunks, and store the file ids of the data chunks into a meta chunk. This is managed by "weed upload/download" tool, and the weed master or volume servers are agnostic about it.
 
+[Back to TOC](#table-of-contents)
 
 ### Compared to GlusterFS, Ceph ###
 
@@ -297,7 +344,7 @@ The architectures are mostly the same. SeaweedFS aims to store and read files fa
 
 * SeaweedFS optimizes for small files, ensuring O(1) disk seek operation, and can also handle large files.
 * SeaweedFS statically assigns a volume id for a file. Locating file content becomes just a lookup of the volume id, which can be easily cached.
-* SeaweedFS Filer metadata store can be any well-known and proven data stores, e.g., Cassandra, Redis, MySql, Postgres, etc, and is easy to customized.
+* SeaweedFS Filer metadata store can be any well-known and proven data stores, e.g., Cassandra, Redis, Etcd, MySql, Postgres, MemSql, TiDB, CockroachDB, etc, and is easy to customized.
 * SeaweedFS Volume server also communicates directly with clients via HTTP, supporting range queries, direct uploads, etc.
 
 | System         | File Meta                       | File Content Read| POSIX  | REST API | Optimized for small files |
@@ -307,25 +354,29 @@ The architectures are mostly the same. SeaweedFS aims to store and read files fa
 | GlusterFS      | hashing          |                  | FUSE, NFS          |          |                           |
 | Ceph           | hashing + rules  |                  | FUSE               | Yes      |                           |
 
+[Back to TOC](#table-of-contents)
+
 ### Compared to GlusterFS ###
 
 GlusterFS stores files, both directories and content, in configurable volumes called "bricks".
 
 GlusterFS hashes the path and filename into ids, and assigned to virtual volumes, and then mapped to "bricks".
 
+[Back to TOC](#table-of-contents)
+
 ### Compared to Ceph ###
 
 Ceph can be setup similar to SeaweedFS as a key->blob store. It is much more complicated, with the need to support layers on top of it. [Here is a more detailed comparison](https://github.com/chrislusf/seaweedfs/issues/120)
 
-SeaweedFS has a centralized master group to look up free volumes, while Ceph uses hashing and metadata servers to locate its objects. Having a centralized master makes it easy to code and manage. 
+SeaweedFS has a centralized master group to look up free volumes, while Ceph uses hashing and metadata servers to locate its objects. Having a centralized master makes it easy to code and manage.
 
-Same as SeaweedFS, Ceph is also based on a object store RADOS. Ceph is rather complicated with mixed reviews.
+Same as SeaweedFS, Ceph is also based on the object store RADOS. Ceph is rather complicated with mixed reviews.
 
 Ceph uses CRUSH hashing to automatically manage the data placement. SeaweedFS places data by assigned volumes.
 
 SeaweedFS is optimized for small files. Small files are stored as one continuous block of content, with at most 8 unused bytes between files. Small file access is O(1) disk read.
 
-SeaweedFS Filer uses off-the-shelf stores, such as MySql, Postgres, Redis, Cassandra, to manage file directories. There are proven, scalable, and easier to manage.
+SeaweedFS Filer uses off-the-shelf stores, such as MySql, Postgres, Redis, Etcd, Cassandra, MemSql, TiDB, CockroachCB, to manage file directories. There are proven, scalable, and easier to manage.
 
 | SeaweedFS         | comparable to Ceph | advantage |
 | -------------  | ------------- | ---------------- |
@@ -333,53 +384,67 @@ SeaweedFS Filer uses off-the-shelf stores, such as MySql, Postgres, Redis, Cassa
 | Volume  | OSD | optimized for small files |
 | Filer  | Ceph FS | linearly scalable, Customizable, O(1) or O(logN) |
 
+[Back to TOC](#table-of-contents)
 
-## Dev plan ##
+## Dev Plan ##
 
 More tools and documentation, on how to maintain and scale the system. For example, how to move volumes, automatically balancing data, how to grow volumes, how to check system status, etc.
 Other key features include: Erasure Encoding, JWT security.
 
-This is a super exciting project! And I need helpers and [support](https://www.patreon.com/seaweedfs)!
+This is a super exciting project! And we need helpers and [support](https://www.patreon.com/seaweedfs)!
 
+BTW, We suggest run the code style check script `util/gostd` before you push your branch to remote, it will make SeaweedFS easy to review, maintain and develop:
 
-## Installation guide for users who are not familiar with golang
+```
+$ ./util/gostd
+```
 
-step 1: install go on your machine and setup the environment by following the instructions from the following link:
+[Back to TOC](#table-of-contents)
+
+## Installation Guide ##
+
+> Installation guide for users who are not familiar with golang
+
+Step 1: install go on your machine and setup the environment by following the instructions at:
 
 https://golang.org/doc/install
 
 make sure you set up your $GOPATH
 
 
-step 2: also you may need to install Mercurial by following the instructions below
+Step 2: also you may need to install Mercurial by following the instructions at:
 
 http://mercurial.selenic.com/downloads
 
-step 3: download, compile, and install the project by executing the following command
+Step 3: download, compile, and install the project by executing the following command
 
 ```bash
 go get github.com/chrislusf/seaweedfs/weed
 ```
 
-once this is done, you should see the executable "weed" under `$GOPATH/bin`
+Once this is done, you will find the executable "weed" in your `$GOPATH/bin` directory
 
-step 4: after you modify your code locally, you could start a local build by calling `go install` under 
+Step 4: after you modify your code locally, you could start a local build by calling `go install` under
 
 ```
 $GOPATH/src/github.com/chrislusf/seaweedfs/weed
 ```
 
-## Disk Related topics ##
+[Back to TOC](#table-of-contents)
+
+## Disk Related Topics ##
 
 ### Hard Drive Performance ###
 
-When testing read performance on SeaweedFS, it basically becomes performance test for your hard drive's random read speed. Hard Drive usually get 100MB/s~200MB/s.
+When testing read performance on SeaweedFS, it basically becomes a performance test of your hard drive's random read speed. Hard drives usually get 100MB/s~200MB/s.
 
-### Solid State Disk
+### Solid State Disk ###
 
 To modify or delete small files, SSD must delete a whole block at a time, and move content in existing blocks to a new block. SSD is fast when brand new, but will get fragmented over time and you have to garbage collect, compacting blocks. SeaweedFS is friendly to SSD since it is append-only. Deletion and compaction are done on volume level in the background, not slowing reading and not causing fragmentation.
 
-## Benchmark
+[Back to TOC](#table-of-contents)
+
+## Benchmark ##
 
 My Own Unscientific Single Machine Results on Mac Book with Solid State Disk, CPU: 1 Intel Core i7 2.6GHz.
 
@@ -432,8 +497,9 @@ Percentage of the requests served within a certain time (ms)
   100%     20.7 ms
 ```
 
+[Back to TOC](#table-of-contents)
 
-## License
+## License ##
 
 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.
@@ -447,7 +513,8 @@ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License.
 
+[Back to TOC](#table-of-contents)
 
-## Stargazers over time
+## Stargazers over time ##
 
 [![Stargazers over time](https://starcharts.herokuapp.com/chrislusf/seaweedfs.svg)](https://starcharts.herokuapp.com/chrislusf/seaweedfs)

backers.md

@@ -4,6 +4,7 @@
 
 <h2 align="center">Generous Backers ($50+)</h2>
 
+- [4Sight Imaging](https://www.4sightimaging.com/)
 - [Evercam Camera Management Software](https://evercam.io/)
 
 <h2 align="center">Backers</h2>

docker/Dockerfile

@@ -1,9 +1,20 @@
-FROM frolvlad/alpine-glibc:alpine-3.5
+FROM frolvlad/alpine-glibc
 
+# Supercronic install settings
+ENV SUPERCRONIC_URL=https://github.com/aptible/supercronic/releases/download/v0.1.8/supercronic-linux-amd64 \
+    SUPERCRONIC=supercronic-linux-amd64 \
+    SUPERCRONIC_SHA1SUM=be43e64c45acd6ec4fce5831e03759c89676a0ea
+
+# Install SeaweedFS and Supercronic ( for cron job mode )
 # Tried to use curl only (curl -o /tmp/linux_amd64.tar.gz ...), however it turned out that the following tar command failed with "gzip: stdin: not in gzip format"
 RUN apk add --no-cache --virtual build-dependencies --update wget curl ca-certificates && \
     wget -P /tmp https://github.com/$(curl -s -L https://github.com/chrislusf/seaweedfs/releases/latest | egrep -o 'chrislusf/seaweedfs/releases/download/.*/linux_amd64.tar.gz') && \
     tar -C /usr/bin/ -xzvf /tmp/linux_amd64.tar.gz && \
+    curl -fsSLO "$SUPERCRONIC_URL" && \
+    echo "${SUPERCRONIC_SHA1SUM}  ${SUPERCRONIC}" | sha1sum -c - && \
+    chmod +x "$SUPERCRONIC" && \
+    mv "$SUPERCRONIC" "/usr/local/bin/${SUPERCRONIC}" && \
+    ln -s "/usr/local/bin/${SUPERCRONIC}" /usr/local/bin/supercronic && \
     apk del build-dependencies && \
     rm -rf /tmp/*
 
@@ -15,8 +26,14 @@ EXPOSE 8080
 EXPOSE 18888
 # filer server http port
 EXPOSE 8888
-# master server shared gprc+http port
+# master server shared gprc port
+EXPOSE 19333
+# master server shared http port
 EXPOSE 9333
+# s3 server http port
+EXPOSE 8333
+
+RUN mkdir -p /data/filerldb2
 
 VOLUME /data
 

docker/Dockerfile.go_build

@@ -0,0 +1,29 @@
+FROM golang:latest
+RUN go get github.com/chrislusf/seaweedfs/weed
+
+# volume server gprc port
+EXPOSE 18080
+# volume server http port
+EXPOSE 8080
+# filer server gprc port
+EXPOSE 18888
+# filer server http port
+EXPOSE 8888
+# master server shared gprc port
+EXPOSE 19333
+# master server shared http port
+EXPOSE 9333
+# s3 server http port
+EXPOSE 8333
+
+RUN mkdir -p /data/filerldb2
+
+VOLUME /data
+
+RUN mkdir -p /etc/seaweedfs
+RUN cp /go/src/github.com/chrislusf/seaweedfs/docker/filer.toml /etc/seaweedfs/filer.toml
+RUN cp /go/src/github.com/chrislusf/seaweedfs/docker/entrypoint.sh /entrypoint.sh
+RUN chmod +x /entrypoint.sh
+RUN cp /go/bin/weed /usr/bin/
+
+ENTRYPOINT ["/entrypoint.sh"]

docker/README.md

@@ -0,0 +1,21 @@
+# Docker
+
+
+## Try it out
+
+```bash
+
+wget https://raw.githubusercontent.com/chrislusf/seaweedfs/master/docker/seaweedfs-compose.yml
+
+docker-compose -f seaweedfs-compose.yml -p seaweedfs up
+
+```
+
+## Development
+
+```bash
+cd $GOPATH/src/github.com/chrislusf/seaweedfs/docker
+
+docker-compose -f dev-compose.yml -p seaweedfs up
+
+```

docker/dev-compose.yml

@@ -0,0 +1,43 @@
+version: '2'
+
+services:
+  master:
+    build:
+      context: .
+      dockerfile: Dockerfile.go_build
+    ports:
+    - 9333:9333
+    - 19333:19333
+    command: "master -ip=master"
+  volume:
+    build:
+      context: .
+      dockerfile: Dockerfile.go_build
+    ports:
+    - 8080:8080
+    - 18080:18080
+    command: 'volume -max=5 -mserver="master:9333" -port=8080'
+    depends_on:
+    - master
+  filer:
+    build:
+      context: .
+      dockerfile: Dockerfile.go_build
+    ports:
+    - 8888:8888
+    - 18888:18888
+    command: 'filer -master="master:9333"'
+    depends_on:
+    - master
+    - volume
+  s3:
+    build:
+      context: .
+      dockerfile: Dockerfile.go_build
+    ports:
+    - 8333:8333
+    command: 's3 -filer="filer:8888"'
+    depends_on:
+    - master
+    - volume
+    - filer

docker/docker-compose.yml

@@ -1,40 +0,0 @@
-version: '2'
-
-services:
-  master:
-    #image: chrislusf/seaweedfs # use a remote image
-    build: . # build our container from the local Dockerfile
-    ports:
-    - 9333:9333
-    command: "master"
-    networks:
-      default:
-        aliases:
-        - seaweed_master
-  volume:
-    #image: chrislusf/seaweedfs # use a remote image
-    build: . # build our container from the local Dockerfile
-    ports:
-    - 8080:8080
-    - 18080:18080
-    command: 'volume -max=5 -mserver="master:9333" -port=8080'
-    depends_on:
-    - master
-    networks:
-      default:
-        aliases:
-        - seaweed_volume
-  filer:
-    #image: chrislusf/seaweedfs # use a remote image
-    build: . # build our container from the local Dockerfile
-    ports:
-    - 8888:8888
-    - 18888:18888
-    command: 'filer -master="master:9333"'
-    depends_on:
-    - master
-    - volume
-    networks:
-      default:
-        aliases:
-        - seaweed_filer

docker/entrypoint.sh

@@ -3,7 +3,7 @@
 case "$1" in
 
   'master')
-  	ARGS="-ip `hostname -i` -mdir /data"
+  	ARGS="-mdir /data"
 	# Is this instance linked with an other master? (Docker commandline "--link master1:master")
 	if [ -n "$MASTER_PORT_9333_TCP_ADDR" ] ; then
 		ARGS="$ARGS -peers=$MASTER_PORT_9333_TCP_ADDR:$MASTER_PORT_9333_TCP_PORT"
@@ -29,13 +29,31 @@ case "$1" in
   	;;
 
   'filer')
-  	ARGS="-ip `hostname -i`"
+  	ARGS=""
   	if [ -n "$MASTER_PORT_9333_TCP_ADDR" ] ; then
 		ARGS="$ARGS -master=$MASTER_PORT_9333_TCP_ADDR:$MASTER_PORT_9333_TCP_PORT"
 	fi
   	exec /usr/bin/weed $@ $ARGS
 	;;
 
+  's3')
+  	ARGS="-domainName=$S3_DOMAIN_NAME -key.file=$S3_KEY_FILE -cert.file=$S3_CERT_FILE"
+  	if [ -n "$FILER_PORT_8888_TCP_ADDR" ] ; then
+		ARGS="$ARGS -filer=$FILER_PORT_8888_TCP_ADDR:$FILER_PORT_8888_TCP_PORT"
+	fi
+  	exec /usr/bin/weed $@ $ARGS
+	;;
+
+  'cronjob')
+	MASTER=${WEED_MASTER-localhost:9333}
+	FIX_REPLICATION_CRON_SCHEDULE=${CRON_SCHEDULE-*/7 * * * * *}
+	echo "$FIX_REPLICATION_CRON_SCHEDULE" 'echo "volume.fix.replication" | weed shell -master='$MASTER > /crontab
+	BALANCING_CRON_SCHEDULE=${CRON_SCHEDULE-25 * * * * *}
+	echo "$BALANCING_CRON_SCHEDULE" 'echo "volume.balance -c ALL -force" | weed shell -master='$MASTER >> /crontab
+	echo "Running Crontab:"
+	cat /crontab
+	exec supercronic /crontab
+	;;
   *)
   	exec /usr/bin/weed $@
 	;;

docker/filer.toml

@@ -1,3 +1,3 @@
-[leveldb]
+[leveldb2]
 enabled = true
-dir = "."
+dir = "/data/filerldb2"

docker/seaweedfs-compose.yml

@@ -0,0 +1,47 @@
+version: '2'
+
+services:
+  master:
+    image: chrislusf/seaweedfs # use a remote image
+    ports:
+    - 9333:9333
+    - 19333:19333
+    command: "master -ip=master"
+  volume:
+    image: chrislusf/seaweedfs # use a remote image
+    ports:
+    - 8080:8080
+    - 18080:18080
+    command: 'volume -max=15 -mserver="master:9333" -port=8080'
+    depends_on:
+    - master
+  filer:
+    image: chrislusf/seaweedfs # use a remote image
+    ports:
+    - 8888:8888
+    - 18888:18888
+    command: 'filer -master="master:9333"'
+    tty: true
+    stdin_open: true
+    depends_on:
+    - master
+    - volume
+  cronjob:
+    image: chrislusf/seaweedfs # use a remote image
+    command: 'cronjob'
+    environment:
+      # Run re-replication every 2 minutes
+      CRON_SCHEDULE: '*/2 * * * * *' # Default: '*/5 * * * * *'
+      WEED_MASTER: master:9333 # Default: localhost:9333
+    depends_on:
+    - master
+    - volume
+  s3:
+    image: chrislusf/seaweedfs # use a remote image
+    ports:
+    - 8333:8333
+    command: 's3 -filer="filer:8888"'
+    depends_on:
+    - master
+    - volume
+    - filer

go.mod

@@ -0,0 +1,99 @@
+module github.com/chrislusf/seaweedfs
+
+go 1.12
+
+require (
+	cloud.google.com/go v0.44.3
+	github.com/Azure/azure-pipeline-go v0.2.2 // indirect
+	github.com/Azure/azure-storage-blob-go v0.8.0
+	github.com/DataDog/zstd v1.4.1 // indirect
+	github.com/Shopify/sarama v1.23.1
+	github.com/aws/aws-sdk-go v1.23.13
+	github.com/chrislusf/raft v0.0.0-20190225081310-10d6e2182d92
+	github.com/coreos/etcd v3.3.15+incompatible // indirect
+	github.com/coreos/go-semver v0.3.0 // indirect
+	github.com/coreos/go-systemd v0.0.0-20190719114852-fd7a80b32e1f // indirect
+	github.com/dgrijalva/jwt-go v3.2.0+incompatible
+	github.com/disintegration/imaging v1.6.1
+	github.com/dustin/go-humanize v1.0.0
+	github.com/eapache/go-resiliency v1.2.0 // indirect
+	github.com/facebookgo/clock v0.0.0-20150410010913-600d898af40a
+	github.com/facebookgo/stats v0.0.0-20151006221625-1b76add642e4
+	github.com/frankban/quicktest v1.7.2 // indirect
+	github.com/gabriel-vasile/mimetype v1.0.0
+	github.com/go-redis/redis v6.15.2+incompatible
+	github.com/go-sql-driver/mysql v1.4.1
+	github.com/gocql/gocql v0.0.0-20190829130954-e163eff7a8c6
+	github.com/gogo/protobuf v1.2.2-0.20190730201129-28a6bbf47e48 // indirect
+	github.com/golang/protobuf v1.3.2
+	github.com/google/btree v1.0.0
+	github.com/google/uuid v1.1.1
+	github.com/gorilla/mux v1.7.3
+	github.com/gorilla/websocket v1.4.1 // indirect
+	github.com/grpc-ecosystem/grpc-gateway v1.11.0 // indirect
+	github.com/hashicorp/golang-lru v0.5.3 // indirect
+	github.com/jacobsa/daemonize v0.0.0-20160101105449-e460293e890f
+	github.com/jcmturner/gofork v1.0.0 // indirect
+	github.com/karlseguin/ccache v2.0.3+incompatible
+	github.com/karlseguin/expect v1.0.1 // indirect
+	github.com/klauspost/cpuid v1.2.1 // indirect
+	github.com/klauspost/crc32 v1.2.0
+	github.com/klauspost/reedsolomon v1.9.2
+	github.com/konsorten/go-windows-terminal-sequences v1.0.2 // indirect
+	github.com/kurin/blazer v0.5.3
+	github.com/lib/pq v1.2.0
+	github.com/magiconair/properties v1.8.1 // indirect
+	github.com/mattn/go-ieproxy v0.0.0-20190805055040-f9202b1cfdeb // indirect
+	github.com/mattn/go-runewidth v0.0.4 // indirect
+	github.com/nats-io/nats-server/v2 v2.0.4 // indirect
+	github.com/onsi/ginkgo v1.10.1 // indirect
+	github.com/onsi/gomega v1.7.0 // indirect
+	github.com/opentracing/opentracing-go v1.1.0 // indirect
+	github.com/pelletier/go-toml v1.4.0 // indirect
+	github.com/peterh/liner v1.1.0
+	github.com/pierrec/lz4 v2.2.7+incompatible // indirect
+	github.com/pingcap/kvproto v0.0.0-20191022073741-81b2a2d9e986 // indirect
+	github.com/pingcap/tidb v1.1.0-beta.0.20191023070859-58fc7d44f73b
+	github.com/pingcap/tipb v0.0.0-20191022094114-a2e8c3fa634b // indirect
+	github.com/prometheus/client_golang v1.1.0
+	github.com/prometheus/client_model v0.0.0-20190812154241-14fe0d1b01d4 // indirect
+	github.com/prometheus/procfs v0.0.4 // indirect
+	github.com/rakyll/statik v0.1.6
+	github.com/rcrowley/go-metrics v0.0.0-20190826022208-cac0b30c2563 // indirect
+	github.com/remyoudompheng/bigfft v0.0.0-20190728182440-6a916e37a237 // indirect
+	github.com/rwcarlsen/goexif v0.0.0-20190401172101-9e8deecbddbd
+	github.com/seaweedfs/fuse v0.0.0-20190510212405-310228904eff
+	github.com/sirupsen/logrus v1.4.2 // indirect
+	github.com/spaolacci/murmur3 v1.1.0 // indirect
+	github.com/spf13/afero v1.2.2 // indirect
+	github.com/spf13/jwalterweatherman v1.1.0 // indirect
+	github.com/spf13/viper v1.4.0
+	github.com/streadway/amqp v0.0.0-20190827072141-edfb9018d271 // indirect
+	github.com/stretchr/testify v1.4.0 // indirect
+	github.com/syndtr/goleveldb v1.0.0
+	github.com/tidwall/gjson v1.3.2
+	github.com/tidwall/match v1.0.1
+	github.com/uber-go/atomic v1.4.0 // indirect
+	github.com/uber/jaeger-client-go v2.17.0+incompatible // indirect
+	github.com/uber/jaeger-lib v2.0.0+incompatible // indirect
+	github.com/willf/bitset v1.1.10 // indirect
+	github.com/willf/bloom v2.0.3+incompatible
+	github.com/wsxiaoys/terminal v0.0.0-20160513160801-0940f3fc43a0 // indirect
+	go.etcd.io/etcd v3.3.15+incompatible
+	gocloud.dev v0.16.0
+	gocloud.dev/pubsub/natspubsub v0.16.0
+	gocloud.dev/pubsub/rabbitpubsub v0.16.0
+	golang.org/x/image v0.0.0-20190829233526-b3c06291d021 // indirect
+	golang.org/x/net v0.0.0-20190909003024-a7b16738d86b
+	golang.org/x/sys v0.0.0-20190910064555-bbd175535a8b
+	golang.org/x/tools v0.0.0-20190911022129-16c5e0f7d110
+	google.golang.org/api v0.9.0
+	google.golang.org/appengine v1.6.2 // indirect
+	google.golang.org/grpc v1.23.0
+	gopkg.in/jcmturner/goidentity.v3 v3.0.0 // indirect
+	gopkg.in/jcmturner/gokrb5.v7 v7.3.0 // indirect
+	gopkg.in/karlseguin/expect.v1 v1.0.1 // indirect
+	sigs.k8s.io/yaml v1.1.0 // indirect
+)
+
+replace github.com/satori/go.uuid v1.2.0 => github.com/satori/go.uuid v0.0.0-20181028125025-b2ce2384e17b

go.sum

@@ -0,0 +1,776 @@
+cloud.google.com/go v0.26.0/go.mod h1:aQUYkXzVsufM+DwF1aE+0xfcU+56JwCaLick0ClmMTw=
+cloud.google.com/go v0.34.0/go.mod h1:aQUYkXzVsufM+DwF1aE+0xfcU+56JwCaLick0ClmMTw=
+cloud.google.com/go v0.38.0/go.mod h1:990N+gfupTy94rShfmMCWGDn0LpTmnzTp2qbd1dvSRU=
+cloud.google.com/go v0.39.0/go.mod h1:rVLT6fkc8chs9sfPtFc1SBH6em7n+ZoXaG+87tDISts=
+cloud.google.com/go v0.44.1/go.mod h1:iSa0KzasP4Uvy3f1mN/7PiObzGgflwredwwASm/v6AU=
+cloud.google.com/go v0.44.3 h1:0sMegbmn/8uTwpNkB0q9cLEpZ2W5a6kl+wtBQgPWBJQ=
+cloud.google.com/go v0.44.3/go.mod h1:60680Gw3Yr4ikxnPRS/oxxkBccT6SA1yMk63TGekxKY=
+cloud.google.com/go/datastore v1.0.0/go.mod h1:LXYbyblFSglQ5pkeyhO+Qmw7ukd3C+pD7TKLgZqpHYE=
+contrib.go.opencensus.io/exporter/aws v0.0.0-20181029163544-2befc13012d0/go.mod h1:uu1P0UCM/6RbsMrgPa98ll8ZcHM858i/AD06a9aLRCA=
+contrib.go.opencensus.io/exporter/ocagent v0.5.0 h1:TKXjQSRS0/cCDrP7KvkgU6SmILtF/yV2TOs/02K/WZQ=
+contrib.go.opencensus.io/exporter/ocagent v0.5.0/go.mod h1:ImxhfLRpxoYiSq891pBrLVhN+qmP8BTVvdH2YLs7Gl0=
+contrib.go.opencensus.io/exporter/stackdriver v0.12.1/go.mod h1:iwB6wGarfphGGe/e5CWqyUk/cLzKnWsOKPVW3no6OTw=
+contrib.go.opencensus.io/integrations/ocsql v0.1.4/go.mod h1:8DsSdjz3F+APR+0z0WkU1aRorQCFfRxvqjUUPMbF3fE=
+contrib.go.opencensus.io/resource v0.1.1/go.mod h1:F361eGI91LCmW1I/Saf+rX0+OFcigGlFvXwEGEnkRLA=
+github.com/Azure/azure-amqp-common-go/v2 v2.1.0/go.mod h1:R8rea+gJRuJR6QxTir/XuEd+YuKoUiazDC/N96FiDEU=
+github.com/Azure/azure-pipeline-go v0.1.8/go.mod h1:XA1kFWRVhSK+KNFiOhfv83Fv8L9achrP7OxIzeTn1Yg=
+github.com/Azure/azure-pipeline-go v0.1.9/go.mod h1:XA1kFWRVhSK+KNFiOhfv83Fv8L9achrP7OxIzeTn1Yg=
+github.com/Azure/azure-pipeline-go v0.2.1 h1:OLBdZJ3yvOn2MezlWvbrBMTEUQC72zAftRZOMdj5HYo=
+github.com/Azure/azure-pipeline-go v0.2.1/go.mod h1:UGSo8XybXnIGZ3epmeBw7Jdz+HiUVpqIlpz/HKHylF4=
+github.com/Azure/azure-pipeline-go v0.2.2 h1:6oiIS9yaG6XCCzhgAgKFfIWyo4LLCiDhZot6ltoThhY=
+github.com/Azure/azure-pipeline-go v0.2.2/go.mod h1:4rQ/NZncSvGqNkkOsNpOU1tgoNuIlp9AfUH5G1tvCHc=
+github.com/Azure/azure-sdk-for-go v29.0.0+incompatible/go.mod h1:9XXNKU+eRnpl9moKnB4QOLf1HestfXbmab5FXxiDBjc=
+github.com/Azure/azure-sdk-for-go v30.1.0+incompatible/go.mod h1:9XXNKU+eRnpl9moKnB4QOLf1HestfXbmab5FXxiDBjc=
+github.com/Azure/azure-service-bus-go v0.9.1/go.mod h1:yzBx6/BUGfjfeqbRZny9AQIbIe3AcV9WZbAdpkoXOa0=
+github.com/Azure/azure-storage-blob-go v0.6.0/go.mod h1:oGfmITT1V6x//CswqY2gtAHND+xIP64/qL7a5QJix0Y=
+github.com/Azure/azure-storage-blob-go v0.8.0 h1:53qhf0Oxa0nOjgbDeeYPUeyiNmafAFEY95rZLK0Tj6o=
+github.com/Azure/azure-storage-blob-go v0.8.0/go.mod h1:lPI3aLPpuLTeUwh1sViKXFxwl2B6teiRqI0deQUvsw0=
+github.com/Azure/go-autorest v12.0.0+incompatible h1:N+VqClcomLGD/sHb3smbSYYtNMgKpVV3Cd5r5i8z6bQ=
+github.com/Azure/go-autorest v12.0.0+incompatible/go.mod h1:r+4oMnoxhatjLLJ6zxSWATqVooLgysK6ZNox3g/xq24=
+github.com/BurntSushi/toml v0.3.1 h1:WXkYYl6Yr3qBf1K79EBnL4mak0OimBfB0XUf9Vl28OQ=
+github.com/BurntSushi/toml v0.3.1/go.mod h1:xHWCNGjB5oqiDr8zfno3MHue2Ht5sIBksp03qcyfWMU=
+github.com/BurntSushi/xgb v0.0.0-20160522181843-27f122750802/go.mod h1:IVnqGOEym/WlBOVXweHU+Q+/VP0lqqI8lqeDx9IjBqo=
+github.com/DataDog/zstd v1.3.6-0.20190409195224-796139022798 h1:2T/jmrHeTezcCM58lvEQXs0UpQJCo5SoGAcg+mbSTIg=
+github.com/DataDog/zstd v1.3.6-0.20190409195224-796139022798/go.mod h1:1jcaCB/ufaK+sKp1NBhlGmpz41jOoPQ35bpF36t7BBo=
+github.com/DataDog/zstd v1.4.1 h1:3oxKN3wbHibqx897utPC2LTQU4J+IHWWJO+glkAkpFM=
+github.com/DataDog/zstd v1.4.1/go.mod h1:1jcaCB/ufaK+sKp1NBhlGmpz41jOoPQ35bpF36t7BBo=
+github.com/GoogleCloudPlatform/cloudsql-proxy v0.0.0-20190605020000-c4ba1fdf4d36/go.mod h1:aJ4qN3TfrelA6NZ6AXsXRfmEVaYin3EDbSPJrKS8OXo=
+github.com/OneOfOne/xxhash v1.2.2/go.mod h1:HSdplMjZKSmBqAxg5vPj2TmRDmfkzw+cTzAElWljhcU=
+github.com/Shopify/sarama v1.23.1 h1:XxJBCZEoWJtoWjf/xRbmGUpAmTZGnuuF0ON0EvxxBrs=
+github.com/Shopify/sarama v1.23.1/go.mod h1:XLH1GYJnLVE0XCr6KdJGVJRTwY30moWNJ4sERjXX6fs=
+github.com/Shopify/toxiproxy v2.1.4+incompatible h1:TKdv8HiTLgE5wdJuEML90aBgNWsokNbMijUGhmcoBJc=
+github.com/Shopify/toxiproxy v2.1.4+incompatible/go.mod h1:OXgGpZ6Cli1/URJOF1DMxUHB2q5Ap20/P/eIdh4G0pI=
+github.com/StackExchange/wmi v0.0.0-20180725035823-b12b22c5341f h1:5ZfJxyXo8KyX8DgGXC5B7ILL8y51fci/qYz2B4j8iLY=
+github.com/StackExchange/wmi v0.0.0-20180725035823-b12b22c5341f/go.mod h1:3eOhrUMpNV+6aFIbp5/iudMxNCF27Vw2OZgy4xEx0Fg=
+github.com/alecthomas/template v0.0.0-20160405071501-a0175ee3bccc/go.mod h1:LOuyumcjzFXgccqObfd/Ljyb9UuFJ6TxHnclSeseNhc=
+github.com/alecthomas/units v0.0.0-20151022065526-2efee857e7cf/go.mod h1:ybxpYRFXyAe+OPACYpWeL0wqObRcbAqCMya13uyzqw0=
+github.com/armon/consul-api v0.0.0-20180202201655-eb2c6b5be1b6/go.mod h1:grANhF5doyWs3UAsr3K4I6qtAmlQcZDesFNEHPZAzj8=
+github.com/aws/aws-sdk-go v1.15.27/go.mod h1:mFuSZ37Z9YOHbQEwBWztmVzqXrEkub65tZoCYDt7FT0=
+github.com/aws/aws-sdk-go v1.19.18/go.mod h1:KmX6BPdI08NWTb3/sm4ZGu5ShLoqVDhKgpiN924inxo=
+github.com/aws/aws-sdk-go v1.19.45/go.mod h1:KmX6BPdI08NWTb3/sm4ZGu5ShLoqVDhKgpiN924inxo=
+github.com/aws/aws-sdk-go v1.23.13 h1:l/NG+mgQFRGG3dsFzEj0jw9JIs/zYdtU6MXhY1WIDmM=
+github.com/aws/aws-sdk-go v1.23.13/go.mod h1:KmX6BPdI08NWTb3/sm4ZGu5ShLoqVDhKgpiN924inxo=
+github.com/beorn7/perks v0.0.0-20180321164747-3a771d992973/go.mod h1:Dwedo/Wpr24TaqPxmxbtue+5NUziq4I4S80YR8gNf3Q=
+github.com/beorn7/perks v1.0.0 h1:HWo1m869IqiPhD389kmkxeTalrjNbbJTC8LXupb+sl0=
+github.com/beorn7/perks v1.0.0/go.mod h1:KWe93zE9D1o94FZ5RNwFwVgaQK1VOXiVxmqh+CedLV8=
+github.com/beorn7/perks v1.0.1 h1:VlbKKnNfV8bJzeqoa4cOKqO6bYr3WgKZxO8Z16+hsOM=
+github.com/beorn7/perks v1.0.1/go.mod h1:G2ZrVWU2WbWT9wwq4/hrbKbnv/1ERSJQ0ibhJ6rlkpw=
+github.com/bgentry/speakeasy v0.1.0/go.mod h1:+zsyZBPWlz7T6j88CTgSN5bM796AkVf0kBD4zp0CCIs=
+github.com/bitly/go-hostpool v0.0.0-20171023180738-a3a6125de932 h1:mXoPYz/Ul5HYEDvkta6I8/rnYM5gSdSV2tJ6XbZuEtY=
+github.com/bitly/go-hostpool v0.0.0-20171023180738-a3a6125de932/go.mod h1:NOuUCSz6Q9T7+igc/hlvDOUdtWKryOrtFyIVABv/p7k=
+github.com/blacktear23/go-proxyprotocol v0.0.0-20180807104634-af7a81e8dd0d/go.mod h1:VKt7CNAQxpFpSDz3sXyj9hY/GbVsQCr0sB3w59nE7lU=
+github.com/bmizerany/assert v0.0.0-20160611221934-b7ed37b82869 h1:DDGfHa7BWjL4YnC6+E63dPcxHo2sUxDIu8g3QgEJdRY=
+github.com/bmizerany/assert v0.0.0-20160611221934-b7ed37b82869/go.mod h1:Ekp36dRnpXw/yCqJaO+ZrUyxD+3VXMFFr56k5XYrpB4=
+github.com/census-instrumentation/opencensus-proto v0.2.0 h1:LzQXZOgg4CQfE6bFvXGM30YZL1WW/M337pXml+GrcZ4=
+github.com/census-instrumentation/opencensus-proto v0.2.0/go.mod h1:f6KPmirojxKA12rnyqOA5BBL4O983OfeGPqjHWSTneU=
+github.com/cespare/xxhash v1.1.0/go.mod h1:XrSqR1VqqWfGrhpAt58auRo0WTKS1nRRg3ghfAqPWnc=
+github.com/chrislusf/raft v0.0.0-20190225081310-10d6e2182d92 h1:lM9SFsh0EPXkyJyrTJqLZPAIJBtNFP6LNkYXu2MnSZI=
+github.com/chrislusf/raft v0.0.0-20190225081310-10d6e2182d92/go.mod h1:4jyiUCD5y548+yKW+oiHtccBiMaLCCbFBpK2t7X4eUo=
+github.com/chzyer/logex v1.1.10/go.mod h1:+Ywpsq7O8HXn0nuIou7OrIPyXbp3wmkHB+jjWRnGsAI=
+github.com/chzyer/readline v0.0.0-20171208011716-f6d7a1f6fbf3/go.mod h1:nSuG5e5PlCu98SY8svDHJxuZscDgtXS6KTTbou5AhLI=
+github.com/chzyer/test v0.0.0-20180213035817-a1ea475d72b1/go.mod h1:Q3SI9o4m/ZMnBNeIyt5eFwwo7qiLfzFZmjNmxjkiQlU=
+github.com/client9/misspell v0.3.4/go.mod h1:qj6jICC3Q7zFZvVWo7KLAzC3yx5G7kyvSDkc90ppPyw=
+github.com/codahale/hdrhistogram v0.0.0-20161010025455-3a0bb77429bd h1:qMd81Ts1T2OTKmB4acZcyKaMtRnY5Y44NuXGX2GFJ1w=
+github.com/codahale/hdrhistogram v0.0.0-20161010025455-3a0bb77429bd/go.mod h1:sE/e/2PUdi/liOCUjSTXgM1o87ZssimdTWN964YiIeI=
+github.com/coreos/bbolt v1.3.2 h1:wZwiHHUieZCquLkDL0B8UhzreNWsPHooDAG3q34zk0s=
+github.com/coreos/bbolt v1.3.2/go.mod h1:iRUV2dpdMOn7Bo10OQBFzIJO9kkE559Wcmn+qkEiiKk=
+github.com/coreos/bbolt v1.3.3 h1:n6AiVyVRKQFNb6mJlwESEvvLoDyiTzXX7ORAUlkeBdY=
+github.com/coreos/bbolt v1.3.3/go.mod h1:iRUV2dpdMOn7Bo10OQBFzIJO9kkE559Wcmn+qkEiiKk=
+github.com/coreos/etcd v3.3.10+incompatible h1:jFneRYjIvLMLhDLCzuTuU4rSJUjRplcJQ7pD7MnhC04=
+github.com/coreos/etcd v3.3.10+incompatible/go.mod h1:uF7uidLiAD3TWHmW31ZFd/JWoc32PjwdhPthX9715RE=
+github.com/coreos/etcd v3.3.13+incompatible/go.mod h1:uF7uidLiAD3TWHmW31ZFd/JWoc32PjwdhPthX9715RE=
+github.com/coreos/etcd v3.3.15+incompatible h1:+9RjdC18gMxNQVvSiXvObLu29mOFmkgdsB4cRTlV+EE=
+github.com/coreos/etcd v3.3.15+incompatible/go.mod h1:uF7uidLiAD3TWHmW31ZFd/JWoc32PjwdhPthX9715RE=
+github.com/coreos/go-semver v0.2.0 h1:3Jm3tLmsgAYcjC+4Up7hJrFBPr+n7rAqYeSw/SZazuY=
+github.com/coreos/go-semver v0.2.0/go.mod h1:nnelYz7RCh+5ahJtPPxZlU+153eP4D4r3EedlOD2RNk=
+github.com/coreos/go-semver v0.3.0 h1:wkHLiw0WNATZnSG7epLsujiMCgPAc9xhjJ4tgnAxmfM=
+github.com/coreos/go-semver v0.3.0/go.mod h1:nnelYz7RCh+5ahJtPPxZlU+153eP4D4r3EedlOD2RNk=
+github.com/coreos/go-systemd v0.0.0-20180511133405-39ca1b05acc7/go.mod h1:F5haX7vjVVG0kc13fIWeqUViNPyEJxv/OmvnBo0Yme4=
+github.com/coreos/go-systemd v0.0.0-20181031085051-9002847aa142/go.mod h1:F5haX7vjVVG0kc13fIWeqUViNPyEJxv/OmvnBo0Yme4=
+github.com/coreos/go-systemd v0.0.0-20190321100706-95778dfbb74e h1:Wf6HqHfScWJN9/ZjdUKyjop4mf3Qdd+1TvvltAvM3m8=
+github.com/coreos/go-systemd v0.0.0-20190321100706-95778dfbb74e/go.mod h1:F5haX7vjVVG0kc13fIWeqUViNPyEJxv/OmvnBo0Yme4=
+github.com/coreos/go-systemd v0.0.0-20190719114852-fd7a80b32e1f h1:JOrtw2xFKzlg+cbHpyrpLDmnN1HqhBfnX7WDiW7eG2c=
+github.com/coreos/go-systemd v0.0.0-20190719114852-fd7a80b32e1f/go.mod h1:F5haX7vjVVG0kc13fIWeqUViNPyEJxv/OmvnBo0Yme4=
+github.com/coreos/pkg v0.0.0-20160727233714-3ac0863d7acf/go.mod h1:E3G3o1h8I7cfcXa63jLwjI0eiQQMgzzUDFVpN/nH/eA=
+github.com/coreos/pkg v0.0.0-20180928190104-399ea9e2e55f h1:lBNOc5arjvs8E5mO2tbpBpLoyyu8B6e44T7hJy6potg=
+github.com/coreos/pkg v0.0.0-20180928190104-399ea9e2e55f/go.mod h1:E3G3o1h8I7cfcXa63jLwjI0eiQQMgzzUDFVpN/nH/eA=
+github.com/cznic/mathutil v0.0.0-20181122101859-297441e03548 h1:iwZdTE0PVqJCos1vaoKsclOGD3ADKpshg3SRtYBbwso=
+github.com/cznic/mathutil v0.0.0-20181122101859-297441e03548/go.mod h1:e6NPNENfs9mPDVNRekM7lKScauxd5kXTr1Mfyig6TDM=
+github.com/cznic/sortutil v0.0.0-20150617083342-4c7342852e65 h1:hxuZop6tSoOi0sxFzoGGYdRqNrPubyaIf9KoBG9tPiE=
+github.com/cznic/sortutil v0.0.0-20150617083342-4c7342852e65/go.mod h1:q2w6Bg5jeox1B+QkJ6Wp/+Vn0G/bo3f1uY7Fn3vivIQ=
+github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
+github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/devigned/tab v0.1.1/go.mod h1:XG9mPq0dFghrYvoBF3xdRrJzSTX1b7IQrvaL9mzjeJY=
+github.com/dgrijalva/jwt-go v3.2.0+incompatible h1:7qlOGliEKZXTDg6OTjfoBKDXWrumCAMpl/TFQ4/5kLM=
+github.com/dgrijalva/jwt-go v3.2.0+incompatible/go.mod h1:E3ru+11k8xSBh+hMPgOLZmtrrCbhqsmaPHjLKYnJCaQ=
+github.com/dgryski/go-farm v0.0.0-20190104051053-3adb47b1fb0f h1:dDxpBYafY/GYpcl+LS4Bn3ziLPuEdGRkRjYAbSlWxSA=
+github.com/dgryski/go-farm v0.0.0-20190104051053-3adb47b1fb0f/go.mod h1:SqUrOPUnsFjfmXRMNPybcSiG0BgUW2AuFH8PAnS2iTw=
+github.com/dgryski/go-sip13 v0.0.0-20181026042036-e10d5fee7954/go.mod h1:vAd38F8PWV+bWy6jNmig1y/TA+kYO4g3RSRF0IAv0no=
+github.com/dimchansky/utfbom v1.1.0/go.mod h1:rO41eb7gLfo8SF1jd9F8HplJm1Fewwi4mQvIirEdv+8=
+github.com/disintegration/imaging v1.6.1 h1:JnBbK6ECIZb1NsWIikP9pd8gIlTIRx7fuDNpU9fsxOE=
+github.com/disintegration/imaging v1.6.1/go.mod h1:xuIt+sRxDFrHS0drzXUlCJthkJ8k7lkkUojDSR247MQ=
+github.com/dustin/go-humanize v0.0.0-20171111073723-bb3d318650d4/go.mod h1:HtrtbFcZ19U5GC7JDqmcUSB87Iq5E25KnS6fMYU6eOk=
+github.com/dustin/go-humanize v0.0.0-20180421182945-02af3965c54e/go.mod h1:HtrtbFcZ19U5GC7JDqmcUSB87Iq5E25KnS6fMYU6eOk=
+github.com/dustin/go-humanize v1.0.0 h1:VSnTsYCnlFHaM2/igO1h6X3HA71jcobQuxemgkq4zYo=
+github.com/dustin/go-humanize v1.0.0/go.mod h1:HtrtbFcZ19U5GC7JDqmcUSB87Iq5E25KnS6fMYU6eOk=
+github.com/eapache/go-resiliency v1.1.0 h1:1NtRmCAqadE2FN4ZcN6g90TP3uk8cg9rn9eNK2197aU=
+github.com/eapache/go-resiliency v1.1.0/go.mod h1:kFI+JgMyC7bLPUVY133qvEBtVayf5mFgVsvEsIPBvNs=
+github.com/eapache/go-resiliency v1.2.0 h1:v7g92e/KSN71Rq7vSThKaWIq68fL4YHvWyiUKorFR1Q=
+github.com/eapache/go-resiliency v1.2.0/go.mod h1:kFI+JgMyC7bLPUVY133qvEBtVayf5mFgVsvEsIPBvNs=
+github.com/eapache/go-xerial-snappy v0.0.0-20180814174437-776d5712da21 h1:YEetp8/yCZMuEPMUDHG0CW/brkkEp8mzqk2+ODEitlw=
+github.com/eapache/go-xerial-snappy v0.0.0-20180814174437-776d5712da21/go.mod h1:+020luEh2TKB4/GOp8oxxtq0Daoen/Cii55CzbTV6DU=
+github.com/eapache/queue v1.1.0 h1:YOEu7KNc61ntiQlcEeUIoDTJ2o8mQznoNvUhiigpIqc=
+github.com/eapache/queue v1.1.0/go.mod h1:6eCeP0CKFpHLu8blIFXhExK/dRa7WDZfr6jVFPTqq+I=
+github.com/eknkc/amber v0.0.0-20171010120322-cdade1c07385 h1:clC1lXBpe2kTj2VHdaIu9ajZQe4kcEY9j0NsnDDBZ3o=
+github.com/eknkc/amber v0.0.0-20171010120322-cdade1c07385/go.mod h1:0vRUJqYpeSZifjYj7uP3BG/gKcuzL9xWVV/Y+cK33KM=
+github.com/facebookgo/clock v0.0.0-20150410010913-600d898af40a h1:yDWHCSQ40h88yih2JAcL6Ls/kVkSE8GFACTGVnMPruw=
+github.com/facebookgo/clock v0.0.0-20150410010913-600d898af40a/go.mod h1:7Ga40egUymuWXxAe151lTNnCv97MddSOVsjpPPkityA=
+github.com/facebookgo/stats v0.0.0-20151006221625-1b76add642e4 h1:0YtRCqIZs2+Tz49QuH6cJVw/IFqzo39gEqZ0iYLxD2M=
+github.com/facebookgo/stats v0.0.0-20151006221625-1b76add642e4/go.mod h1:vsJz7uE339KUCpBXx3JAJzSRH7Uk4iGGyJzR529qDIA=
+github.com/fatih/color v1.7.0/go.mod h1:Zm6kSWBoL9eyXnKyktHP6abPY2pDugNf5KwzbycvMj4=
+github.com/fortytw2/leaktest v1.2.0/go.mod h1:jDsjWgpAGjm2CA7WthBh/CdZYEPF31XHquHwclZch5g=
+github.com/fortytw2/leaktest v1.3.0/go.mod h1:jDsjWgpAGjm2CA7WthBh/CdZYEPF31XHquHwclZch5g=
+github.com/frankban/quicktest v1.7.2 h1:2QxQoC1TS09S7fhCPsrvqYdvP1H5M1P1ih5ABm3BTYk=
+github.com/frankban/quicktest v1.7.2/go.mod h1:jaStnuzAqU1AJdCO0l53JDCJrVDKcS03DbaAcR7Ks/o=
+github.com/fsnotify/fsnotify v1.4.7 h1:IXs+QLmnXW2CcXuY+8Mzv/fWEsPGWxqefPtCP5CnV9I=
+github.com/fsnotify/fsnotify v1.4.7/go.mod h1:jwhsz4b93w/PPRr/qN1Yymfu8t87LnFCMoQvtojpjFo=
+github.com/gabriel-vasile/mimetype v0.3.17 h1:NGWgggJJqTofUcTV1E7hkk2zVjZ54EfJa1z5O3z6By4=
+github.com/gabriel-vasile/mimetype v0.3.17/go.mod h1:kMJbg3SlWZCsj4R73F1WDzbT9AyGCOVmUtIxxwO5pmI=
+github.com/gabriel-vasile/mimetype v1.0.0 h1:0QKnAQQhG6oOsb4GK7iPlet7RtjHi9us8RF/nXoTxhI=
+github.com/gabriel-vasile/mimetype v1.0.0/go.mod h1:6CDPel/o/3/s4+bp6kIbsWATq8pmgOisOPG40CJa6To=
+github.com/ghodss/yaml v1.0.0 h1:wQHKEahhL6wmXdzwWG11gIVCkOv05bNOh+Rxn0yngAk=
+github.com/ghodss/yaml v1.0.0/go.mod h1:4dBDuWmgqj2HViK6kFavaiC9ZROes6MMH2rRYeMEF04=
+github.com/ghodss/yaml v1.0.1-0.20190212211648-25d852aebe32/go.mod h1:GIjDIg/heH5DOkXY3YJ/wNhfHsQHoXGjl8G8amsYQ1I=
+github.com/go-ini/ini v1.25.4/go.mod h1:ByCAeIL28uOIIG0E3PJtZPDL8WnHpFKFOtgjp+3Ies8=
+github.com/go-kit/kit v0.8.0/go.mod h1:xBxKIO96dXMWWy0MnWVtmwkA9/13aqxPnvrjFYMA2as=
+github.com/go-logfmt/logfmt v0.3.0/go.mod h1:Qt1PoO58o5twSAckw1HlFXLmHsOX5/0LbT9GBnD5lWE=
+github.com/go-logfmt/logfmt v0.4.0/go.mod h1:3RMwSq7FuexP4Kalkev3ejPJsZTpXXBr9+V4qmtdjCk=
+github.com/go-ole/go-ole v1.2.1 h1:2lOsA72HgjxAuMlKpFiCbHTvu44PIVkZ5hqm3RSdI/E=
+github.com/go-ole/go-ole v1.2.1/go.mod h1:7FAglXiTm7HKlQRDeOQ6ZNUHidzCWXuZWq/1dTyBNF8=
+github.com/go-playground/overalls v0.0.0-20180201144345-22ec1a223b7c/go.mod h1:UqxAgEOt89sCiXlrc/ycnx00LVvUO/eS8tMUkWX4R7w=
+github.com/go-redis/redis v6.15.2+incompatible h1:9SpNVG76gr6InJGxoZ6IuuxaCOQwDAhzyXg+Bs+0Sb4=
+github.com/go-redis/redis v6.15.2+incompatible/go.mod h1:NAIEuMOZ/fxfXJIrKDQDz8wamY7mA7PouImQ2Jvg6kA=
+github.com/go-sql-driver/mysql v0.0.0-20170715192408-3955978caca4/go.mod h1:zAC/RDZ24gD3HViQzih4MyKcchzm+sOG5ZlKdlhCg5w=
+github.com/go-sql-driver/mysql v1.4.1 h1:g24URVg0OFbNUTx9qqY1IRZ9D9z3iPyi5zKhQZpNwpA=
+github.com/go-sql-driver/mysql v1.4.1/go.mod h1:zAC/RDZ24gD3HViQzih4MyKcchzm+sOG5ZlKdlhCg5w=
+github.com/go-stack/stack v1.8.0/go.mod h1:v0f6uXyyMGvRgIKkXu+yp6POWl0qKG85gN/melR3HDY=
+github.com/gocql/gocql v0.0.0-20190829130954-e163eff7a8c6 h1:P66kRWyEoIx6URKgAC3ijx9jo9gEid7bEhLQ/Z0G65A=
+github.com/gocql/gocql v0.0.0-20190829130954-e163eff7a8c6/go.mod h1:Q7Sru5153KG8D9zwueuQJB3ccJf9/bIwF/x8b3oKgT8=
+github.com/gogo/protobuf v0.0.0-20180717141946-636bf0302bc9/go.mod h1:r8qH/GZQm5c6nD/R0oafs1akxWv10x8SbQlK7atdtwQ=
+github.com/gogo/protobuf v1.0.0/go.mod h1:r8qH/GZQm5c6nD/R0oafs1akxWv10x8SbQlK7atdtwQ=
+github.com/gogo/protobuf v1.1.1/go.mod h1:r8qH/GZQm5c6nD/R0oafs1akxWv10x8SbQlK7atdtwQ=
+github.com/gogo/protobuf v1.2.0/go.mod h1:r8qH/GZQm5c6nD/R0oafs1akxWv10x8SbQlK7atdtwQ=
+github.com/gogo/protobuf v1.2.1 h1:/s5zKNz0uPFCZ5hddgPdo2TK2TVrUNMn0OOX8/aZMTE=
+github.com/gogo/protobuf v1.2.1/go.mod h1:hp+jE20tsWTFYpLwKvXlhS1hjn+gTNwPg2I6zVXpSg4=
+github.com/gogo/protobuf v1.2.2-0.20190730201129-28a6bbf47e48 h1:X+zN6RZXsvnrSJaAIQhZezPfAfvsqihKKR8oiLHid34=
+github.com/gogo/protobuf v1.2.2-0.20190730201129-28a6bbf47e48/go.mod h1:SlYgWuQ5SjCEi6WLHjHCa1yvBfUnHcTbrrZtXPKa29o=
+github.com/golang/glog v0.0.0-20160126235308-23def4e6c14b h1:VKtxabqXZkF25pY9ekfRL6a582T4P37/31XEstQ5p58=
+github.com/golang/glog v0.0.0-20160126235308-23def4e6c14b/go.mod h1:SBH7ygxi8pfUlaOkMMuAQtPIUF8ecWP5IEl/CR7VP2Q=
+github.com/golang/groupcache v0.0.0-20160516000752-02826c3e7903/go.mod h1:cIg4eruTrX1D+g88fzRXU5OdNfaM+9IcxsU14FzY7Hc=
+github.com/golang/groupcache v0.0.0-20181024230925-c65c006176ff/go.mod h1:cIg4eruTrX1D+g88fzRXU5OdNfaM+9IcxsU14FzY7Hc=
+github.com/golang/groupcache v0.0.0-20190129154638-5b532d6fd5ef h1:veQD95Isof8w9/WXiA+pa3tz3fJXkt5B7QaRBrM62gk=
+github.com/golang/groupcache v0.0.0-20190129154638-5b532d6fd5ef/go.mod h1:cIg4eruTrX1D+g88fzRXU5OdNfaM+9IcxsU14FzY7Hc=
+github.com/golang/groupcache v0.0.0-20190702054246-869f871628b6 h1:ZgQEtGgCBiWRM39fZuwSd1LwSqqSW0hOdXCYYDX0R3I=
+github.com/golang/groupcache v0.0.0-20190702054246-869f871628b6/go.mod h1:cIg4eruTrX1D+g88fzRXU5OdNfaM+9IcxsU14FzY7Hc=
+github.com/golang/mock v1.1.1/go.mod h1:oTYuIxOrZwtPieC+H1uAHpcLFnEyAGVDL/k47Jfbm0A=
+github.com/golang/mock v1.2.0/go.mod h1:oTYuIxOrZwtPieC+H1uAHpcLFnEyAGVDL/k47Jfbm0A=
+github.com/golang/mock v1.3.1/go.mod h1:sBzyDLLjw3U8JLTeZvSv8jJB+tU5PVekmnlKIyFUx0Y=
+github.com/golang/protobuf v0.0.0-20180814211427-aa810b61a9c7/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
+github.com/golang/protobuf v1.2.0/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
+github.com/golang/protobuf v1.3.1/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
+github.com/golang/protobuf v1.3.2 h1:6nsPYzhq5kReh6QImI3k5qWzO4PEbvbIW2cwSfR/6xs=
+github.com/golang/protobuf v1.3.2/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
+github.com/golang/snappy v0.0.0-20170215233205-553a64147049/go.mod h1:/XxbfmMg8lxefKM7IXC3fBNl/7bRcc72aCRzEWrmP2Q=
+github.com/golang/snappy v0.0.0-20180518054509-2e65f85255db/go.mod h1:/XxbfmMg8lxefKM7IXC3fBNl/7bRcc72aCRzEWrmP2Q=
+github.com/golang/snappy v0.0.1 h1:Qgr9rKW7uDUkrbSmQeiDsGa8SjGyCOGtuasMWwvp2P4=
+github.com/golang/snappy v0.0.1/go.mod h1:/XxbfmMg8lxefKM7IXC3fBNl/7bRcc72aCRzEWrmP2Q=
+github.com/google/btree v0.0.0-20180124185431-e89373fe6b4a/go.mod h1:lNA+9X1NB3Zf8V7Ke586lFgjr2dZNuvo3lPJSGZ5JPQ=
+github.com/google/btree v0.0.0-20180813153112-4030bb1f1f0c/go.mod h1:lNA+9X1NB3Zf8V7Ke586lFgjr2dZNuvo3lPJSGZ5JPQ=
+github.com/google/btree v1.0.0 h1:0udJVsspx3VBr5FwtLhQQtuAsVc79tTq0ocGIPAU6qo=
+github.com/google/btree v1.0.0/go.mod h1:lNA+9X1NB3Zf8V7Ke586lFgjr2dZNuvo3lPJSGZ5JPQ=
+github.com/google/go-cmp v0.2.0/go.mod h1:oXzfMopK8JAjlY9xF4vHSVASa0yLyX7SntLO5aqRK0M=
+github.com/google/go-cmp v0.3.0 h1:crn/baboCvb5fXaQ0IJ1SGTsTVrWpDsCWC8EGETZijY=
+github.com/google/go-cmp v0.3.0/go.mod h1:8QqcDgzrUqlUb/G2PQTWiueGozuR1884gddMywk6iLU=
+github.com/google/go-cmp v0.3.1 h1:Xye71clBPdm5HgqGwUkwhbynsUJZhDbS20FvLhQ2izg=
+github.com/google/go-cmp v0.3.1/go.mod h1:8QqcDgzrUqlUb/G2PQTWiueGozuR1884gddMywk6iLU=
+github.com/google/go-replayers/grpcreplay v0.1.0 h1:eNb1y9rZFmY4ax45uEEECSa8fsxGRU+8Bil52ASAwic=
+github.com/google/go-replayers/grpcreplay v0.1.0/go.mod h1:8Ig2Idjpr6gifRd6pNVggX6TC1Zw6Jx74AKp7QNH2QE=
+github.com/google/go-replayers/httpreplay v0.1.0 h1:AX7FUb4BjrrzNvblr/OlgwrmFiep6soj5K2QSDW7BGk=
+github.com/google/go-replayers/httpreplay v0.1.0/go.mod h1:YKZViNhiGgqdBlUbI2MwGpq4pXxNmhJLPHQ7cv2b5no=
+github.com/google/gofuzz v1.0.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg=
+github.com/google/martian v2.1.0+incompatible/go.mod h1:9I4somxYTbIHy5NJKHRl3wXiIaQGbYVAs8BPL6v8lEs=
+github.com/google/martian v2.1.1-0.20190517191504-25dcb96d9e51+incompatible h1:xmapqc1AyLoB+ddYT6r04bD9lIjlOqGaREovi0SzFaE=
+github.com/google/martian v2.1.1-0.20190517191504-25dcb96d9e51+incompatible/go.mod h1:9I4somxYTbIHy5NJKHRl3wXiIaQGbYVAs8BPL6v8lEs=
+github.com/google/pprof v0.0.0-20181206194817-3ea8567a2e57/go.mod h1:zfwlbNMJ+OItoe0UupaVj+oy1omPYYDuagoSzA8v9mc=
+github.com/google/pprof v0.0.0-20190515194954-54271f7e092f/go.mod h1:zfwlbNMJ+OItoe0UupaVj+oy1omPYYDuagoSzA8v9mc=
+github.com/google/shlex v0.0.0-20181106134648-c34317bd91bf/go.mod h1:RpwtwJQFrIEPstU94h88MWPXP2ektJZ8cZ0YntAmXiE=
+github.com/google/subcommands v1.0.1/go.mod h1:ZjhPrFU+Olkh9WazFPsl27BQ4UPiG37m3yTrtFlrHVk=
+github.com/google/uuid v1.0.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
+github.com/google/uuid v1.1.1 h1:Gkbcsh/GbpXz7lPftLA3P6TYMwjCLYm83jiFQZF/3gY=
+github.com/google/uuid v1.1.1/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
+github.com/google/wire v0.3.0 h1:imGQZGEVEHpje5056+K+cgdO72p0LQv2xIIFXNGUf60=
+github.com/google/wire v0.3.0/go.mod h1:i1DMg/Lu8Sz5yYl25iOdmc5CT5qusaa+zmRWs16741s=
+github.com/googleapis/gax-go v2.0.2+incompatible h1:silFMLAnr330+NRuag/VjIGF7TLp/LBrV2CJKFLWEww=
+github.com/googleapis/gax-go v2.0.2+incompatible/go.mod h1:SFVmujtThgffbyetf+mdk2eWhX2bMyUtNHzFKcPA9HY=
+github.com/googleapis/gax-go/v2 v2.0.4/go.mod h1:0Wqv26UfaUD9n4G6kQubkQ+KchISgw+vpHVxEJEs9eg=
+github.com/googleapis/gax-go/v2 v2.0.5 h1:sjZBwGj9Jlw33ImPtvFviGYvseOtDM7hkSKB7+Tv3SM=
+github.com/googleapis/gax-go/v2 v2.0.5/go.mod h1:DWXyrwAJ9X0FpwwEdw+IPEYBICEFu5mhpdKc/us6bOk=
+github.com/gorilla/context v0.0.0-20160226214623-1ea25387ff6f/go.mod h1:kBGZzfjB9CEq2AlWe17Uuf7NDRt0dE0s8S51q0aT7Yg=
+github.com/gorilla/context v1.1.1/go.mod h1:kBGZzfjB9CEq2AlWe17Uuf7NDRt0dE0s8S51q0aT7Yg=
+github.com/gorilla/mux v1.6.1/go.mod h1:1lud6UwP+6orDFRuTfBEV8e9/aOM/c4fVVCaMa2zaAs=
+github.com/gorilla/mux v1.6.2/go.mod h1:1lud6UwP+6orDFRuTfBEV8e9/aOM/c4fVVCaMa2zaAs=
+github.com/gorilla/mux v1.7.3 h1:gnP5JzjVOuiZD07fKKToCAOjS0yOpj/qPETTXCCS6hw=
+github.com/gorilla/mux v1.7.3/go.mod h1:1lud6UwP+6orDFRuTfBEV8e9/aOM/c4fVVCaMa2zaAs=
+github.com/gorilla/websocket v0.0.0-20170926233335-4201258b820c/go.mod h1:E7qHFY5m1UJ88s3WnNqhKjPHQ0heANvMoAMk2YaljkQ=
+github.com/gorilla/websocket v1.2.0/go.mod h1:E7qHFY5m1UJ88s3WnNqhKjPHQ0heANvMoAMk2YaljkQ=
+github.com/gorilla/websocket v1.4.0 h1:WDFjx/TMzVgy9VdMMQi2K2Emtwi2QcUQsztZ/zLaH/Q=
+github.com/gorilla/websocket v1.4.0/go.mod h1:E7qHFY5m1UJ88s3WnNqhKjPHQ0heANvMoAMk2YaljkQ=
+github.com/gorilla/websocket v1.4.1 h1:q7AeDBpnBk8AogcD4DSag/Ukw/KV+YhzLj2bP5HvKCM=
+github.com/gorilla/websocket v1.4.1/go.mod h1:YR8l580nyteQvAITg2hZ9XVh4b55+EU/adAjf1fMHhE=
+github.com/grpc-ecosystem/go-grpc-middleware v1.0.0 h1:Iju5GlWwrvL6UBg4zJJt3btmonfrMlCDdsejg4CZE7c=
+github.com/grpc-ecosystem/go-grpc-middleware v1.0.0/go.mod h1:FiyG127CGDf3tlThmgyCl78X/SZQqEOJBCDaAfeWzPs=
+github.com/grpc-ecosystem/go-grpc-middleware v1.0.1-0.20190118093823-f849b5445de4 h1:z53tR0945TRRQO/fLEVPI6SMv7ZflF0TEaTAoU7tOzg=
+github.com/grpc-ecosystem/go-grpc-middleware v1.0.1-0.20190118093823-f849b5445de4/go.mod h1:FiyG127CGDf3tlThmgyCl78X/SZQqEOJBCDaAfeWzPs=
+github.com/grpc-ecosystem/go-grpc-prometheus v1.2.0 h1:Ovs26xHkKqVztRpIrF/92BcuyuQ/YW4NSIpoGtfXNho=
+github.com/grpc-ecosystem/go-grpc-prometheus v1.2.0/go.mod h1:8NvIoxWQoOIhqOTXgfV/d3M/q6VIi02HzZEHgUlZvzk=
+github.com/grpc-ecosystem/grpc-gateway v1.4.1/go.mod h1:RSKVYQBd5MCa4OVpNdGskqpgL2+G+NZTnrVHpWWfpdw=
+github.com/grpc-ecosystem/grpc-gateway v1.5.1/go.mod h1:RSKVYQBd5MCa4OVpNdGskqpgL2+G+NZTnrVHpWWfpdw=
+github.com/grpc-ecosystem/grpc-gateway v1.8.5/go.mod h1:vNeuVxBJEsws4ogUvrchl83t/GYV9WGTSLVdBhOQFDY=
+github.com/grpc-ecosystem/grpc-gateway v1.9.0 h1:bM6ZAFZmc/wPFaRDi0d5L7hGEZEx/2u+Tmr2evNHDiI=
+github.com/grpc-ecosystem/grpc-gateway v1.9.0/go.mod h1:vNeuVxBJEsws4ogUvrchl83t/GYV9WGTSLVdBhOQFDY=
+github.com/grpc-ecosystem/grpc-gateway v1.9.2/go.mod h1:vNeuVxBJEsws4ogUvrchl83t/GYV9WGTSLVdBhOQFDY=
+github.com/grpc-ecosystem/grpc-gateway v1.11.0 h1:aT5ISUniaOTErogCQ+4pGoYNBB6rm6Fq3g1v8QwYGas=
+github.com/grpc-ecosystem/grpc-gateway v1.11.0/go.mod h1:vNeuVxBJEsws4ogUvrchl83t/GYV9WGTSLVdBhOQFDY=
+github.com/hailocab/go-hostpool v0.0.0-20160125115350-e80d13ce29ed h1:5upAirOpQc1Q53c0bnx2ufif5kANL7bfZWcc6VJWJd8=
+github.com/hailocab/go-hostpool v0.0.0-20160125115350-e80d13ce29ed/go.mod h1:tMWxXQ9wFIaZeTI9F+hmhFiGpFmhOHzyShyFUhRm0H4=
+github.com/hashicorp/go-uuid v1.0.1 h1:fv1ep09latC32wFoVwnqcnKJGnMSdBanPczbHAYm1BE=
+github.com/hashicorp/go-uuid v1.0.1/go.mod h1:6SBZvOh/SIDV7/2o3Jml5SYk/TvGqwFJ/bN7x4byOro=
+github.com/hashicorp/golang-lru v0.5.0/go.mod h1:/m3WP610KZHVQ1SGc6re/UDhFvYD7pJ4Ao+sR/qLZy8=
+github.com/hashicorp/golang-lru v0.5.1 h1:0hERBMJE1eitiLkihrMvRVBYAkpHzc/J3QdDN+dAcgU=
+github.com/hashicorp/golang-lru v0.5.1/go.mod h1:/m3WP610KZHVQ1SGc6re/UDhFvYD7pJ4Ao+sR/qLZy8=
+github.com/hashicorp/golang-lru v0.5.3 h1:YPkqC67at8FYaadspW/6uE0COsBxS2656RLEr8Bppgk=
+github.com/hashicorp/golang-lru v0.5.3/go.mod h1:iADmTwqILo4mZ8BN3D2Q6+9jd8WM5uGBxy+E8yxSoD4=
+github.com/hashicorp/hcl v1.0.0 h1:0Anlzjpi4vEasTeNFn2mLJgTSwt0+6sfsiTG8qcWGx4=
+github.com/hashicorp/hcl v1.0.0/go.mod h1:E5yfLk+7swimpb2L/Alb/PJmXilQ/rhwaUYs4T20WEQ=
+github.com/hpcloud/tail v1.0.0 h1:nfCOvKYfkgYP8hkirhJocXT2+zOD8yUNjXaWfTlyFKI=
+github.com/hpcloud/tail v1.0.0/go.mod h1:ab1qPbhIpdTxEkNHXyeSf5vhxWSCs/tWer42PpOxQnU=
+github.com/inconshreveable/mousetrap v1.0.0/go.mod h1:PxqpIevigyE2G7u3NXJIT2ANytuPF1OarO4DADm73n8=
+github.com/jacobsa/daemonize v0.0.0-20160101105449-e460293e890f h1:X+tnaqoCcBgAwSTJtoYW6p0qKiuPyMfofEHEFUf2kdU=
+github.com/jacobsa/daemonize v0.0.0-20160101105449-e460293e890f/go.mod h1:Ip4fOwzCrnDVuluHBd7FXIMb7SHOKfkt9/UDrYSZvqI=
+github.com/jcmturner/gofork v0.0.0-20190328161633-dc7c13fece03 h1:FUwcHNlEqkqLjLBdCp5PRlCFijNjvcYANOZXzCfXwCM=
+github.com/jcmturner/gofork v0.0.0-20190328161633-dc7c13fece03/go.mod h1:MK8+TM0La+2rjBD4jE12Kj1pCCxK7d2LK/UM3ncEo0o=
+github.com/jcmturner/gofork v1.0.0 h1:J7uCkflzTEhUZ64xqKnkDxq3kzc96ajM1Gli5ktUem8=
+github.com/jcmturner/gofork v1.0.0/go.mod h1:MK8+TM0La+2rjBD4jE12Kj1pCCxK7d2LK/UM3ncEo0o=
+github.com/jmespath/go-jmespath v0.0.0-20160202185014-0b12d6b521d8/go.mod h1:Nht3zPeWKUH0NzdCt2Blrr5ys8VGpn0CEB0cQHVjt7k=
+github.com/jmespath/go-jmespath v0.0.0-20180206201540-c2b33e8439af h1:pmfjZENx5imkbgOkpRUYLnmbU7UEFbjtDA2hxJ1ichM=
+github.com/jmespath/go-jmespath v0.0.0-20180206201540-c2b33e8439af/go.mod h1:Nht3zPeWKUH0NzdCt2Blrr5ys8VGpn0CEB0cQHVjt7k=
+github.com/joho/godotenv v1.3.0/go.mod h1:7hK45KPybAkOC6peb+G5yklZfMxEjkZhHbwpqxOKXbg=
+github.com/jonboulle/clockwork v0.1.0 h1:VKV+ZcuP6l3yW9doeqz6ziZGgcynBVQO+obU0+0hcPo=
+github.com/jonboulle/clockwork v0.1.0/go.mod h1:Ii8DK3G1RaLaWxj9trq07+26W01tbo22gdxWY5EU2bo=
+github.com/json-iterator/go v1.1.6/go.mod h1:+SdeFBvtyEkXs7REEP0seUULqWtbJapLOCVDaaPEHmU=
+github.com/json-iterator/go v1.1.7 h1:KfgG9LzI+pYjr4xvmz/5H4FXjokeP+rlHLhv3iH62Fo=
+github.com/json-iterator/go v1.1.7/go.mod h1:KdQUCv79m/52Kvf8AW2vK1V8akMuk1QjK/uOdHXbAo4=
+github.com/jstemmer/go-junit-report v0.0.0-20190106144839-af01ea7f8024/go.mod h1:6v2b51hI/fHJwM22ozAgKL4VKDeJcHhJFhtBdhmNjmU=
+github.com/juju/ratelimit v1.0.1 h1:+7AIFJVQ0EQgq/K9+0Krm7m530Du7tIz0METWzN0RgY=
+github.com/juju/ratelimit v1.0.1/go.mod h1:qapgC/Gy+xNh9UxzV13HGGl/6UXNN+ct+vwSgWNm/qk=
+github.com/julienschmidt/httprouter v1.2.0/go.mod h1:SYymIcj16QtmaHHD7aYtjjsJG7VTCxuUUipMqKk8s4w=
+github.com/karlseguin/ccache v2.0.3+incompatible h1:j68C9tWOROiOLWTS/kCGg9IcJG+ACqn5+0+t8Oh83UU=
+github.com/karlseguin/ccache v2.0.3+incompatible/go.mod h1:CM9tNPzT6EdRh14+jiW8mEF9mkNZuuE51qmgGYUB93w=
+github.com/karlseguin/expect v1.0.1 h1:z4wy4npwwHSWKjGWH85WNJO42VQhovxTCZDSzhjo8hY=
+github.com/karlseguin/expect v1.0.1/go.mod h1:zNBxMY8P21owkeogJELCLeHIt+voOSduHYTFUbwRAV8=
+github.com/kisielk/errcheck v1.1.0/go.mod h1:EZBBE59ingxPouuu3KfxchcWSUPOHkagtvWXihfKN4Q=
+github.com/kisielk/errcheck v1.2.0/go.mod h1:/BMXB+zMLi60iA8Vv6Ksmxu/1UDYcXs4uQLJ+jE2L00=
+github.com/kisielk/gotool v1.0.0/go.mod h1:XhKaO+MFFWcvkIS/tQcRk01m1F5IRFswLeQ+oQHNcck=
+github.com/klauspost/cpuid v0.0.0-20170728055534-ae7887de9fa5/go.mod h1:Pj4uuM528wm8OyEC2QMXAi2YiTZ96dNQPGgoMS4s3ek=
+github.com/klauspost/cpuid v1.2.1 h1:vJi+O/nMdFt0vqm8NZBI6wzALWdA2X+egi0ogNyrC/w=
+github.com/klauspost/cpuid v1.2.1/go.mod h1:Pj4uuM528wm8OyEC2QMXAi2YiTZ96dNQPGgoMS4s3ek=
+github.com/klauspost/crc32 v1.2.0 h1:0VuyqOCruD33/lJ/ojXNvzVyl8Zr5zdTmj9l9qLZ86I=
+github.com/klauspost/crc32 v1.2.0/go.mod h1:+ZoRqAPRLkC4NPOvfYeR5KNOrY6TD+/sAC3HXPZgDYg=
+github.com/klauspost/reedsolomon v1.9.2 h1:E9CMS2Pqbv+C7tsrYad4YC9MfhnMVWhMRsTi7U0UB18=
+github.com/klauspost/reedsolomon v1.9.2/go.mod h1:CwCi+NUr9pqSVktrkN+Ondf06rkhYZ/pcNv7fu+8Un4=
+github.com/konsorten/go-windows-terminal-sequences v1.0.1 h1:mweAR1A6xJ3oS2pRaGiHgQ4OO8tzTaLawm8vnODuwDk=
+github.com/konsorten/go-windows-terminal-sequences v1.0.1/go.mod h1:T0+1ngSBFLxvqU3pZ+m/2kptfBszLMUkC4ZK/EgS/cQ=
+github.com/konsorten/go-windows-terminal-sequences v1.0.2 h1:DB17ag19krx9CFsz4o3enTrPXyIXCl+2iCXH/aMAp9s=
+github.com/konsorten/go-windows-terminal-sequences v1.0.2/go.mod h1:T0+1ngSBFLxvqU3pZ+m/2kptfBszLMUkC4ZK/EgS/cQ=
+github.com/kr/logfmt v0.0.0-20140226030751-b84e30acd515/go.mod h1:+0opPa2QZZtGFBFZlji/RkVcI2GknAs/DXo4wKdlNEc=
+github.com/kr/pretty v0.1.0 h1:L/CwN0zerZDmRFUapSPitk6f+Q3+0za1rQkzVuMiMFI=
+github.com/kr/pretty v0.1.0/go.mod h1:dAy3ld7l9f0ibDNOQOHHMYYIIbhfbHSm3C4ZsoJORNo=
+github.com/kr/pty v1.0.0/go.mod h1:pFQYn66WHrOpPYNljwOMqo10TkYh1fy3cYio2l3bCsQ=
+github.com/kr/pty v1.1.1/go.mod h1:pFQYn66WHrOpPYNljwOMqo10TkYh1fy3cYio2l3bCsQ=
+github.com/kr/text v0.1.0 h1:45sCR5RtlFHMR4UwH9sdQ5TC8v0qDQCHnXt+kaKSTVE=
+github.com/kr/text v0.1.0/go.mod h1:4Jbv+DJW3UT/LiOwJeYQe1efqtUx/iVham/4vfdArNI=
+github.com/kurin/blazer v0.5.3 h1:SAgYv0TKU0kN/ETfO5ExjNAPyMt2FocO2s/UlCHfjAk=
+github.com/kurin/blazer v0.5.3/go.mod h1:4FCXMUWo9DllR2Do4TtBd377ezyAJ51vB5uTBjt0pGU=
+github.com/lib/pq v1.1.1/go.mod h1:5WUZQaWbwv1U+lTReE5YruASi9Al49XbQIvNi/34Woo=
+github.com/lib/pq v1.2.0 h1:LXpIM/LZ5xGFhOpXAQUIMM1HdyqzVYM13zNdjCEEcA0=
+github.com/lib/pq v1.2.0/go.mod h1:5WUZQaWbwv1U+lTReE5YruASi9Al49XbQIvNi/34Woo=
+github.com/magiconair/properties v1.8.0 h1:LLgXmsheXeRoUOBOjtwPQCWIYqM/LU1ayDtDePerRcY=
+github.com/magiconair/properties v1.8.0/go.mod h1:PppfXfuXeibc/6YijjN8zIbojt8czPbwD3XqdrwzmxQ=
+github.com/magiconair/properties v1.8.1 h1:ZC2Vc7/ZFkGmsVC9KvOjumD+G5lXy2RtTKyzRKO2BQ4=
+github.com/magiconair/properties v1.8.1/go.mod h1:PppfXfuXeibc/6YijjN8zIbojt8czPbwD3XqdrwzmxQ=
+github.com/mattn/go-colorable v0.0.9/go.mod h1:9vuHe8Xs5qXnSaW/c/ABM9alt+Vo+STaOChaDxuIBZU=
+github.com/mattn/go-colorable v0.1.2/go.mod h1:U0ppj6V5qS13XJ6of8GYAs25YV2eR4EVcfRqFIhoBtE=
+github.com/mattn/go-ieproxy v0.0.0-20190610004146-91bb50d98149 h1:HfxbT6/JcvIljmERptWhwa8XzP7H3T+Z2N26gTsaDaA=
+github.com/mattn/go-ieproxy v0.0.0-20190610004146-91bb50d98149/go.mod h1:31jz6HNzdxOmlERGGEc4v/dMssOfmp2p5bT/okiKFFc=
+github.com/mattn/go-ieproxy v0.0.0-20190702010315-6dee0af9227d/go.mod h1:31jz6HNzdxOmlERGGEc4v/dMssOfmp2p5bT/okiKFFc=
+github.com/mattn/go-ieproxy v0.0.0-20190805055040-f9202b1cfdeb h1:hXqqXzQtJbENrsb+rsIqkVqcg4FUJL0SQFGw08Dgivw=
+github.com/mattn/go-ieproxy v0.0.0-20190805055040-f9202b1cfdeb/go.mod h1:31jz6HNzdxOmlERGGEc4v/dMssOfmp2p5bT/okiKFFc=
+github.com/mattn/go-isatty v0.0.4/go.mod h1:M+lRXTBqGeGNdLjl/ufCoiOlB5xdOkqRJdNxMWT7Zi4=
+github.com/mattn/go-isatty v0.0.8/go.mod h1:Iq45c/XA43vh69/j3iqttzPXn0bhXyGjM0Hdxcsrc5s=
+github.com/mattn/go-runewidth v0.0.2/go.mod h1:LwmH8dsx7+W8Uxz3IHJYH5QSwggIsqBzpuz5H//U1FU=
+github.com/mattn/go-runewidth v0.0.3 h1:a+kO+98RDGEfo6asOGMmpodZq4FNtnGP54yps8BzLR4=
+github.com/mattn/go-runewidth v0.0.3/go.mod h1:LwmH8dsx7+W8Uxz3IHJYH5QSwggIsqBzpuz5H//U1FU=
+github.com/mattn/go-runewidth v0.0.4 h1:2BvfKmzob6Bmd4YsL0zygOqfdFnK7GR4QL06Do4/p7Y=
+github.com/mattn/go-runewidth v0.0.4/go.mod h1:LwmH8dsx7+W8Uxz3IHJYH5QSwggIsqBzpuz5H//U1FU=
+github.com/mattn/go-shellwords v1.0.3/go.mod h1:3xCvwCdWdlDJUrvuMn7Wuy9eWs4pE8vqg+NOMyg4B2o=
+github.com/matttproud/golang_protobuf_extensions v1.0.0/go.mod h1:D8He9yQNgCq6Z5Ld7szi9bcBfOoFv/3dc6xSMkL2PC0=
+github.com/matttproud/golang_protobuf_extensions v1.0.1 h1:4hp9jkHxhMHkqkrB3Ix0jegS5sx/RkqARlsWZ6pIwiU=
+github.com/matttproud/golang_protobuf_extensions v1.0.1/go.mod h1:D8He9yQNgCq6Z5Ld7szi9bcBfOoFv/3dc6xSMkL2PC0=
+github.com/mitchellh/go-homedir v1.1.0/go.mod h1:SfyaCUpYCn1Vlf4IUYiD9fPX4A5wJrkLzIz1N1q0pr0=
+github.com/mitchellh/mapstructure v1.1.2 h1:fmNYVwqnSfB9mZU6OS2O6GsXM+wcskZDuKQzvN1EDeE=
+github.com/mitchellh/mapstructure v1.1.2/go.mod h1:FVVH3fgwuzCH5S8UJGiWEs2h04kUh9fWfEaFds41c1Y=
+github.com/modern-go/concurrent v0.0.0-20180228061459-e0a39a4cb421/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
+github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd h1:TRLaZ9cD/w8PVh93nsPXa1VrQ6jlwL5oN8l14QlcNfg=
+github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
+github.com/modern-go/reflect2 v0.0.0-20180701023420-4b7aa43c6742/go.mod h1:bx2lNnkwVCuqBIxFjflWJWanXIb3RllmbCylyMrvgv0=
+github.com/modern-go/reflect2 v1.0.1 h1:9f412s+6RmYXLWZSEzVVgPGK7C2PphHj5RJrvfx9AWI=
+github.com/modern-go/reflect2 v1.0.1/go.mod h1:bx2lNnkwVCuqBIxFjflWJWanXIb3RllmbCylyMrvgv0=
+github.com/montanaflynn/stats v0.0.0-20151014174947-eeaced052adb/go.mod h1:wL8QJuTMNUDYhXwkmfOly8iTdp5TEcJFWZD2D7SIkUc=
+github.com/montanaflynn/stats v0.0.0-20180911141734-db72e6cae808 h1:pmpDGKLw4n82EtrNiLqB+xSz/JQwFOaZuMALYUHwX5s=
+github.com/montanaflynn/stats v0.0.0-20180911141734-db72e6cae808/go.mod h1:wL8QJuTMNUDYhXwkmfOly8iTdp5TEcJFWZD2D7SIkUc=
+github.com/mwitkow/go-conntrack v0.0.0-20161129095857-cc309e4a2223/go.mod h1:qRWi+5nqEBWmkhHvq77mSJWrCKwh8bxhgT7d/eI7P4U=
+github.com/nats-io/jwt v0.2.6/go.mod h1:mQxQ0uHQ9FhEVPIcTSKwx2lqZEpXWWcCgA7R6NrWvvY=
+github.com/nats-io/jwt v0.2.14 h1:wA50KvFz/JXGXMHRygTWsRGh/ixxgC5E3kHvmtGLNf4=
+github.com/nats-io/jwt v0.2.14/go.mod h1:fRYCDE99xlTsqUzISS1Bi75UBJ6ljOJQOAAu5VglpSg=
+github.com/nats-io/nats-server/v2 v2.0.0/go.mod h1:RyVdsHHvY4B6c9pWG+uRLpZ0h0XsqiuKp2XCTurP5LI=
+github.com/nats-io/nats-server/v2 v2.0.4 h1:XOMeQRbhl1lGNTIctPhih6pTa15NGif54Uas6ZW5q7g=
+github.com/nats-io/nats-server/v2 v2.0.4/go.mod h1:AWdGEVbjKRS9ZIx4DSP5eKW48nfFm7q3uiSkP/1KD7M=
+github.com/nats-io/nats.go v1.8.1 h1:6lF/f1/NN6kzUDBz6pyvQDEXO39jqXcWRLu/tKjtOUQ=
+github.com/nats-io/nats.go v1.8.1/go.mod h1:BrFz9vVn0fU3AcH9Vn4Kd7W0NpJ651tD5omQ3M8LwxM=
+github.com/nats-io/nkeys v0.0.2 h1:+qM7QpgXnvDDixitZtQUBDY9w/s9mu1ghS+JIbsrx6M=
+github.com/nats-io/nkeys v0.0.2/go.mod h1:dab7URMsZm6Z/jp9Z5UGa87Uutgc2mVpXLC4B7TDb/4=
+github.com/nats-io/nkeys v0.1.0 h1:qMd4+pRHgdr1nAClu+2h/2a5F2TmKcCzjCDazVgRoX4=
+github.com/nats-io/nkeys v0.1.0/go.mod h1:xpnFELMwJABBLVhffcfd1MZx6VsNRFpEugbxziKVo7w=
+github.com/nats-io/nuid v1.0.1 h1:5iA8DT8V7q8WK2EScv2padNa/rTESc1KdnPw4TC2paw=
+github.com/nats-io/nuid v1.0.1/go.mod h1:19wcPz3Ph3q0Jbyiqsd0kePYG7A95tJPxeL+1OSON2c=
+github.com/ngaut/pools v0.0.0-20180318154953-b7bc8c42aac7 h1:7KAv7KMGTTqSmYZtNdcNTgsos+vFzULLwyElndwn+5c=
+github.com/ngaut/pools v0.0.0-20180318154953-b7bc8c42aac7/go.mod h1:iWMfgwqYW+e8n5lC/jjNEhwcjbRDpl5NT7n2h+4UNcI=
+github.com/ngaut/sync2 v0.0.0-20141008032647-7a24ed77b2ef h1:K0Fn+DoFqNqktdZtdV3bPQ/0cuYh2H4rkg0tytX/07k=
+github.com/ngaut/sync2 v0.0.0-20141008032647-7a24ed77b2ef/go.mod h1:7WjlapSfwQyo6LNmIvEWzsW1hbBQfpUO4JWnuQRmva8=
+github.com/nicksnyder/go-i18n v1.10.0/go.mod h1:HrK7VCrbOvQoUAQ7Vpy7i87N7JZZZ7R2xBGjv0j365Q=
+github.com/oklog/ulid v1.3.1/go.mod h1:CirwcVhetQ6Lv90oh/F+FBtV6XMibvdAFo93nm5qn4U=
+github.com/olekukonko/tablewriter v0.0.0-20170122224234-a0225b3f23b5/go.mod h1:vsDQFd/mU46D+Z4whnwzcISnGGzXWMclvtLoiIKAKIo=
+github.com/onsi/ginkgo v1.6.0/go.mod h1:lLunBs/Ym6LB5Z9jYTR76FiuTmxDTDusOGeTQH+WWjE=
+github.com/onsi/ginkgo v1.7.0 h1:WSHQ+IS43OoUrWtD1/bbclrwK8TTH5hzp+umCiuxHgs=
+github.com/onsi/ginkgo v1.7.0/go.mod h1:lLunBs/Ym6LB5Z9jYTR76FiuTmxDTDusOGeTQH+WWjE=
+github.com/onsi/ginkgo v1.10.1 h1:q/mM8GF/n0shIN8SaAZ0V+jnLPzen6WIVZdiwrRlMlo=
+github.com/onsi/ginkgo v1.10.1/go.mod h1:lLunBs/Ym6LB5Z9jYTR76FiuTmxDTDusOGeTQH+WWjE=
+github.com/onsi/gomega v1.4.2/go.mod h1:ex+gbHU/CVuBBDIJjb2X0qEXbFg53c61hWP/1CpauHY=
+github.com/onsi/gomega v1.4.3 h1:RE1xgDvH7imwFD45h+u2SgIfERHlS2yNG4DObb5BSKU=
+github.com/onsi/gomega v1.4.3/go.mod h1:ex+gbHU/CVuBBDIJjb2X0qEXbFg53c61hWP/1CpauHY=
+github.com/onsi/gomega v1.7.0 h1:XPnZz8VVBHjVsy1vzJmRwIcSwiUO+JFfrv/xGiigmME=
+github.com/onsi/gomega v1.7.0/go.mod h1:ex+gbHU/CVuBBDIJjb2X0qEXbFg53c61hWP/1CpauHY=
+github.com/opentracing/basictracer-go v1.0.0 h1:YyUAhaEfjoWXclZVJ9sGoNct7j4TVk7lZWlQw5UXuoo=
+github.com/opentracing/basictracer-go v1.0.0/go.mod h1:QfBfYuafItcjQuMwinw9GhYKwFXS9KnPs5lxoYwgW74=
+github.com/opentracing/opentracing-go v1.0.2/go.mod h1:UkNAQd3GIcIGf0SeVgPpRdFStlNbqXla1AfSYxPUl2o=
+github.com/opentracing/opentracing-go v1.1.0 h1:pWlfV3Bxv7k65HYwkikxat0+s3pV4bsqf19k25Ur8rU=
+github.com/opentracing/opentracing-go v1.1.0/go.mod h1:UkNAQd3GIcIGf0SeVgPpRdFStlNbqXla1AfSYxPUl2o=
+github.com/pelletier/go-toml v1.2.0 h1:T5zMGML61Wp+FlcbWjRDT7yAxhJNAiPPLOFECq181zc=
+github.com/pelletier/go-toml v1.2.0/go.mod h1:5z9KED0ma1S8pY6P1sdut58dfprrGBbd/94hg7ilaic=
+github.com/pelletier/go-toml v1.3.0/go.mod h1:PN7xzY2wHTK0K9p34ErDQMlFxa51Fk0OUruD3k1mMwo=
+github.com/pelletier/go-toml v1.4.0 h1:u3Z1r+oOXJIkxqw34zVhyPgjBsm6X2wn21NWs/HfSeg=
+github.com/pelletier/go-toml v1.4.0/go.mod h1:PN7xzY2wHTK0K9p34ErDQMlFxa51Fk0OUruD3k1mMwo=
+github.com/peterh/liner v1.1.0 h1:f+aAedNJA6uk7+6rXsYBnhdo4Xux7ESLe+kcuVUF5os=
+github.com/peterh/liner v1.1.0/go.mod h1:CRroGNssyjTd/qIG2FyxByd2S8JEAZXBl4qUrZf8GS0=
+github.com/pierrec/lz4 v0.0.0-20190327172049-315a67e90e41/go.mod h1:3/3N9NVKO0jef7pBehbT1qWhCMrIgbYNnFAZCqQ5LRc=
+github.com/pierrec/lz4 v2.2.7+incompatible h1:Eerk9aiqeZo2QzsbWOAsELUf9ddvAxEdMY9LYze/DEc=
+github.com/pierrec/lz4 v2.2.7+incompatible/go.mod h1:pdkljMzZIN41W+lC3N2tnIh5sFi+IEE17M5jbnwPHcY=
+github.com/pingcap/check v0.0.0-20190102082844-67f458068fc8 h1:USx2/E1bX46VG32FIw034Au6seQ2fY9NEILmNh/UlQg=
+github.com/pingcap/check v0.0.0-20190102082844-67f458068fc8/go.mod h1:B1+S9LNcuMyLH/4HMTViQOJevkGiik3wW2AN9zb2fNQ=
+github.com/pingcap/errcode v0.0.0-20180921232412-a1a7271709d9 h1:KH4f4Si9XK6/IW50HtoaiLIFHGkapOM6w83za47UYik=
+github.com/pingcap/errcode v0.0.0-20180921232412-a1a7271709d9/go.mod h1:4b2X8xSqxIroj/IZ9MX/VGZhAwc11wB9wRIzHvz6SeM=
+github.com/pingcap/errors v0.11.0/go.mod h1:Oi8TUi2kEtXXLMJk9l1cGmz20kV3TaQ0usTwv5KuLY8=
+github.com/pingcap/errors v0.11.4 h1:lFuQV/oaUMGcD2tqt+01ROSmJs75VG1ToEOkZIZ4nE4=
+github.com/pingcap/errors v0.11.4/go.mod h1:Oi8TUi2kEtXXLMJk9l1cGmz20kV3TaQ0usTwv5KuLY8=
+github.com/pingcap/failpoint v0.0.0-20190512135322-30cc7431d99c h1:hvQd3aOLKLF7xvRV6DzvPkKY4QXzfVbjU1BhW0d9yL8=
+github.com/pingcap/failpoint v0.0.0-20190512135322-30cc7431d99c/go.mod h1:DNS3Qg7bEDhU6EXNHF+XSv/PGznQaMJ5FWvctpm6pQI=
+github.com/pingcap/goleveldb v0.0.0-20171020122428-b9ff6c35079e h1:P73/4dPCL96rGrobssy1nVy2VaVpNCuLpCbr+FEaTA8=
+github.com/pingcap/goleveldb v0.0.0-20171020122428-b9ff6c35079e/go.mod h1:O17XtbryoCJhkKGbT62+L2OlrniwqiGLSqrmdHCMzZw=
+github.com/pingcap/kvproto v0.0.0-20190822090350-11ea838aedf7/go.mod h1:QMdbTAXCHzzygQzqcG9uVUgU2fKeSN1GmfMiykdSzzY=
+github.com/pingcap/kvproto v0.0.0-20190910074005-0e61b6f435c1/go.mod h1:QMdbTAXCHzzygQzqcG9uVUgU2fKeSN1GmfMiykdSzzY=
+github.com/pingcap/kvproto v0.0.0-20191022073741-81b2a2d9e986 h1:XFh7n8Cheo00pakfhpUofnlptHCuz9lkp4p/jXPb8lM=
+github.com/pingcap/kvproto v0.0.0-20191022073741-81b2a2d9e986/go.mod h1:QMdbTAXCHzzygQzqcG9uVUgU2fKeSN1GmfMiykdSzzY=
+github.com/pingcap/log v0.0.0-20190715063458-479153f07ebd/go.mod h1:WpHUKhNZ18v116SvGrmjkA9CBhYmuUTKL+p8JC9ANEw=
+github.com/pingcap/log v0.0.0-20191012051959-b742a5d432e9 h1:AJD9pZYm72vMgPcQDww9rkZ1DnWfl0pXV3BOWlkYIjA=
+github.com/pingcap/log v0.0.0-20191012051959-b742a5d432e9/go.mod h1:4rbK1p9ILyIfb6hU7OG2CiWSqMXnp3JMbiaVJ6mvoY8=
+github.com/pingcap/parser v0.0.0-20191021083151-7c64f78a5100 h1:TRyps2d+2TsJv1Vk4S2D+5COMDVKClnAO5aNmGGVyj0=
+github.com/pingcap/parser v0.0.0-20191021083151-7c64f78a5100/go.mod h1:1FNvfp9+J0wvc4kl8eGNh7Rqrxveg15jJoWo/a0uHwA=
+github.com/pingcap/pd v1.1.0-beta.0.20190923032047-5c648dc365e0 h1:GIEq+wZfrl2bcJxpuSrEH4H7/nlf5YdmpS+dU9lNIt8=
+github.com/pingcap/pd v1.1.0-beta.0.20190923032047-5c648dc365e0/go.mod h1:G/6rJpnYwM0LKMec2rI82/5Kg6GaZMvlfB+e6/tvYmI=
+github.com/pingcap/tidb v1.1.0-beta.0.20191023070859-58fc7d44f73b h1:6GfcYOX9/CCxPnNOivVxiDYXbZrCHU1mRp691iw9EYs=
+github.com/pingcap/tidb v1.1.0-beta.0.20191023070859-58fc7d44f73b/go.mod h1:YfrHdQ613A+E2FSugyXOdJmeZQbXNjpXX2doNe8MGj8=
+github.com/pingcap/tidb-tools v2.1.3-0.20190321065848-1e8b48f5c168+incompatible h1:MkWCxgZpJBgY2f4HtwWMMFzSBb3+JPzeJgF3VrXE/bU=
+github.com/pingcap/tidb-tools v2.1.3-0.20190321065848-1e8b48f5c168+incompatible/go.mod h1:XGdcy9+yqlDSEMTpOXnwf3hiTeqrV6MN/u1se9N8yIM=
+github.com/pingcap/tipb v0.0.0-20191015023537-709b39e7f8bb/go.mod h1:RtkHW8WbcNxj8lsbzjaILci01CtYnYbIkQhjyZWrWVI=
+github.com/pingcap/tipb v0.0.0-20191022094114-a2e8c3fa634b h1:DZ0cTsn4lGMNaRjkUFKBtHn4s2F8KFMm83lWvSo+x7c=
+github.com/pingcap/tipb v0.0.0-20191022094114-a2e8c3fa634b/go.mod h1:RtkHW8WbcNxj8lsbzjaILci01CtYnYbIkQhjyZWrWVI=
+github.com/pkg/errors v0.8.0/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
+github.com/pkg/errors v0.8.1 h1:iURUrRGxPUNPdy5/HRSm+Yj6okJ6UtLINN0Q9M4+h3I=
+github.com/pkg/errors v0.8.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
+github.com/pkg/profile v1.2.1/go.mod h1:hJw3o1OdXxsrSjjVksARp5W95eeEaEfptyVZyv6JUPA=
+github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
+github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/prometheus/client_golang v0.8.0/go.mod h1:7SWBe2y4D6OKWSNQJUaRYU/AaXPKyh/dDVn+NZz0KFw=
+github.com/prometheus/client_golang v0.9.0/go.mod h1:7SWBe2y4D6OKWSNQJUaRYU/AaXPKyh/dDVn+NZz0KFw=
+github.com/prometheus/client_golang v0.9.1/go.mod h1:7SWBe2y4D6OKWSNQJUaRYU/AaXPKyh/dDVn+NZz0KFw=
+github.com/prometheus/client_golang v0.9.3/go.mod h1:/TN21ttK/J9q6uSwhBd54HahCDft0ttaMvbicHlPoso=
+github.com/prometheus/client_golang v1.0.0 h1:vrDKnkGzuGvhNAL56c7DBz29ZL+KxnoR0x7enabFceM=
+github.com/prometheus/client_golang v1.0.0/go.mod h1:db9x61etRT2tGnBNRi70OPL5FsnadC4Ky3P0J6CfImo=
+github.com/prometheus/client_golang v1.1.0 h1:BQ53HtBmfOitExawJ6LokA4x8ov/z0SYYb0+HxJfRI8=
+github.com/prometheus/client_golang v1.1.0/go.mod h1:I1FGZT9+L76gKKOs5djB6ezCbFQP1xR9D75/vuwEF3g=
+github.com/prometheus/client_model v0.0.0-20170216185247-6f3806018612/go.mod h1:MbSGuTsp3dbXC40dX6PRTWyKYBIrTGTE9sqQNg2J8bo=
+github.com/prometheus/client_model v0.0.0-20171117100541-99fa1f4be8e5/go.mod h1:MbSGuTsp3dbXC40dX6PRTWyKYBIrTGTE9sqQNg2J8bo=
+github.com/prometheus/client_model v0.0.0-20180712105110-5c3871d89910/go.mod h1:MbSGuTsp3dbXC40dX6PRTWyKYBIrTGTE9sqQNg2J8bo=
+github.com/prometheus/client_model v0.0.0-20190129233127-fd36f4220a90 h1:S/YWwWx/RA8rT8tKFRuGUZhuA90OyIBpPCXkcbwU8DE=
+github.com/prometheus/client_model v0.0.0-20190129233127-fd36f4220a90/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
+github.com/prometheus/client_model v0.0.0-20190812154241-14fe0d1b01d4 h1:gQz4mCbXsO+nc9n1hCxHcGA3Zx3Eo+UHZoInFGUIXNM=
+github.com/prometheus/client_model v0.0.0-20190812154241-14fe0d1b01d4/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
+github.com/prometheus/common v0.0.0-20180518154759-7600349dcfe1/go.mod h1:daVV7qP5qjZbuso7PdcryaAu0sAZbrN9i7WWcTMWvro=
+github.com/prometheus/common v0.0.0-20181020173914-7e9e6cabbd39/go.mod h1:daVV7qP5qjZbuso7PdcryaAu0sAZbrN9i7WWcTMWvro=
+github.com/prometheus/common v0.0.0-20181113130724-41aa239b4cce/go.mod h1:daVV7qP5qjZbuso7PdcryaAu0sAZbrN9i7WWcTMWvro=
+github.com/prometheus/common v0.4.0/go.mod h1:TNfzLD0ON7rHzMJeJkieUDPYmFC7Snx/y86RQel1bk4=
+github.com/prometheus/common v0.4.1 h1:K0MGApIoQvMw27RTdJkPbr3JZ7DNbtxQNyi5STVM6Kw=
+github.com/prometheus/common v0.4.1/go.mod h1:TNfzLD0ON7rHzMJeJkieUDPYmFC7Snx/y86RQel1bk4=
+github.com/prometheus/common v0.6.0 h1:kRhiuYSXR3+uv2IbVbZhUxK5zVD/2pp3Gd2PpvPkpEo=
+github.com/prometheus/common v0.6.0/go.mod h1:eBmuwkDJBwy6iBfxCBob6t6dR6ENT/y+J+Zk0j9GMYc=
+github.com/prometheus/procfs v0.0.0-20180612222113-7d6f385de8be/go.mod h1:c3At6R/oaqEKCNdg8wHV1ftS6bRYblBhIjjI8uT2IGk=
+github.com/prometheus/procfs v0.0.0-20181005140218-185b4288413d/go.mod h1:c3At6R/oaqEKCNdg8wHV1ftS6bRYblBhIjjI8uT2IGk=
+github.com/prometheus/procfs v0.0.0-20190507164030-5867b95ac084/go.mod h1:TjEm7ze935MbeOT/UhFTIMYKhuLP4wbCsTZCD3I8kEA=
+github.com/prometheus/procfs v0.0.2 h1:6LJUbpNm42llc4HRCuvApCSWB/WfhuNo9K98Q9sNGfs=
+github.com/prometheus/procfs v0.0.2/go.mod h1:TjEm7ze935MbeOT/UhFTIMYKhuLP4wbCsTZCD3I8kEA=
+github.com/prometheus/procfs v0.0.3/go.mod h1:4A/X28fw3Fc593LaREMrKMqOKvUAntwMDaekg4FpcdQ=
+github.com/prometheus/procfs v0.0.4 h1:w8DjqFMJDjuVwdZBQoOozr4MVWOnwF7RcL/7uxBjY78=
+github.com/prometheus/procfs v0.0.4/go.mod h1:4A/X28fw3Fc593LaREMrKMqOKvUAntwMDaekg4FpcdQ=
+github.com/prometheus/tsdb v0.7.1/go.mod h1:qhTCs0VvXwvX/y3TZrWD7rabWM+ijKTux40TwIPHuXU=
+github.com/rakyll/statik v0.1.6 h1:uICcfUXpgqtw2VopbIncslhAmE5hwc4g20TEyEENBNs=
+github.com/rakyll/statik v0.1.6/go.mod h1:OEi9wJV/fMUAGx1eNjq75DKDsJVuEv1U0oYdX6GX8Zs=
+github.com/rcrowley/go-metrics v0.0.0-20181016184325-3113b8401b8a h1:9ZKAASQSHhDYGoxY8uLVpewe1GDZ2vu2Tr/vTdVAkFQ=
+github.com/rcrowley/go-metrics v0.0.0-20181016184325-3113b8401b8a/go.mod h1:bCqnVzQkZxMG4s8nGwiZ5l3QUCyqpo9Y+/ZMZ9VjZe4=
+github.com/rcrowley/go-metrics v0.0.0-20190826022208-cac0b30c2563 h1:dY6ETXrvDG7Sa4vE8ZQG4yqWg6UnOcbqTAahkV813vQ=
+github.com/rcrowley/go-metrics v0.0.0-20190826022208-cac0b30c2563/go.mod h1:bCqnVzQkZxMG4s8nGwiZ5l3QUCyqpo9Y+/ZMZ9VjZe4=
+github.com/remyoudompheng/bigfft v0.0.0-20190512091148-babf20351dd7/go.mod h1:qqbHyh8v60DhA7CoWK5oRCqLrMHRGoxYCSS9EjAz6Eo=
+github.com/remyoudompheng/bigfft v0.0.0-20190728182440-6a916e37a237 h1:HQagqIiBmr8YXawX/le3+O26N+vPPC1PtjaF3mwnook=
+github.com/remyoudompheng/bigfft v0.0.0-20190728182440-6a916e37a237/go.mod h1:qqbHyh8v60DhA7CoWK5oRCqLrMHRGoxYCSS9EjAz6Eo=
+github.com/rogpeppe/fastuuid v0.0.0-20150106093220-6724a57986af/go.mod h1:XWv6SoW27p1b0cqNHllgS5HIMJraePCO15w5zCzIWYg=
+github.com/rwcarlsen/goexif v0.0.0-20190401172101-9e8deecbddbd h1:CmH9+J6ZSsIjUK3dcGsnCnO41eRBOnY12zwkn5qVwgc=
+github.com/rwcarlsen/goexif v0.0.0-20190401172101-9e8deecbddbd/go.mod h1:hPqNNc0+uJM6H+SuU8sEs5K5IQeKccPqeSjfgcKGgPk=
+github.com/seaweedfs/fuse v0.0.0-20190510212405-310228904eff h1:uLd5zBvf5OA67wcVRePHrFt60bR4LSskaVhgVwyk0Jg=
+github.com/seaweedfs/fuse v0.0.0-20190510212405-310228904eff/go.mod h1:cubdLmQFqEUZ9vNJrznhgc3m3VMAJi/nY2Ix2axXkG0=
+github.com/sergi/go-diff v1.0.1-0.20180205163309-da645544ed44/go.mod h1:0CfEIISq7TuYL3j771MWULgwwjU+GofnZX9QAmXWZgo=
+github.com/shirou/gopsutil v2.18.10+incompatible h1:cy84jW6EVRPa5g9HAHrlbxMSIjBhDSX0OFYyMYminYs=
+github.com/shirou/gopsutil v2.18.10+incompatible/go.mod h1:5b4v6he4MtMOwMlS0TUMTu2PcXUg8+E1lC7eC3UO/RA=
+github.com/shurcooL/httpfs v0.0.0-20171119174359-809beceb2371/go.mod h1:ZY1cvUeJuFPAdZ/B6v7RHavJWZn2YPVFQ1OSXhCGOkg=
+github.com/shurcooL/vfsgen v0.0.0-20181020040650-a97a25d856ca/go.mod h1:TrYk7fJVaAttu97ZZKrO9UbRa8izdowaMIZcxYMbVaw=
+github.com/sirupsen/logrus v1.0.5/go.mod h1:pMByvHTf9Beacp5x1UXfOR9xyW/9antXMhjMPG0dEzc=
+github.com/sirupsen/logrus v1.2.0 h1:juTguoYk5qI21pwyTXY3B3Y5cOTH3ZUyZCg1v/mihuo=
+github.com/sirupsen/logrus v1.2.0/go.mod h1:LxeOpSwHxABJmUn/MG1IvRgCAasNZTLOkJPxbbu5VWo=
+github.com/sirupsen/logrus v1.4.2 h1:SPIRibHv4MatM3XXNO2BJeFLZwZ2LvZgfQ5+UNI2im4=
+github.com/sirupsen/logrus v1.4.2/go.mod h1:tLMulIdttU9McNUspp0xgXVQah82FyeX6MwdIuYE2rE=
+github.com/soheilhy/cmux v0.1.4 h1:0HKaf1o97UwFjHH9o5XsHUOF+tqmdA7KEzXLpiyaw0E=
+github.com/soheilhy/cmux v0.1.4/go.mod h1:IM3LyeVVIOuxMH7sFAkER9+bJ4dT7Ms6E4xg4kGIyLM=
+github.com/spaolacci/murmur3 v0.0.0-20180118202830-f09979ecbc72/go.mod h1:JwIasOWyU6f++ZhiEuf87xNszmSA2myDM2Kzu9HwQUA=
+github.com/spaolacci/murmur3 v1.1.0 h1:7c1g84S4BPRrfL5Xrdp6fOJ206sU9y293DDHaoy0bLI=
+github.com/spaolacci/murmur3 v1.1.0/go.mod h1:JwIasOWyU6f++ZhiEuf87xNszmSA2myDM2Kzu9HwQUA=
+github.com/spf13/afero v1.1.2 h1:m8/z1t7/fwjysjQRYbP0RD+bUIF/8tJwPdEZsI83ACI=
+github.com/spf13/afero v1.1.2/go.mod h1:j4pytiNVoe2o6bmDsKpLACNPDBIoEAkihy7loJ1B0CQ=
+github.com/spf13/afero v1.2.2 h1:5jhuqJyZCZf2JRofRvN/nIFgIWNzPa3/Vz8mYylgbWc=
+github.com/spf13/afero v1.2.2/go.mod h1:9ZxEEn6pIJ8Rxe320qSDBk6AsU0r9pR7Q4OcevTdifk=
+github.com/spf13/cast v1.3.0 h1:oget//CVOEoFewqQxwr0Ej5yjygnqGkvggSE/gB35Q8=
+github.com/spf13/cast v1.3.0/go.mod h1:Qx5cxh0v+4UWYiBimWS+eyWzqEqokIECu5etghLkUJE=
+github.com/spf13/cobra v0.0.3/go.mod h1:1l0Ry5zgKvJasoi3XT1TypsSe7PqH0Sj9dhYf7v3XqQ=
+github.com/spf13/jwalterweatherman v1.0.0 h1:XHEdyB+EcvlqZamSM4ZOMGlc93t6AcsBEu9Gc1vn7yk=
+github.com/spf13/jwalterweatherman v1.0.0/go.mod h1:cQK4TGJAtQXfYWX+Ddv3mKDzgVb68N+wFjFa4jdeBTo=
+github.com/spf13/jwalterweatherman v1.1.0 h1:ue6voC5bR5F8YxI5S67j9i582FU4Qvo2bmqnqMYADFk=
+github.com/spf13/jwalterweatherman v1.1.0/go.mod h1:aNWZUN0dPAAO/Ljvb5BEdw96iTZ0EXowPYD95IqWIGo=
+github.com/spf13/pflag v1.0.1/go.mod h1:DYY7MBk1bdzusC3SYhjObp+wFpr4gzcvqqNjLnInEg4=
+github.com/spf13/pflag v1.0.3 h1:zPAT6CGy6wXeQ7NtTnaTerfKOsV6V6F8agHXFiazDkg=
+github.com/spf13/pflag v1.0.3/go.mod h1:DYY7MBk1bdzusC3SYhjObp+wFpr4gzcvqqNjLnInEg4=
+github.com/spf13/viper v1.4.0 h1:yXHLWeravcrgGyFSyCgdYpXQ9dR9c/WED3pg1RhxqEU=
+github.com/spf13/viper v1.4.0/go.mod h1:PTJ7Z/lr49W6bUbkmS1V3by4uWynFiR9p7+dSq/yZzE=
+github.com/streadway/amqp v0.0.0-20190404075320-75d898a42a94 h1:0ngsPmuP6XIjiFRNFYlvKwSr5zff2v+uPHaffZ6/M4k=
+github.com/streadway/amqp v0.0.0-20190404075320-75d898a42a94/go.mod h1:AZpEONHx3DKn8O/DFsRAY58/XVQiIPMTMB1SddzLXVw=
+github.com/streadway/amqp v0.0.0-20190827072141-edfb9018d271 h1:WhxRHzgeVGETMlmVfqhRn8RIeeNoPr2Czh33I4Zdccw=
+github.com/streadway/amqp v0.0.0-20190827072141-edfb9018d271/go.mod h1:AZpEONHx3DKn8O/DFsRAY58/XVQiIPMTMB1SddzLXVw=
+github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
+github.com/stretchr/objx v0.1.1 h1:2vfRuCMp5sSVIDSqO8oNnWJq7mPa6KVP3iPIwFBuy8A=
+github.com/stretchr/objx v0.1.1/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
+github.com/stretchr/objx v0.2.0 h1:Hbg2NidpLE8veEBkEZTL3CvlkUIVzuU9jDplZO54c48=
+github.com/stretchr/objx v0.2.0/go.mod h1:qt09Ya8vawLte6SNmTgCsAVtYtaKzEcn8ATUoHMkEqE=
+github.com/stretchr/testify v1.2.2/go.mod h1:a8OnRcib4nhh0OaRAV+Yts87kKdq0PP7pXfy6kDkUVs=
+github.com/stretchr/testify v1.3.0 h1:TivCn/peBQ7UY8ooIcPgZFpTNSz0Q2U6UrFlUfqbe0Q=
+github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
+github.com/stretchr/testify v1.4.0 h1:2E4SXV/wtOkTonXsotYi4li6zVWxYlZuYNCXe9XRJyk=
+github.com/stretchr/testify v1.4.0/go.mod h1:j7eGeouHqKxXV5pUuKE4zz7dFj8WfuZ+81PSLYec5m4=
+github.com/struCoder/pidusage v0.1.2/go.mod h1:pWBlW3YuSwRl6h7R5KbvA4N8oOqe9LjaKW5CwT1SPjI=
+github.com/syndtr/goleveldb v0.0.0-20180815032940-ae2bd5eed72d/go.mod h1:Z4AUp2Km+PwemOoO/VB5AOx9XSsIItzFjoJlOSiYmn0=
+github.com/syndtr/goleveldb v1.0.0 h1:fBdIW9lB4Iz0n9khmH8w27SJ3QEJ7+IgjPEwGSZiFdE=
+github.com/syndtr/goleveldb v1.0.0/go.mod h1:ZVVdQEZoIme9iO1Ch2Jdy24qqXrMMOU6lpPAyBWyWuQ=
+github.com/tiancaiamao/appdash v0.0.0-20181126055449-889f96f722a2/go.mod h1:2PfKggNGDuadAa0LElHrByyrz4JPZ9fFx6Gs7nx7ZZU=
+github.com/tidwall/gjson v1.3.2 h1:+7p3qQFaH3fOMXAJSrdZwGKcOO/lYdGS0HqGhPqDdTI=
+github.com/tidwall/gjson v1.3.2/go.mod h1:P256ACg0Mn+j1RXIDXoss50DeIABTYK1PULOJHhxOls=
+github.com/tidwall/match v1.0.1 h1:PnKP62LPNxHKTwvHHZZzdOAOCtsJTjo6dZLCwpKm5xc=
+github.com/tidwall/match v1.0.1/go.mod h1:LujAq0jyVjBy028G1WhWfIzbpQfMO8bBZ6Tyb0+pL9E=
+github.com/tidwall/pretty v1.0.0 h1:HsD+QiTn7sK6flMKIvNmpqz1qrpP3Ps6jOKIKMooyg4=
+github.com/tidwall/pretty v1.0.0/go.mod h1:XNkn88O1ChpSDQmQeStsy+sBenx6DDtFZJxhVysOjyk=
+github.com/tmc/grpc-websocket-proxy v0.0.0-20170815181823-89b8d40f7ca8/go.mod h1:ncp9v5uamzpCO7NfCPTXjqaC+bZgJeR0sMTm6dMHP7U=
+github.com/tmc/grpc-websocket-proxy v0.0.0-20171017195756-830351dc03c6/go.mod h1:ncp9v5uamzpCO7NfCPTXjqaC+bZgJeR0sMTm6dMHP7U=
+github.com/tmc/grpc-websocket-proxy v0.0.0-20190109142713-0ad062ec5ee5 h1:LnC5Kc/wtumK+WB441p7ynQJzVuNRJiqddSIE3IlSEQ=
+github.com/tmc/grpc-websocket-proxy v0.0.0-20190109142713-0ad062ec5ee5/go.mod h1:ncp9v5uamzpCO7NfCPTXjqaC+bZgJeR0sMTm6dMHP7U=
+github.com/uber-go/atomic v1.3.2/go.mod h1:/Ct5t2lcmbJ4OSe/waGBoaVvVqtO0bmtfVNex1PFV8g=
+github.com/uber-go/atomic v1.4.0 h1:yOuPqEq4ovnhEjpHmfFwsqBXDYbQeT6Nb0bwD6XnD5o=
+github.com/uber-go/atomic v1.4.0/go.mod h1:/Ct5t2lcmbJ4OSe/waGBoaVvVqtO0bmtfVNex1PFV8g=
+github.com/uber/jaeger-client-go v2.15.0+incompatible/go.mod h1:WVhlPFC8FDjOFMMWRy2pZqQJSXxYSwNYOkTr/Z6d3Kk=
+github.com/uber/jaeger-client-go v2.17.0+incompatible h1:35tpDuT3k0oBiN/aGoSWuiFaqKgKZSciSMnWrazhSHE=
+github.com/uber/jaeger-client-go v2.17.0+incompatible/go.mod h1:WVhlPFC8FDjOFMMWRy2pZqQJSXxYSwNYOkTr/Z6d3Kk=
+github.com/uber/jaeger-lib v1.5.0/go.mod h1:ComeNDZlWwrWnDv8aPp0Ba6+uUTzImX/AauajbLI56U=
+github.com/uber/jaeger-lib v2.0.0+incompatible h1:iMSCV0rmXEogjNWPh2D0xk9YVKvrtGoHJNe9ebLu/pw=
+github.com/uber/jaeger-lib v2.0.0+incompatible/go.mod h1:ComeNDZlWwrWnDv8aPp0Ba6+uUTzImX/AauajbLI56U=
+github.com/ugorji/go v1.1.2/go.mod h1:hnLbHMwcvSihnDhEfx2/BzKp2xb0Y+ErdfYcrs9tkJQ=
+github.com/ugorji/go v1.1.4 h1:j4s+tAvLfL3bZyefP2SEWmhBzmuIlH/eqNuPdFPgngw=
+github.com/ugorji/go v1.1.4/go.mod h1:uQMGLiO92mf5W77hV/PUCpI3pbzQx3CRekS0kk+RGrc=
+github.com/ugorji/go/codec v0.0.0-20190204201341-e444a5086c43/go.mod h1:iT03XoTwV7xq/+UGwKO3UbC1nNNlopQiY61beSdrtOA=
+github.com/unrolled/render v0.0.0-20171102162132-65450fb6b2d3/go.mod h1:tu82oB5W2ykJRVioYsB+IQKcft7ryBr7w12qMBUPyXg=
+github.com/unrolled/render v0.0.0-20180914162206-b9786414de4d h1:ggUgChAeyge4NZ4QUw6lhHsVymzwSDJOZcE0s2X8S20=
+github.com/unrolled/render v0.0.0-20180914162206-b9786414de4d/go.mod h1:tu82oB5W2ykJRVioYsB+IQKcft7ryBr7w12qMBUPyXg=
+github.com/urfave/cli v1.20.0/go.mod h1:70zkFmudgCuE/ngEzBv17Jvp/497gISqfk5gWijbERA=
+github.com/urfave/negroni v0.3.0/go.mod h1:Meg73S6kFm/4PpbYdq35yYWoCZ9mS/YSx+lKnmiohz4=
+github.com/willf/bitset v1.1.10 h1:NotGKqX0KwQ72NUzqrjZq5ipPNDQex9lo3WpaS8L2sc=
+github.com/willf/bitset v1.1.10/go.mod h1:RjeCKbqT1RxIR/KWY6phxZiaY1IyutSBfGjNPySAYV4=
+github.com/willf/bloom v2.0.3+incompatible h1:QDacWdqcAUI1MPOwIQZRy9kOR7yxfyEmxX8Wdm2/JPA=
+github.com/willf/bloom v2.0.3+incompatible/go.mod h1:MmAltL9pDMNTrvUkxdg0k0q5I0suxmuwp3KbyrZLOZ8=
+github.com/wsxiaoys/terminal v0.0.0-20160513160801-0940f3fc43a0 h1:3UeQBvD0TFrlVjOeLOBz+CPAI8dnbqNSVwUwRrkp7vQ=
+github.com/wsxiaoys/terminal v0.0.0-20160513160801-0940f3fc43a0/go.mod h1:IXCdmsXIht47RaVFLEdVnh1t+pgYtTAhQGj73kz+2DM=
+github.com/xdg/scram v0.0.0-20180814205039-7eeb5667e42c/go.mod h1:lB8K/P019DLNhemzwFU4jHLhdvlE6uDZjXFejJXr49I=
+github.com/xdg/stringprep v1.0.0/go.mod h1:Jhud4/sHMO4oL310DaZAKk9ZaJ08SJfe+sJh0HrGL1Y=
+github.com/xiang90/probing v0.0.0-20190116061207-43a291ad63a2 h1:eY9dn8+vbi4tKz5Qo6v2eYzo7kUS51QINcR5jNpbZS8=
+github.com/xiang90/probing v0.0.0-20190116061207-43a291ad63a2/go.mod h1:UETIi67q53MR2AWcXfiuqkDkRtnGDLqkBTpCHuJHxtU=
+github.com/xordataexchange/crypt v0.0.3-0.20170626215501-b2862e3d0a77/go.mod h1:aYKd//L2LvnjZzWKhF00oedf4jCCReLcmhLdhm1A27Q=
+github.com/yookoala/realpath v1.0.0/go.mod h1:gJJMA9wuX7AcqLy1+ffPatSCySA1FQ2S8Ya9AIoYBpE=
+go.etcd.io/bbolt v1.3.2 h1:Z/90sZLPOeCy2PwprqkFa25PdkusRzaj9P8zm/KNyvk=
+go.etcd.io/bbolt v1.3.2/go.mod h1:IbVyRI1SCnLcuJnV2u8VeU0CEYM7e686BmAb1XKL+uU=
+go.etcd.io/bbolt v1.3.3 h1:MUGmc65QhB3pIlaQ5bB4LwqSj6GIonVJXpZiaKNyaKk=
+go.etcd.io/bbolt v1.3.3/go.mod h1:IbVyRI1SCnLcuJnV2u8VeU0CEYM7e686BmAb1XKL+uU=
+go.etcd.io/etcd v0.0.0-20190320044326-77d4b742cdbf/go.mod h1:KSGwdbiFchh5KIC9My2+ZVl5/3ANcwohw50dpPwa2cw=
+go.etcd.io/etcd v3.3.15+incompatible h1:0VpOVCF6EFnJptt8Jh0EWEHO4j2fepyV1fpu9xz/UoQ=
+go.etcd.io/etcd v3.3.15+incompatible/go.mod h1:yaeTdrJi5lOmYerz05bd8+V7KubZs8YSFZfzsF9A6aI=
+go.opencensus.io v0.15.0/go.mod h1:UffZAU+4sDEINUGP/B7UfBBkq4fqLu9zXAX7ke6CHW0=
+go.opencensus.io v0.21.0/go.mod h1:mSImk1erAIZhrmZN+AvHh14ztQfjbGwt4TtuofqLduU=
+go.opencensus.io v0.22.0 h1:C9hSCOW830chIVkdja34wa6Ky+IzWllkUinR+BtRZd4=
+go.opencensus.io v0.22.0/go.mod h1:+kGneAE2xo2IficOXnaByMWTGM9T73dGwxeWcUqIpI8=
+go.uber.org/atomic v1.3.2/go.mod h1:gD2HeocX3+yG+ygLZcrzQJaqmWj9AIm7n08wl/qW/PE=
+go.uber.org/atomic v1.4.0 h1:cxzIVoETapQEqDhQu3QfnvXAV4AlzcvUCxkVUFw3+EU=
+go.uber.org/atomic v1.4.0/go.mod h1:gD2HeocX3+yG+ygLZcrzQJaqmWj9AIm7n08wl/qW/PE=
+go.uber.org/multierr v1.1.0 h1:HoEmRHQPVSqub6w2z2d2EOVs2fjyFRGyofhKuyDq0QI=
+go.uber.org/multierr v1.1.0/go.mod h1:wR5kodmAFQ0UK8QlbwjlSNy0Z68gJhDJUG5sjR94q/0=
+go.uber.org/multierr v1.2.0 h1:6I+W7f5VwC5SV9dNrZ3qXrDB9mD0dyGOi/ZJmYw03T4=
+go.uber.org/multierr v1.2.0/go.mod h1:wR5kodmAFQ0UK8QlbwjlSNy0Z68gJhDJUG5sjR94q/0=
+go.uber.org/zap v1.9.1/go.mod h1:vwi/ZaCAaUcBkycHslxD9B2zi4UTXhF60s6SWpuDF0Q=
+go.uber.org/zap v1.10.0 h1:ORx85nbTijNz8ljznvCMR1ZBIPKFn3jQrag10X2AsuM=
+go.uber.org/zap v1.10.0/go.mod h1:vwi/ZaCAaUcBkycHslxD9B2zi4UTXhF60s6SWpuDF0Q=
+gocloud.dev v0.16.0 h1:hWeaQWxamGerwsU7B9xSWvUjx0p7TwG8fcHro2TzbbM=
+gocloud.dev v0.16.0/go.mod h1:xWGXD8t7bEhqPIuyAUFyXV9qHn+PvEY2F2GKPh7i/O0=
+gocloud.dev/pubsub/natspubsub v0.16.0 h1:MoBGXULDzb1fVaZsGWO5cUCgr6yoI/DHhau8OPGaGEI=
+gocloud.dev/pubsub/natspubsub v0.16.0/go.mod h1:0n7pT7PkLMClBUHDrOkHfOFVr/o/6kawNMwsyAbwadI=
+gocloud.dev/pubsub/rabbitpubsub v0.16.0 h1:Bkv2njMSl2tmT3tGbvbwpiIDAXBIpqzP9dmts+rhD4E=
+gocloud.dev/pubsub/rabbitpubsub v0.16.0/go.mod h1:JJVdUUIqwgaaMJg/1xHQza0g4sI/4KHHSNiGE+pn4JM=
+golang.org/x/crypto v0.0.0-20180608092829-8ac0e0d97ce4/go.mod h1:6SG95UA2DQfeDnfUPMdvaQW0Q7yPrPDi9nlGo2tz2b4=
+golang.org/x/crypto v0.0.0-20180904163835-0709b304e793/go.mod h1:6SG95UA2DQfeDnfUPMdvaQW0Q7yPrPDi9nlGo2tz2b4=
+golang.org/x/crypto v0.0.0-20181203042331-505ab145d0a9/go.mod h1:6SG95UA2DQfeDnfUPMdvaQW0Q7yPrPDi9nlGo2tz2b4=
+golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
+golang.org/x/crypto v0.0.0-20190404164418-38d8ce5564a5/go.mod h1:WFFai1msRO1wXaEeE5yQxYXgSfI8pQAWXbQop6sCtWE=
+golang.org/x/crypto v0.0.0-20190530122614-20be4c3c3ed5/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
+golang.org/x/crypto v0.0.0-20190605123033-f99c8df09eb5 h1:58fnuSXlxZmFdJyvtTFVmVhcMLU6v5fEb/ok4wyqtNU=
+golang.org/x/crypto v0.0.0-20190605123033-f99c8df09eb5/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
+golang.org/x/crypto v0.0.0-20190701094942-4def268fd1a4/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
+golang.org/x/crypto v0.0.0-20190909091759-094676da4a83/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
+golang.org/x/crypto v0.0.0-20190911031432-227b76d455e7 h1:0hQKqeLdqlt5iIwVOBErRisrHJAN57yOiPRQItI20fU=
+golang.org/x/crypto v0.0.0-20190911031432-227b76d455e7/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
+golang.org/x/exp v0.0.0-20190121172915-509febef88a4/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
+golang.org/x/exp v0.0.0-20190510132918-efd6b22b2522/go.mod h1:ZjyILWgesfNpC6sMxTJOJm9Kp84zZh5NQWvqDGG3Qr8=
+golang.org/x/image v0.0.0-20180708004352-c73c2afc3b81/go.mod h1:ux5Hcp/YLpHSI86hEcLt0YII63i6oz57MZXIpbrjZUs=
+golang.org/x/image v0.0.0-20190227222117-0694c2d4d067 h1:KYGJGHOQy8oSi1fDlSpcZF0+juKwk/hEMv5SiwHogR0=
+golang.org/x/image v0.0.0-20190227222117-0694c2d4d067/go.mod h1:kZ7UVZpmo3dzQBMxlp+ypCbDeSB+sBbTgSJuh5dn5js=
+golang.org/x/image v0.0.0-20190829233526-b3c06291d021 h1:j6QOxNFMpEL1wIQX6TUdBPNfGZKmBOJS/vfSm8a7tdM=
+golang.org/x/image v0.0.0-20190829233526-b3c06291d021/go.mod h1:FeLwcggjj3mMvU+oOTbSwawSJRM1uh48EjtB4UJZlP0=
+golang.org/x/lint v0.0.0-20181026193005-c67002cb31c3/go.mod h1:UVdnD1Gm6xHRNCYTkRU2/jEulfH38KcIWyp/GAMgvoE=
+golang.org/x/lint v0.0.0-20190227174305-5b3e6a55c961/go.mod h1:wehouNa3lNwaWXcvxsM5YxQ5yQlVC4a0KAMCusXpPoU=
+golang.org/x/lint v0.0.0-20190301231843-5614ed5bae6f/go.mod h1:UVdnD1Gm6xHRNCYTkRU2/jEulfH38KcIWyp/GAMgvoE=
+golang.org/x/lint v0.0.0-20190313153728-d0100b6bd8b3/go.mod h1:6SW0HCj/g11FgYtHlgUYUwCkIfeOF89ocIRzGO/8vkc=
+golang.org/x/lint v0.0.0-20190409202823-959b441ac422/go.mod h1:6SW0HCj/g11FgYtHlgUYUwCkIfeOF89ocIRzGO/8vkc=
+golang.org/x/mobile v0.0.0-20190312151609-d3739f865fa6/go.mod h1:z+o9i4GpDbdi3rU15maQ/Ox0txvL9dWGYEHz965HBQE=
+golang.org/x/net v0.0.0-20180724234803-3673e40ba225/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20180826012351-8a410e7b638d/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20180906233101-161cd47e91fd/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20181005035420-146acd28ed58/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20181114220301-adae6a3d119a/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20181220203305-927f97764cc3/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20190108225652-1e06a53dbb7e/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20190213061140-3a22650c66bd/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
+golang.org/x/net v0.0.0-20190311183353-d8887717615a/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
+golang.org/x/net v0.0.0-20190404232315-eb5bcb51f2a3/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
+golang.org/x/net v0.0.0-20190501004415-9ce7a6920f09/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
+golang.org/x/net v0.0.0-20190503192946-f4e77d36d62c/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
+golang.org/x/net v0.0.0-20190522155817-f3200d17e092/go.mod h1:HSz+uSET+XFnRR8LxR5pz3Of3rY3CfYBVs4xY44aLks=
+golang.org/x/net v0.0.0-20190603091049-60506f45cf65/go.mod h1:HSz+uSET+XFnRR8LxR5pz3Of3rY3CfYBVs4xY44aLks=
+golang.org/x/net v0.0.0-20190613194153-d28f0bde5980/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
+golang.org/x/net v0.0.0-20190619014844-b5b0513f8c1b/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
+golang.org/x/net v0.0.0-20190620200207-3b0461eec859/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
+golang.org/x/net v0.0.0-20190909003024-a7b16738d86b h1:XfVGCX+0T4WOStkaOsJRllbsiImhB2jgVBGc9L0lPGc=
+golang.org/x/net v0.0.0-20190909003024-a7b16738d86b/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
+golang.org/x/oauth2 v0.0.0-20180821212333-d2e6202438be/go.mod h1:N/0e6XlmueqKjAGxoOufVs8QHGRruUQn6yWY3a++T0U=
+golang.org/x/oauth2 v0.0.0-20190226205417-e64efc72b421/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=
+golang.org/x/oauth2 v0.0.0-20190402181905-9f3314589c9a/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=
+golang.org/x/oauth2 v0.0.0-20190604053449-0f29369cfe45 h1:SVwTIAaPC2U/AvvLNZ2a7OVsmBpC8L5BlwK1whH3hm0=
+golang.org/x/oauth2 v0.0.0-20190604053449-0f29369cfe45/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=
+golang.org/x/sync v0.0.0-20180314180146-1d60e4601c6f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+golang.org/x/sync v0.0.0-20181108010431-42b317875d0f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+golang.org/x/sync v0.0.0-20181221193216-37e7f081c4d4/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+golang.org/x/sync v0.0.0-20190227155943-e225da77a7e6/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+golang.org/x/sync v0.0.0-20190423024810-112230192c58 h1:8gQV6CLnAEikrhgkHFbMAEhagSSnXWGV915qUMm9mrU=
+golang.org/x/sync v0.0.0-20190423024810-112230192c58/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+golang.org/x/sys v0.0.0-20180830151530-49385e6e1522/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20180905080454-ebe1bf3edb33/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20180909124046-d0be0721c37e/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20181107165924-66b7b1311ac8/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20181116152217-5ac8a444bdc5/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20190215142949-d0b11bdaac8a/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20190222072716-a9d3bda3a223/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
+golang.org/x/sys v0.0.0-20190312061237-fead79001313/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190403152447-81d4e9dc473e/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190412213103-97732733099d/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190422165155-953cdadca894/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190502145724-3ef323f4f1fd/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190507160741-ecd444e8653b/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190606165138-5da285871e9c/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190620070143-6f217b454f45/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190624142023-c5567b49c5d0 h1:HyfiK1WMnHj5FXFXatD+Qs1A/xC2Run6RzeW1SyHxpc=
+golang.org/x/sys v0.0.0-20190624142023-c5567b49c5d0/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190726091711-fc99dfbffb4e/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190801041406-cbf593c0f2f3/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190909082730-f460065e899a/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20190910064555-bbd175535a8b h1:3S2h5FadpNr0zUUCVZjlKIEYF+KaX/OBplTGo89CYHI=
+golang.org/x/sys v0.0.0-20190910064555-bbd175535a8b/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
+golang.org/x/text v0.3.1-0.20180807135948-17ff2d5776d2/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
+golang.org/x/text v0.3.2 h1:tW2bmiBqwgJj/UpqtC8EpXEZVYOwU0yG4iWbprSVAcs=
+golang.org/x/text v0.3.2/go.mod h1:bEr9sfX3Q8Zfm5fL9x+3itogRgK3+ptLWKqgva+5dAk=
+golang.org/x/time v0.0.0-20180412165947-fbb02b2291d2/go.mod h1:tRJNPiyCQ0inRvYxbN9jk5I+vvW/OXSQhTDSoE431IQ=
+golang.org/x/time v0.0.0-20181108054448-85acf8d2951c/go.mod h1:tRJNPiyCQ0inRvYxbN9jk5I+vvW/OXSQhTDSoE431IQ=
+golang.org/x/time v0.0.0-20190308202827-9d24e82272b4 h1:SvFZT6jyqRaOeXpc5h/JSfZenJ2O330aBsf7JfSUXmQ=
+golang.org/x/time v0.0.0-20190308202827-9d24e82272b4/go.mod h1:tRJNPiyCQ0inRvYxbN9jk5I+vvW/OXSQhTDSoE431IQ=
+golang.org/x/tools v0.0.0-20180221164845-07fd8470d635/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
+golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
+golang.org/x/tools v0.0.0-20181030221726-6c7e314b6563/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
+golang.org/x/tools v0.0.0-20190114222345-bf090417da8b/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
+golang.org/x/tools v0.0.0-20190226205152-f727befe758c/go.mod h1:9Yl7xja0Znq3iFh3HoIrodX9oNMXvdceNzlUR8zjMvY=
+golang.org/x/tools v0.0.0-20190311212946-11955173bddd/go.mod h1:LCzVGOaR6xXOjkQ3onu1FJEFr0SW1gC7cKk1uF8kGRs=
+golang.org/x/tools v0.0.0-20190312151545-0bb0c0a6e846/go.mod h1:LCzVGOaR6xXOjkQ3onu1FJEFr0SW1gC7cKk1uF8kGRs=
+golang.org/x/tools v0.0.0-20190312170243-e65039ee4138/go.mod h1:LCzVGOaR6xXOjkQ3onu1FJEFr0SW1gC7cKk1uF8kGRs=
+golang.org/x/tools v0.0.0-20190422233926-fe54fb35175b/go.mod h1:LCzVGOaR6xXOjkQ3onu1FJEFr0SW1gC7cKk1uF8kGRs=
+golang.org/x/tools v0.0.0-20190425150028-36563e24a262/go.mod h1:RgjU9mgBXZiqYHBnxXauZ1Gv1EHHAz9KjViQ78xBX0Q=
+golang.org/x/tools v0.0.0-20190506145303-2d16b83fe98c/go.mod h1:RgjU9mgBXZiqYHBnxXauZ1Gv1EHHAz9KjViQ78xBX0Q=
+golang.org/x/tools v0.0.0-20190524140312-2c0ae7006135/go.mod h1:RgjU9mgBXZiqYHBnxXauZ1Gv1EHHAz9KjViQ78xBX0Q=
+golang.org/x/tools v0.0.0-20190606124116-d0a3d012864b/go.mod h1:/rFqwRUd4F7ZHNgwSSTFct+R/Kf4OFW1sUzUTQQTgfc=
+golang.org/x/tools v0.0.0-20190628153133-6cdbf07be9d0/go.mod h1:/rFqwRUd4F7ZHNgwSSTFct+R/Kf4OFW1sUzUTQQTgfc=
+golang.org/x/tools v0.0.0-20190911022129-16c5e0f7d110 h1:6S6bidS7O4yAwA5ORRbRIjvNQ9tGbLd5e+LRIaTeVDQ=
+golang.org/x/tools v0.0.0-20190911022129-16c5e0f7d110/go.mod h1:b+2E5dAYhXwXZwtnZ6UAqBI28+e2cm9otk0dWdXHAEo=
+golang.org/x/xerrors v0.0.0-20190513163551-3ee3066db522/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
+golang.org/x/xerrors v0.0.0-20190717185122-a985d3407aa7 h1:9zdDQZ7Thm29KFXgAX/+yaf3eVbP7djjWp/dXAppNCc=
+golang.org/x/xerrors v0.0.0-20190717185122-a985d3407aa7/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
+google.golang.org/api v0.4.0/go.mod h1:8k5glujaEP+g9n7WNsDg8QP6cUVNI86fCNMcbazEtwE=
+google.golang.org/api v0.5.0/go.mod h1:8k5glujaEP+g9n7WNsDg8QP6cUVNI86fCNMcbazEtwE=
+google.golang.org/api v0.6.0/go.mod h1:btoxGiFvQNVUZQ8W08zLtrVS08CNpINPEfxXxgJL1Q4=
+google.golang.org/api v0.7.0 h1:9sdfJOzWlkqPltHAuzT2Cp+yrBeY1KRVYgms8soxMwM=
+google.golang.org/api v0.7.0/go.mod h1:WtwebWUNSVBH/HAw79HIFXZNqEvBhG+Ra+ax0hx3E3M=
+google.golang.org/api v0.8.0/go.mod h1:o4eAsZoiT+ibD93RtjEohWalFOjRDx6CVaqeizhEnKg=
+google.golang.org/api v0.9.0 h1:jbyannxz0XFD3zdjgrSUsaJbgpH4eTrkdhRChkHPfO8=
+google.golang.org/api v0.9.0/go.mod h1:o4eAsZoiT+ibD93RtjEohWalFOjRDx6CVaqeizhEnKg=
+google.golang.org/appengine v1.1.0/go.mod h1:EbEs0AVv82hx2wNQdGPgUI5lhzA/G0D9YwlJXL52JkM=
+google.golang.org/appengine v1.4.0/go.mod h1:xpcJRLb0r/rnEns0DIKYYv+WjYCduHsrkT7/EB5XEv4=
+google.golang.org/appengine v1.5.0/go.mod h1:xpcJRLb0r/rnEns0DIKYYv+WjYCduHsrkT7/EB5XEv4=
+google.golang.org/appengine v1.6.1 h1:QzqyMA1tlu6CgqCDUtU9V+ZKhLFT2dkJuANu5QaxI3I=
+google.golang.org/appengine v1.6.1/go.mod h1:i06prIuMbXzDqacNJfV5OdTW448YApPu5ww/cMBSeb0=
+google.golang.org/appengine v1.6.2 h1:j8RI1yW0SkI+paT6uGwMlrMI/6zwYA6/CFil8rxOzGI=
+google.golang.org/appengine v1.6.2/go.mod h1:i06prIuMbXzDqacNJfV5OdTW448YApPu5ww/cMBSeb0=
+google.golang.org/genproto v0.0.0-20180608181217-32ee49c4dd80/go.mod h1:JiN7NxoALGmiZfu7CAH4rXhgtRTLTxftemlI0sWmxmc=
+google.golang.org/genproto v0.0.0-20180817151627-c66870c02cf8/go.mod h1:JiN7NxoALGmiZfu7CAH4rXhgtRTLTxftemlI0sWmxmc=
+google.golang.org/genproto v0.0.0-20181004005441-af9cb2a35e7f/go.mod h1:JiN7NxoALGmiZfu7CAH4rXhgtRTLTxftemlI0sWmxmc=
+google.golang.org/genproto v0.0.0-20190307195333-5fe7a883aa19/go.mod h1:VzzqZJRnGkLBvHegQrXjBqPurQTc5/KpmUdxsrq26oE=
+google.golang.org/genproto v0.0.0-20190418145605-e7d98fc518a7/go.mod h1:VzzqZJRnGkLBvHegQrXjBqPurQTc5/KpmUdxsrq26oE=
+google.golang.org/genproto v0.0.0-20190425155659-357c62f0e4bb/go.mod h1:VzzqZJRnGkLBvHegQrXjBqPurQTc5/KpmUdxsrq26oE=
+google.golang.org/genproto v0.0.0-20190502173448-54afdca5d873/go.mod h1:VzzqZJRnGkLBvHegQrXjBqPurQTc5/KpmUdxsrq26oE=
+google.golang.org/genproto v0.0.0-20190508193815-b515fa19cec8/go.mod h1:VzzqZJRnGkLBvHegQrXjBqPurQTc5/KpmUdxsrq26oE=
+google.golang.org/genproto v0.0.0-20190530194941-fb225487d101/go.mod h1:z3L6/3dTEVtUr6QSP8miRzeRqwQOioJ9I66odjN4I7s=
+google.golang.org/genproto v0.0.0-20190620144150-6af8c5fc6601/go.mod h1:z3L6/3dTEVtUr6QSP8miRzeRqwQOioJ9I66odjN4I7s=
+google.golang.org/genproto v0.0.0-20190801165951-fa694d86fc64/go.mod h1:DMBHOl98Agz4BDEuKkezgsaosCRResVns1a3J2ZsMNc=
+google.golang.org/genproto v0.0.0-20190905072037-92dd089d5514 h1:oFSK4421fpCKRrpzIpybyBVWyht05NegY9+L/3TLAZs=
+google.golang.org/genproto v0.0.0-20190905072037-92dd089d5514/go.mod h1:DMBHOl98Agz4BDEuKkezgsaosCRResVns1a3J2ZsMNc=
+google.golang.org/grpc v0.0.0-20180607172857-7a6a684ca69e/go.mod h1:yo6s7OP7yaDglbqo1J04qKzAhqBH6lvTonzMVmEdcZw=
+google.golang.org/grpc v1.14.0/go.mod h1:yo6s7OP7yaDglbqo1J04qKzAhqBH6lvTonzMVmEdcZw=
+google.golang.org/grpc v1.19.0/go.mod h1:mqu4LbDTu4XGKhr4mRzUsmM4RtVoemTSY81AxZiDr8c=
+google.golang.org/grpc v1.20.1/go.mod h1:10oTOabMzJvdu6/UiuZezV6QK5dSlG84ov/aaiqXj38=
+google.golang.org/grpc v1.21.0/go.mod h1:oYelfM1adQP15Ek0mdvEgi9Df8B9CZIaU1084ijfRaM=
+google.golang.org/grpc v1.21.1/go.mod h1:oYelfM1adQP15Ek0mdvEgi9Df8B9CZIaU1084ijfRaM=
+google.golang.org/grpc v1.23.0 h1:AzbTB6ux+okLTzP8Ru1Xs41C303zdcfEht7MQnYJt5A=
+google.golang.org/grpc v1.23.0/go.mod h1:Y5yQAOtifL1yxbo5wqy6BxZv8vAUGQwXBOALyacEbxg=
+gopkg.in/airbrake/gobrake.v2 v2.0.9/go.mod h1:/h5ZAUhDkGaJfjzjKLSjv6zCL6O0LLBxU4K+aSYdM/U=
+gopkg.in/alecthomas/gometalinter.v2 v2.0.12/go.mod h1:NDRytsqEZyolNuAgTzJkZMkSQM7FIKyzVzGhjB/qfYo=
+gopkg.in/alecthomas/kingpin.v2 v2.2.6/go.mod h1:FMv+mEhP44yOT+4EoQTLFTRgOQ1FBLkstjWtayDeSgw=
+gopkg.in/alecthomas/kingpin.v3-unstable v3.0.0-20180810215634-df19058c872c/go.mod h1:3HH7i1SgMqlzxCcBmUHW657sD4Kvv9sC3HpL3YukzwA=
+gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/check.v1 v1.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/check.v1 v1.0.0-20180628173108-788fd7840127 h1:qIbj1fsPNlZgppZ+VLlY7N33q108Sa+fhmuc+sWQYwY=
+gopkg.in/check.v1 v1.0.0-20180628173108-788fd7840127/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/check.v1 v1.0.0-20190902080502-41f04d3bba15 h1:YR8cESwS4TdDjEe65xsg0ogRM/Nc3DYOhEAlW+xobZo=
+gopkg.in/check.v1 v1.0.0-20190902080502-41f04d3bba15/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/cheggaaa/pb.v1 v1.0.25/go.mod h1:V/YB90LKu/1FcN3WVnfiiE5oMCibMjukxqG/qStrOgw=
+gopkg.in/fsnotify.v1 v1.4.7 h1:xOHLXZwVvI9hhs+cLKq5+I5onOuwQLhQwiu63xxlHs4=
+gopkg.in/fsnotify.v1 v1.4.7/go.mod h1:Tz8NjZHkW78fSQdbUxIjBTcgA1z1m8ZHf0WmKUhAMys=
+gopkg.in/gemnasium/logrus-airbrake-hook.v2 v2.1.2/go.mod h1:Xk6kEKp8OKb+X14hQBKWaSkCsqBpgog8nAV2xsGOxlo=
+gopkg.in/inf.v0 v0.9.1 h1:73M5CoZyi3ZLMOyDlQh031Cx6N9NDJ2Vvfl76EDAgDc=
+gopkg.in/inf.v0 v0.9.1/go.mod h1:cWUDdTG/fYaXco+Dcufb5Vnc6Gp2YChqWtbxRZE0mXw=
+gopkg.in/jcmturner/aescts.v1 v1.0.1 h1:cVVZBK2b1zY26haWB4vbBiZrfFQnfbTVrE3xZq6hrEw=
+gopkg.in/jcmturner/aescts.v1 v1.0.1/go.mod h1:nsR8qBOg+OucoIW+WMhB3GspUQXq9XorLnQb9XtvcOo=
+gopkg.in/jcmturner/dnsutils.v1 v1.0.1 h1:cIuC1OLRGZrld+16ZJvvZxVJeKPsvd5eUIvxfoN5hSM=
+gopkg.in/jcmturner/dnsutils.v1 v1.0.1/go.mod h1:m3v+5svpVOhtFAP/wSz+yzh4Mc0Fg7eRhxkJMWSIz9Q=
+gopkg.in/jcmturner/goidentity.v3 v3.0.0 h1:1duIyWiTaYvVx3YX2CYtpJbUFd7/UuPYCfgXtQ3VTbI=
+gopkg.in/jcmturner/goidentity.v3 v3.0.0/go.mod h1:oG2kH0IvSYNIu80dVAyu/yoefjq1mNfM5bm88whjWx4=
+gopkg.in/jcmturner/gokrb5.v7 v7.2.3 h1:hHMV/yKPwMnJhPuPx7pH2Uw/3Qyf+thJYlisUc44010=
+gopkg.in/jcmturner/gokrb5.v7 v7.2.3/go.mod h1:l8VISx+WGYp+Fp7KRbsiUuXTTOnxIc3Tuvyavf11/WM=
+gopkg.in/jcmturner/gokrb5.v7 v7.3.0 h1:0709Jtq/6QXEuWRfAm260XqlpcwL1vxtO1tUE2qK8Z4=
+gopkg.in/jcmturner/gokrb5.v7 v7.3.0/go.mod h1:l8VISx+WGYp+Fp7KRbsiUuXTTOnxIc3Tuvyavf11/WM=
+gopkg.in/jcmturner/rpc.v1 v1.1.0 h1:QHIUxTX1ISuAv9dD2wJ9HWQVuWDX/Zc0PfeC2tjc4rU=
+gopkg.in/jcmturner/rpc.v1 v1.1.0/go.mod h1:YIdkC4XfD6GXbzje11McwsDuOlZQSb9W4vfLvuNnlv8=
+gopkg.in/karlseguin/expect.v1 v1.0.1 h1:9u0iUltnhFbJTHaSIH0EP+cuTU5rafIgmcsEsg2JQFw=
+gopkg.in/karlseguin/expect.v1 v1.0.1/go.mod h1:uB7QIJBcclvYbwlUDkSCsGjAOMis3fP280LyhuDEf2I=
+gopkg.in/natefinch/lumberjack.v2 v2.0.0 h1:1Lc07Kr7qY4U2YPouBjpCLxpiyxIVoxqXgkXLknAOE8=
+gopkg.in/natefinch/lumberjack.v2 v2.0.0/go.mod h1:l0ndWWf7gzL7RNwBG7wST/UCcT4T24xpD6X8LsfU/+k=
+gopkg.in/resty.v1 v1.12.0/go.mod h1:mDo4pnntr5jdWRML875a/NmxYqAlA73dVijT2AXvQQo=
+gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7 h1:uRGJdciOHaEIrze2W8Q3AKkepLTh2hOroT7a+7czfdQ=
+gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7/go.mod h1:dt/ZhP58zS4L8KSrWDmTeBkI65Dw0HsyUHuEVlX15mw=
+gopkg.in/yaml.v2 v2.0.0-20170812160011-eb3733d160e7/go.mod h1:JAlM8MvJe8wmxCU4Bli9HhUf9+ttbYbLASfIpnQbh74=
+gopkg.in/yaml.v2 v2.2.1/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI=
+gopkg.in/yaml.v2 v2.2.2 h1:ZCJp+EgiOT7lHqUV2J862kp8Qj64Jo6az82+3Td9dZw=
+gopkg.in/yaml.v2 v2.2.2/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI=
+honnef.co/go/tools v0.0.0-20190102054323-c2f93a96b099/go.mod h1:rf3lG4BRIbNafJWhAfAdb/ePZxsR/4RtNHQocxwk9r4=
+honnef.co/go/tools v0.0.0-20190106161140-3f1c8253044a/go.mod h1:rf3lG4BRIbNafJWhAfAdb/ePZxsR/4RtNHQocxwk9r4=
+honnef.co/go/tools v0.0.0-20190418001031-e561f6794a2a/go.mod h1:rf3lG4BRIbNafJWhAfAdb/ePZxsR/4RtNHQocxwk9r4=
+honnef.co/go/tools v0.0.0-20190523083050-ea95bdfd59fc/go.mod h1:rf3lG4BRIbNafJWhAfAdb/ePZxsR/4RtNHQocxwk9r4=
+pack.ag/amqp v0.11.2/go.mod h1:4/cbmt4EJXSKlG6LCfWHoqmN0uFdy5i/+YFz+fTfhV4=
+rsc.io/binaryregexp v0.2.0/go.mod h1:qTv7/COck+e2FymRvadv62gMdZztPaShugOCi3I+8D8=
+sigs.k8s.io/yaml v1.1.0 h1:4A07+ZFc2wgJwo8YNlQpr1rVlgUDlxXHhPJciaPY5gs=
+sigs.k8s.io/yaml v1.1.0/go.mod h1:UJmg0vDUVViEyp3mgSv9WPwZCDxu4rQW1olrI1uml+o=
+sourcegraph.com/sourcegraph/appdash v0.0.0-20180531100431-4c381bd170b4 h1:VO9oZbbkvTwqLimlQt15QNdOOBArT2dw/bvzsMZBiqQ=
+sourcegraph.com/sourcegraph/appdash v0.0.0-20180531100431-4c381bd170b4/go.mod h1:hI742Nqp5OhwiqlzhgfbWU4mW4yO10fP+LoT9WOswdU=
+sourcegraph.com/sourcegraph/appdash-data v0.0.0-20151005221446-73f23eafcf67/go.mod h1:L5q+DGLGOQFpo1snNEkLOJT2d1YTW66rWNzatr3He1k=

note/memory_usage.txt

@@ -1,278 +0,0 @@
-64 32G volumes consumes 10G memory
-Each volume has 25M index, so each cost 160MB memory
-
-
-
-Things happened when I use lots of threads ( almost 120 ) keeping read file from SeaweedFS.
-But I'm not so familiar with linux so I can't tell you exactly what happened.
-Next I'll show you things I know , if you need more info , contact me
-
-My SeaweedFS version is about 0.12
-
-1. top
-
-top - 12:07:37 up 1 day,  3:17,  2 users,  load average: 0.00, 0.00, 0.00
-Tasks: 152 total,   1 running, 151 sleeping,   0 stopped,   0 zombie
-Cpu(s):  0.0%us,  0.0%sy,  0.0%ni, 99.8%id,  0.1%wa,  0.0%hi,  0.0%si,  0.0%st
-Mem:  16269880k total, 16192364k used,    77516k free,    58172k buffers
-Swap:  2064376k total,    12324k used,  2052052k free,  2827520k cached
-
-  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
- 1499 root      20   0 11.6g  10g 1424 S  0.0 65.6   7:32.53 weedvolume
- 1498 root      20   0 3204m 2.1g 1428 S  0.0 13.5   4:36.59 weedvolume
- 1737 root      20   0 98868 4932 2920 S  0.0  0.0   0:00.56 sshd
- 1497 root      20   0  151m 4404 1152 S  0.0  0.0   1:21.40 weedmaster
- 1335 root      20   0 97816 3044 2896 S  0.0  0.0   0:00.76 sshd
-
-After system became steady ,  weedvolume used 65.6% memory .
-
-2. free -m
-
-             total       used       free     shared    buffers     cached
-Mem:         15888      15809         79          0         56       2758
--/+ buffers/cache:      12994       2894
-Swap:         2015         12       2003
-
-3. startup cmd
-
-screen -d -m /opt/weed/weedmaster -mdir /data/weeddata/ > /data/logs/weed/master.log &
-screen -d -m /opt/weed/weedvolume -volumes=0-64 -dir /data/weeddata/ -pulseSeconds=20 -publicUrl="x.y.z:9334" -port 9334 > /data/logs/weed/s01.log &
-screen -d -m /opt/weed/weedvolume -volumes=65-107 -dir /data/weeddata/ -pulseSeconds=20 -publicUrl="x.y.z:9335" -port 9335 > /data/logs/weed/s02.log &
-
-4. du -sh .
-
-32G     0.dat
-26M     0.idx
-8.2G    100.dat
-6.8M    100.idx
-8.2G    101.dat
-6.9M    101.idx
-8.2G    102.dat
-6.8M    102.idx
-8.2G    103.dat
-6.8M    103.idx
-8.2G    104.dat
-6.8M    104.idx
-8.2G    105.dat
-6.9M    105.idx
-8.2G    106.dat
-6.9M    106.idx
-8.2G    107.dat
-6.9M    107.idx
-32G     10.dat
-25M     10.idx
-32G     11.dat
-25M     11.idx
-32G     12.dat
-25M     12.idx
-32G     13.dat
-25M     13.idx
-32G     14.dat
-25M     14.idx
-32G     15.dat
-25M     15.idx
-32G     16.dat
-25M     16.idx
-32G     17.dat
-25M     17.idx
-32G     18.dat
-25M     18.idx
-32G     19.dat
-25M     19.idx
-32G     1.dat
-26M     1.idx
-32G     20.dat
-25M     20.idx
-32G     21.dat
-25M     21.idx
-32G     22.dat
-25M     22.idx
-32G     23.dat
-25M     23.idx
-32G     24.dat
-25M     24.idx
-32G     25.dat
-25M     25.idx
-32G     26.dat
-25M     26.idx
-32G     27.dat
-25M     27.idx
-32G     28.dat
-25M     28.idx
-32G     29.dat
-25M     29.idx
-32G     2.dat
-26M     2.idx
-32G     30.dat
-25M     30.idx
-32G     31.dat
-25M     31.idx
-32G     32.dat
-25M     32.idx
-32G     33.dat
-25M     33.idx
-32G     34.dat
-25M     34.idx
-32G     35.dat
-25M     35.idx
-32G     36.dat
-25M     36.idx
-32G     37.dat
-25M     37.idx
-32G     38.dat
-25M     38.idx
-32G     39.dat
-25M     39.idx
-32G     3.dat
-26M     3.idx
-32G     40.dat
-25M     40.idx
-32G     41.dat
-25M     41.idx
-32G     42.dat
-25M     42.idx
-32G     43.dat
-25M     43.idx
-32G     44.dat
-25M     44.idx
-32G     45.dat
-25M     45.idx
-32G     46.dat
-25M     46.idx
-32G     47.dat
-25M     47.idx
-32G     48.dat
-25M     48.idx
-32G     49.dat
-25M     49.idx
-32G     4.dat
-26M     4.idx
-32G     50.dat
-25M     50.idx
-32G     51.dat
-25M     51.idx
-32G     52.dat
-25M     52.idx
-32G     53.dat
-25M     53.idx
-32G     54.dat
-25M     54.idx
-32G     55.dat
-25M     55.idx
-32G     56.dat
-25M     56.idx
-32G     57.dat
-25M     57.idx
-32G     58.dat
-25M     58.idx
-32G     59.dat
-25M     59.idx
-32G     5.dat
-26M     5.idx
-32G     60.dat
-25M     60.idx
-32G     61.dat
-25M     61.idx
-32G     62.dat
-25M     62.idx
-32G     63.dat
-25M     63.idx
-32G     64.dat
-25M     64.idx
-8.2G    65.dat
-6.9M    65.idx
-8.2G    66.dat
-6.9M    66.idx
-8.2G    67.dat
-6.9M    67.idx
-8.2G    68.dat
-6.8M    68.idx
-8.2G    69.dat
-6.9M    69.idx
-32G     6.dat
-25M     6.idx
-8.2G    70.dat
-6.8M    70.idx
-8.2G    71.dat
-6.9M    71.idx
-8.2G    72.dat
-6.9M    72.idx
-8.2G    73.dat
-6.9M    73.idx
-8.2G    74.dat
-6.9M    74.idx
-8.2G    75.dat
-6.9M    75.idx
-8.1G    76.dat
-6.8M    76.idx
-8.2G    77.dat
-6.8M    77.idx
-8.2G    78.dat
-6.8M    78.idx
-8.1G    79.dat
-6.8M    79.idx
-32G     7.dat
-25M     7.idx
-8.2G    80.dat
-6.8M    80.idx
-8.2G    81.dat
-6.9M    81.idx
-8.2G    82.dat
-6.9M    82.idx
-8.2G    83.dat
-6.9M    83.idx
-8.2G    84.dat
-6.9M    84.idx
-8.2G    85.dat
-6.8M    85.idx
-8.2G    86.dat
-6.9M    86.idx
-8.2G    87.dat
-6.9M    87.idx
-8.2G    88.dat
-6.9M    88.idx
-8.2G    89.dat
-6.8M    89.idx
-32G     8.dat
-25M     8.idx
-8.2G    90.dat
-6.9M    90.idx
-8.1G    91.dat
-6.8M    91.idx
-8.1G    92.dat
-6.8M    92.idx
-8.1G    93.dat
-6.8M    93.idx
-8.2G    94.dat
-6.9M    94.idx
-8.2G    95.dat
-6.9M    95.idx
-8.2G    96.dat
-6.9M    96.idx
-8.2G    97.dat
-6.9M    97.idx
-8.2G    98.dat
-6.9M    98.idx
-8.2G    99.dat
-6.9M    99.idx
-32G     9.dat
-25M     9.idx
-4.0K    directory.seq
-
-You can see the volume 1-64 is now full.
-
-5. more log
-
-see logs.zip
-
-In messages you can see these lines: (Line 51095)
-
-Sep 26 06:14:31 wedb-01 kernel: auditd: page allocation failure. order:0, mode:0x20
-Sep 26 06:14:31 wedb-01 kernel: Pid: 1009, comm: auditd Not tainted 2.6.32-220.el6.x86_64 #1
-Sep 26 06:14:31 wedb-01 kernel: Call Trace:
-
-After those lines , the system deny any new network connect request
-
-6. /dir/status
-
-{"Machines":[{"Server":{"Url":"127.0.0.1:9335","PublicUrl":"x.y.z:9335"},"Volumes":[{"Id":106,"Size":8728909632},{"Id":66,"Size":8729852744},{"Id":90,"Size":8747834896},{"Id":103,"Size":8718106024},{"Id":87,"Size":8732133512},{"Id":96,"Size":8737251904},{"Id":80,"Size":8704130712},{"Id":77,"Size":8717989496},{"Id":70,"Size":8731474744},{"Id":94,"Size":8758656144},{"Id":107,"Size":8729599232},{"Id":67,"Size":8736848088},{"Id":91,"Size":8665847760},{"Id":100,"Size":8703272552},{"Id":84,"Size":8745121528},{"Id":97,"Size":8713031744},{"Id":81,"Size":8726088872},{"Id":74,"Size":8738588152},{"Id":71,"Size":8729349920},{"Id":95,"Size":8741526896},{"Id":104,"Size":8699374736},{"Id":88,"Size":8740362880},{"Id":101,"Size":8711832992},{"Id":85,"Size":8723479552},{"Id":78,"Size":8700345400},{"Id":75,"Size":8727796912},{"Id":68,"Size":8698607440},{"Id":92,"Size":8682683056},{"Id":105,"Size":8741226152},{"Id":65,"Size":8725365752},{"Id":89,"Size":8703062600},{"Id":98,"Size":8742331560},{"Id":82,"Size":8762554952},{"Id":79,"Size":8696300376},{"Id":72,"Size":8708217304},{"Id":69,"Size":8740268144},{"Id":93,"Size":8685060320},{"Id":102,"Size":8708695352},{"Id":86,"Size":8783247776},{"Id":99,"Size":8753463608},{"Id":83,"Size":8725963952},{"Id":76,"Size":8694693536},{"Id":73,"Size":8733560832}]},{"Server":{"Url":"127.0.0.1:9334","PublicUrl":"x.y.z:9334"},"Volumes":[{"Id":34,"Size":33415706800},{"Id":58,"Size":33569224784},{"Id":18,"Size":33474649968},{"Id":55,"Size":33542422680},{"Id":15,"Size":33517247576},{"Id":48,"Size":33574860328},{"Id":8,"Size":33511257144},{"Id":45,"Size":33463948408},{"Id":5,"Size":34317702920},{"Id":29,"Size":33465695776},{"Id":38,"Size":33553119624},{"Id":62,"Size":33448316736},{"Id":22,"Size":33566586296},{"Id":35,"Size":33493733728},{"Id":59,"Size":33498554904},{"Id":19,"Size":33493313784},{"Id":52,"Size":33552978448},{"Id":12,"Size":33505183752},{"Id":49,"Size":33603029896},{"Id":9,"Size":33515778064},{"Id":42,"Size":33500402248},{"Id":2,"Size":34223232992},{"Id":26,"Size":33526519600},{"Id":39,"Size":33580414336},{"Id":63,"Size":33476332456},{"Id":23,"Size":33543872592},{"Id":32,"Size":33515290168},{"Id":56,"Size":33499171184},{"Id":16,"Size":33556591168},{"Id":64,"Size":33495148616},{"Id":53,"Size":33467738560},{"Id":13,"Size":33596873960},{"Id":46,"Size":33508120448},{"Id":6,"Size":33417470256},{"Id":30,"Size":33532933992},{"Id":43,"Size":33591802008},{"Id":3,"Size":34270682080},{"Id":27,"Size":33525736944},{"Id":36,"Size":33443597824},{"Id":60,"Size":33427931336},{"Id":20,"Size":33499083096},{"Id":33,"Size":33531396280},{"Id":57,"Size":33578015104},{"Id":17,"Size":33510525480},{"Id":50,"Size":33503123704},{"Id":10,"Size":33502391608},{"Id":47,"Size":33521868568},{"Id":7,"Size":33497101664},{"Id":31,"Size":33426905232},{"Id":40,"Size":33472978696},{"Id":0,"Size":34337344304},{"Id":24,"Size":33550157192},{"Id":37,"Size":33477162720},{"Id":61,"Size":33537175080},{"Id":21,"Size":33517192456},{"Id":54,"Size":33480720288},{"Id":14,"Size":33513192896},{"Id":51,"Size":33531336080},{"Id":11,"Size":33562385088},{"Id":44,"Size":33554479104},{"Id":4,"Size":34333127520},{"Id":28,"Size":33510503000},{"Id":41,"Size":33574922928},{"Id":1,"Size":34307181368},{"Id":25,"Size":33542834568}]}],"Writers":[106,66,90,103,87,96,80,77,70,94,107,67,91,100,84,97,81,74,71,95,104,88,101,85,78,75,68,92,105,65,89,98,82,79,72,69,93,102,86,99,83,76,73,34,58,18,55,15,48,8,45,5,29,38,62,22,35,59,19,52,12,49,9,42,2,26,39,63,23,32,56,16,64,53,13,46,6,30,43,3,27,36,60,20,33,57,17,50,10,47,7,31,40,0,24,37,61,21,54,14,51,11,44,4,28,41,1,25],"FileIdSequence":110250000}

note/replication.txt

@@ -1,109 +0,0 @@
-1. each file can choose the replication factor
-2. replication granularity is in volume level
-3. if not enough spaces, we can automatically decrease some volume's the replication factor, especially for cold data
-4. plan to support migrating data to cheaper storage
-5. plan to manual volume placement, access-based volume placement, auction based volume placement
-
-When a new volume server is started, it reports 
-  1. how many volumes it can hold
-  2. current list of existing volumes and each volume's replication type
-Each volume server remembers:
-  1. current volume ids
-  2. replica locations are read from the master
-
-The master assign volume ids based on
-  1. replication factor
-     data center, rack
-  2. concurrent write support
-On master, stores the replication configuration
-{
-  replication:{
-    {type:"00", min_volume_count:3, weight:10},
-    {type:"01", min_volume_count:2, weight:20},
-    {type:"10", min_volume_count:2, weight:20},
-    {type:"11", min_volume_count:3, weight:30},
-    {type:"20", min_volume_count:2, weight:20}
-  },
-  port:9333,
-}
-Or manually via command line
-  1. add volume with specified replication factor
-  2. add volume with specified volume id
-
-
-If duplicated volume ids are reported from different volume servers, 
-the master determines the replication factor of the volume,
-if less than the replication factor, the volume is in readonly mode
-if more than the replication factor, the volume will purge the smallest/oldest volume
-if equal, the volume will function as usual
-
-
-Use cases:
-  on volume server
-    1. weed volume -mserver="xx.xx.xx.xx:9333" -publicUrl="good.com:8080" -dir="/tmp" -volumes=50
-  on weed master
-    1. weed master -port=9333
-      generate a default json configuration file if doesn't exist
-      
-Bootstrap
-  1. at the very beginning, the system has no volumes at all.
-When data node starts:
-  1. each data node send to master its existing volumes and max volume blocks
-  2. master remembers the topology/data_center/rack/data_node/volumes
-     for each replication level, stores
-       volume id ~ data node
-       writable volume ids
-If any "assign" request comes in
-  1. find a writable volume with the right replicationLevel
-  2. if not found, grow the volumes with the right replication level
-  3. return a writable volume to the user
-
-  
-  for data node:
-    0. detect existing volumes DONE
-    1. onStartUp, and periodically, send existing volumes and maxVolumeCount  store.Join(), DONE
-    2. accept command to grow a volume( id + replication level)  DONE
-       /admin/assign_volume?volume=some_id&replicationType=01
-    3. accept setting volumeLocationList  DONE
-       /admin/set_volume_locations_list?volumeLocationsList=[{Vid:xxx,Locations:[loc1,loc2,loc3]}]
-    4. for each write, pass the write to the next location, (Step 2)
-       POST method should accept an index, like ttl, get decremented every hop
-  for master:
-    1. accept data node's report of existing volumes and maxVolumeCount ALREADY EXISTS /dir/join
-    2. periodically refresh for active data nodes, and adjust writable volumes
-    3. send command to grow a volume(id + replication level)  DONE
-    5. accept lookup for volume locations    ALREADY EXISTS /dir/lookup
-    6. read topology/datacenter/rack layout
-
-An algorithm to allocate volumes evenly, but may be inefficient if free volumes are plenty:
-input: replication=xyz
-algorithm:
-ret_dcs = []
-foreach dc that has y+z+1 volumes{
-  ret_racks = []
-  foreach rack with z+1 volumes{
-    ret = select z+1 servers with 1 volume
-  if ret.size()==z+1 {
-    ret_racks.append(ret)
-  }
-  }
-  randomly pick one rack from ret_racks
-  ret += select y racks with 1 volume each
-  if ret.size()==y+z+1{
-    ret_dcs.append(ret)
-  }
-}
-randomly pick one dc from ret_dcs
-ret += select x data centers with 1 volume each
-
-A simple replica placement algorithm, but may fail when free volume slots are not plenty:
-ret := []volumes
-dc = randomly pick 1 data center with y+z+1 volumes
-  rack = randomly pick 1 rack with z+1 volumes
-    ret = ret.append(randomly pick z+1 volumes)
-  ret = ret.append(randomly pick y racks with 1 volume)
-ret = ret.append(randomly pick x data centers with 1 volume)
-
-
-TODO:
-  1. replicate content to the other server if the replication type needs replicas

note/security.txt

@@ -1,51 +0,0 @@
-Design for SeaweedFS security
-
-Design Objectives
-	Security can mean many different things. The original vision is that: if you have one machine lying around
-	somewhere with some disk space, it should be able to join your file system to contribute some disk space and
-	network bandwidth, securely!
-
-	To achieve this purpose, the security should be able to:
-	1. Secure the inter-server communication. Only real cluster servers can join and communicate.
-	2. allow clients to securely write to volume servers
-
-Non Objective
-	Multi-tenant support. Avoid filers or clients cross-updating files.
-	User specific access control.
-
-Design Architect
-	master, and volume servers all talk securely via 2-way SSL for admin operations.
-	upon joining, master gives its secret key to volume servers.
-	filer or clients talk to master to get secret key, and use the key to generate JWT to write on volume server.
-	A side benefit:
-		a time limited read feature?
-	4. volume server needs to expose https ports
-
-HTTP Connections
-	clear http
-		filer~>master, need to get a JWT from master
-		filer~>volume
-	2-way https
-		master~ssl~>volume
-		volume~ssl~>master
-
-file uploading:
-	when volume server starts, it asks master for the secret key to decode JWT
-	when filer/clients wants to upload, master generate a JWT
-		filer~>volume(public port)
-		master~>volume(public port)
-
-Currently, volume server has 2 ip addresses: ip and publicUrl.
-	The ip is for admin purpose, and master talk to volume server this way.
-	The publicUrl is for clients to access the server, via http GET/POST/DELETE etc.
-		The write operations are secured by JWT.
-	clients talk to master also via https? possible. Decide on this later.
-
-Dev plan:
-	1. volume server separate admin from public GET/POST/DELETE handlers
-		The step 1 may be good enough for most use cases.
-
-	If 2-way ssl are still needed
-	2. volume server add ssl support
-	3. https connections to operate on volume servers
-

note/weedfs.txt

@@ -1,46 +0,0 @@
-How to submit a content
-1. Find physical volumes
-1.c Create a hash value
-1.d find a write logic volume id, and return [logic volume id, {physical volume ids}]
-2. submit to physical volumes
-2.c
-  generate the cookie
-  generate a unique id as key
-  choose the right altKey
-  send bytes to physical volumes
-2.s each
-  save bytes
-  store map[key uint64, altKey uint32]<offset, size>
-    for updated entry, set old entry's offset to zero
-3.c
-  wait for all physical volumes to finish
-  store the /<logic volume id>/<key>_<cookie>_<altKey>.<ext>
-
-How to retrieve a content
-1.c
-  send logic volume id
-1.d
-  find least busy volume's id
-2.c
-  send URI /<physical volume id>/<key>_<cookie>_<altKey>.<ext>
-
-
-How to submit a content
-1. send bytes to SeaweedFS, got <volume id, key uint64, cookie code>
-   store <key uint64, volume id uint32, cookie code uint32, ext>, and other information
-
-To read a content
-2. use logic volume id to lookup a <machine id>
-   render url as /<machine id>/<volume id>/<key>/<cookie>.ext
-
-The directory server
-0.init
-  load and collect <logic volume id, machine ids> mapping
-1.on submit content
-  find a free logic volume id, start sending content to 3 machines
-  if all of them finishes, return <logic volume id, key, cookie code>
-2.on read content
-  based on logic volume id, pick a machine with less load,
-    return <machine id>
-
-

other/java/client/pom.xml

@@ -0,0 +1,167 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
+    <modelVersion>4.0.0</modelVersion>
+
+    <groupId>com.github.chrislusf</groupId>
+    <artifactId>seaweedfs-client</artifactId>
+    <version>1.2.4</version>
+
+    <parent>
+        <groupId>org.sonatype.oss</groupId>
+        <artifactId>oss-parent</artifactId>
+        <version>9</version>
+    </parent>
+
+    <properties>
+        <protobuf.version>3.9.1</protobuf.version>
+        <!-- follow https://github.com/grpc/grpc-java -->
+        <grpc.version>1.23.0</grpc.version>
+        <guava.version>28.0-jre</guava.version>
+    </properties>
+
+    <dependencies>
+        <dependency>
+            <groupId>com.moandjiezana.toml</groupId>
+            <artifactId>toml4j</artifactId>
+            <version>0.7.2</version>
+        </dependency>
+        <!-- https://mvnrepository.com/artifact/com.google.protobuf/protobuf-java -->
+        <dependency>
+            <groupId>com.google.protobuf</groupId>
+            <artifactId>protobuf-java</artifactId>
+            <version>${protobuf.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>com.google.guava</groupId>
+            <artifactId>guava</artifactId>
+            <version>${guava.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>io.grpc</groupId>
+            <artifactId>grpc-netty-shaded</artifactId>
+            <version>${grpc.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>io.grpc</groupId>
+            <artifactId>grpc-protobuf</artifactId>
+            <version>${grpc.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>io.grpc</groupId>
+            <artifactId>grpc-stub</artifactId>
+            <version>${grpc.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>org.slf4j</groupId>
+            <artifactId>slf4j-api</artifactId>
+            <version>1.7.25</version>
+        </dependency>
+        <dependency>
+            <groupId>org.apache.httpcomponents</groupId>
+            <artifactId>httpmime</artifactId>
+            <version>4.5.6</version>
+        </dependency>
+        <dependency>
+            <groupId>junit</groupId>
+            <artifactId>junit</artifactId>
+            <version>4.12</version>
+            <scope>test</scope>
+        </dependency>
+    </dependencies>
+
+    <distributionManagement>
+        <snapshotRepository>
+            <id>ossrh</id>
+            <url>https://oss.sonatype.org/content/repositories/snapshots</url>
+        </snapshotRepository>
+    </distributionManagement>
+    <build>
+        <extensions>
+            <extension>
+                <groupId>kr.motd.maven</groupId>
+                <artifactId>os-maven-plugin</artifactId>
+                <version>1.6.2</version>
+            </extension>
+        </extensions>
+        <plugins>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-compiler-plugin</artifactId>
+                <configuration>
+                    <source>7</source>
+                    <target>7</target>
+                </configuration>
+            </plugin>
+            <plugin>
+                <groupId>org.xolstice.maven.plugins</groupId>
+                <artifactId>protobuf-maven-plugin</artifactId>
+                <version>0.6.1</version>
+                <configuration>
+                    <protocArtifact>com.google.protobuf:protoc:${protobuf.version}:exe:${os.detected.classifier}</protocArtifact>
+                    <pluginId>grpc-java</pluginId>
+                    <pluginArtifact>io.grpc:protoc-gen-grpc-java:${grpc.version}:exe:${os.detected.classifier}</pluginArtifact>
+                </configuration>
+                <executions>
+                    <execution>
+                        <goals>
+                            <goal>compile</goal>
+                            <goal>compile-custom</goal>
+                        </goals>
+                    </execution>
+                </executions>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-gpg-plugin</artifactId>
+                <version>1.5</version>
+                <executions>
+                    <execution>
+                        <id>sign-artifacts</id>
+                        <phase>verify</phase>
+                        <goals>
+                            <goal>sign</goal>
+                        </goals>
+                    </execution>
+                </executions>
+            </plugin>
+            <plugin>
+                <groupId>org.sonatype.plugins</groupId>
+                <artifactId>nexus-staging-maven-plugin</artifactId>
+                <version>1.6.7</version>
+                <extensions>true</extensions>
+                <configuration>
+                    <serverId>ossrh</serverId>
+                    <nexusUrl>https://oss.sonatype.org/</nexusUrl>
+                    <autoReleaseAfterClose>true</autoReleaseAfterClose>
+                </configuration>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-source-plugin</artifactId>
+                <version>2.2.1</version>
+                <executions>
+                    <execution>
+                        <id>attach-sources</id>
+                        <goals>
+                            <goal>jar-no-fork</goal>
+                        </goals>
+                    </execution>
+                </executions>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-javadoc-plugin</artifactId>
+                <version>2.9.1</version>
+                <executions>
+                    <execution>
+                        <id>attach-javadocs</id>
+                        <goals>
+                            <goal>jar</goal>
+                        </goals>
+                    </execution>
+                </executions>
+            </plugin>
+        </plugins>
+    </build>
+
+</project>

other/java/client/src/main/java/seaweedfs/client/ByteBufferOutputStream.java

@@ -0,0 +1,21 @@
+package seaweedfs.client;
+
+import java.io.IOException;
+import java.io.OutputStream;
+import java.nio.ByteBuffer;
+
+public class ByteBufferOutputStream extends OutputStream {
+    private final ByteBuffer buf;
+
+    public ByteBufferOutputStream(ByteBuffer buf) {
+        this.buf = buf;
+    }
+
+    public void write(int b) throws IOException {
+        this.buf.put((byte)b);
+    }
+
+    public void write(byte[] b, int off, int len) throws IOException {
+        this.buf.put(b, off, len);
+    }
+}

other/java/client/src/main/java/seaweedfs/client/FilerClient.java

@@ -0,0 +1,286 @@
+package seaweedfs.client;
+
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.nio.file.Path;
+import java.nio.file.Paths;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Iterator;
+import java.util.List;
+
+public class FilerClient {
+
+    private static final Logger LOG = LoggerFactory.getLogger(FilerClient.class);
+
+    private FilerGrpcClient filerGrpcClient;
+
+    public FilerClient(String host, int grpcPort) {
+        filerGrpcClient = new FilerGrpcClient(host, grpcPort);
+    }
+
+    public FilerClient(FilerGrpcClient filerGrpcClient) {
+        this.filerGrpcClient = filerGrpcClient;
+    }
+
+    public boolean mkdirs(String path, int mode) {
+        String currentUser = System.getProperty("user.name");
+        return mkdirs(path, mode, 0, 0, currentUser, new String[]{});
+    }
+
+    public boolean mkdirs(String path, int mode, String userName, String[] groupNames) {
+        return mkdirs(path, mode, 0, 0, userName, groupNames);
+    }
+
+    public boolean mkdirs(String path, int mode, int uid, int gid, String userName, String[] groupNames) {
+
+        if ("/".equals(path)) {
+            return true;
+        }
+        Path pathObject = Paths.get(path);
+        String parent = pathObject.getParent().toString();
+        String name = pathObject.getFileName().toString();
+
+        mkdirs(parent, mode, uid, gid, userName, groupNames);
+
+        FilerProto.Entry existingEntry = lookupEntry(parent, name);
+
+        if (existingEntry != null) {
+            return true;
+        }
+
+        return createEntry(
+                parent,
+                newDirectoryEntry(name, mode, uid, gid, userName, groupNames).build()
+        );
+
+    }
+
+    public boolean mv(String oldPath, String newPath) {
+
+        Path oldPathObject = Paths.get(oldPath);
+        String oldParent = oldPathObject.getParent().toString();
+        String oldName = oldPathObject.getFileName().toString();
+
+        Path newPathObject = Paths.get(newPath);
+        String newParent = newPathObject.getParent().toString();
+        String newName = newPathObject.getFileName().toString();
+
+        return atomicRenameEntry(oldParent, oldName, newParent, newName);
+
+    }
+
+    public boolean rm(String path, boolean isRecursive, boolean ignoreRecusiveError) {
+
+        Path pathObject = Paths.get(path);
+        String parent = pathObject.getParent().toString();
+        String name = pathObject.getFileName().toString();
+
+        return deleteEntry(
+                parent,
+                name,
+                true,
+                isRecursive,
+                ignoreRecusiveError);
+    }
+
+    public boolean touch(String path, int mode) {
+        String currentUser = System.getProperty("user.name");
+        return touch(path, mode, 0, 0, currentUser, new String[]{});
+    }
+
+    public boolean touch(String path, int mode, int uid, int gid, String userName, String[] groupNames) {
+
+        Path pathObject = Paths.get(path);
+        String parent = pathObject.getParent().toString();
+        String name = pathObject.getFileName().toString();
+
+        FilerProto.Entry entry = lookupEntry(parent, name);
+        if (entry == null) {
+            return createEntry(
+                    parent,
+                    newFileEntry(name, mode, uid, gid, userName, groupNames).build()
+            );
+        }
+        long now = System.currentTimeMillis() / 1000L;
+        FilerProto.FuseAttributes.Builder attr = entry.getAttributes().toBuilder()
+                .setMtime(now)
+                .setUid(uid)
+                .setGid(gid)
+                .setUserName(userName)
+                .clearGroupName()
+                .addAllGroupName(Arrays.asList(groupNames));
+        return updateEntry(parent, entry.toBuilder().setAttributes(attr).build());
+    }
+
+    public FilerProto.Entry.Builder newDirectoryEntry(String name, int mode,
+                                                      int uid, int gid, String userName, String[] groupNames) {
+
+        long now = System.currentTimeMillis() / 1000L;
+
+        return FilerProto.Entry.newBuilder()
+                .setName(name)
+                .setIsDirectory(true)
+                .setAttributes(FilerProto.FuseAttributes.newBuilder()
+                        .setMtime(now)
+                        .setCrtime(now)
+                        .setUid(uid)
+                        .setGid(gid)
+                        .setFileMode(mode | 1 << 31)
+                        .setUserName(userName)
+                        .clearGroupName()
+                        .addAllGroupName(Arrays.asList(groupNames)));
+    }
+
+    public FilerProto.Entry.Builder newFileEntry(String name, int mode,
+                                                 int uid, int gid, String userName, String[] groupNames) {
+
+        long now = System.currentTimeMillis() / 1000L;
+
+        return FilerProto.Entry.newBuilder()
+                .setName(name)
+                .setIsDirectory(false)
+                .setAttributes(FilerProto.FuseAttributes.newBuilder()
+                        .setMtime(now)
+                        .setCrtime(now)
+                        .setUid(uid)
+                        .setGid(gid)
+                        .setFileMode(mode)
+                        .setUserName(userName)
+                        .clearGroupName()
+                        .addAllGroupName(Arrays.asList(groupNames)));
+    }
+
+    public List<FilerProto.Entry> listEntries(String path) {
+        List<FilerProto.Entry> results = new ArrayList<FilerProto.Entry>();
+        String lastFileName = "";
+        for (int limit = Integer.MAX_VALUE; limit > 0; ) {
+            List<FilerProto.Entry> t = listEntries(path, "", lastFileName, 1024);
+            if (t == null) {
+                break;
+            }
+            int nSize = t.size();
+            if (nSize > 0) {
+                limit -= nSize;
+                lastFileName = t.get(nSize - 1).getName();
+            }
+            results.addAll(t);
+            if (t.size() < 1024) {
+                break;
+            }
+        }
+        return results;
+    }
+
+    public List<FilerProto.Entry> listEntries(String path, String entryPrefix, String lastEntryName, int limit) {
+        Iterator<FilerProto.ListEntriesResponse> iter = filerGrpcClient.getBlockingStub().listEntries(FilerProto.ListEntriesRequest.newBuilder()
+                .setDirectory(path)
+                .setPrefix(entryPrefix)
+                .setStartFromFileName(lastEntryName)
+                .setLimit(limit)
+                .build());
+        List<FilerProto.Entry> entries = new ArrayList<>();
+        while (iter.hasNext()){
+            FilerProto.ListEntriesResponse resp = iter.next();
+            entries.add(fixEntryAfterReading(resp.getEntry()));
+        }
+        return entries;
+    }
+
+    public FilerProto.Entry lookupEntry(String directory, String entryName) {
+        try {
+            FilerProto.Entry entry = filerGrpcClient.getBlockingStub().lookupDirectoryEntry(
+                    FilerProto.LookupDirectoryEntryRequest.newBuilder()
+                            .setDirectory(directory)
+                            .setName(entryName)
+                            .build()).getEntry();
+            return fixEntryAfterReading(entry);
+        } catch (Exception e) {
+            if (e.getMessage().indexOf("filer: no entry is found in filer store")>0){
+                return null;
+            }
+            LOG.warn("lookupEntry {}/{}: {}", directory, entryName, e);
+            return null;
+        }
+    }
+
+
+    public boolean createEntry(String parent, FilerProto.Entry entry) {
+        try {
+            filerGrpcClient.getBlockingStub().createEntry(FilerProto.CreateEntryRequest.newBuilder()
+                    .setDirectory(parent)
+                    .setEntry(entry)
+                    .build());
+        } catch (Exception e) {
+            LOG.warn("createEntry {}/{}: {}", parent, entry.getName(), e);
+            return false;
+        }
+        return true;
+    }
+
+    public boolean updateEntry(String parent, FilerProto.Entry entry) {
+        try {
+            filerGrpcClient.getBlockingStub().updateEntry(FilerProto.UpdateEntryRequest.newBuilder()
+                    .setDirectory(parent)
+                    .setEntry(entry)
+                    .build());
+        } catch (Exception e) {
+            LOG.warn("createEntry {}/{}: {}", parent, entry.getName(), e);
+            return false;
+        }
+        return true;
+    }
+
+    public boolean deleteEntry(String parent, String entryName, boolean isDeleteFileChunk, boolean isRecursive, boolean ignoreRecusiveError) {
+        try {
+            filerGrpcClient.getBlockingStub().deleteEntry(FilerProto.DeleteEntryRequest.newBuilder()
+                    .setDirectory(parent)
+                    .setName(entryName)
+                    .setIsDeleteData(isDeleteFileChunk)
+                    .setIsRecursive(isRecursive)
+                    .setIgnoreRecursiveError(ignoreRecusiveError)
+                    .build());
+        } catch (Exception e) {
+            LOG.warn("deleteEntry {}/{}: {}", parent, entryName, e);
+            return false;
+        }
+        return true;
+    }
+
+    public boolean atomicRenameEntry(String oldParent, String oldName, String newParent, String newName) {
+        try {
+            filerGrpcClient.getBlockingStub().atomicRenameEntry(FilerProto.AtomicRenameEntryRequest.newBuilder()
+                    .setOldDirectory(oldParent)
+                    .setOldName(oldName)
+                    .setNewDirectory(newParent)
+                    .setNewName(newName)
+                    .build());
+        } catch (Exception e) {
+            LOG.warn("atomicRenameEntry {}/{} => {}/{}: {}", oldParent, oldName, newParent, newName, e);
+            return false;
+        }
+        return true;
+    }
+
+    private FilerProto.Entry fixEntryAfterReading(FilerProto.Entry entry) {
+        if (entry.getChunksList().size() <= 0) {
+            return entry;
+        }
+        String fileId = entry.getChunks(0).getFileId();
+        if (fileId != null && fileId.length() != 0) {
+            return entry;
+        }
+        FilerProto.Entry.Builder entryBuilder = entry.toBuilder();
+        entryBuilder.clearChunks();
+        for (FilerProto.FileChunk chunk : entry.getChunksList()) {
+            FilerProto.FileChunk.Builder chunkBuilder = chunk.toBuilder();
+            FilerProto.FileId fid = chunk.getFid();
+            fileId = String.format("%d,%d%x", fid.getVolumeId(), fid.getFileKey(), fid.getCookie());
+            chunkBuilder.setFileId(fileId);
+            entryBuilder.addChunks(chunkBuilder);
+        }
+        return entryBuilder.build();
+    }
+
+}

other/java/client/src/main/java/seaweedfs/client/FilerGrpcClient.java

@@ -0,0 +1,70 @@
+package seaweedfs.client;
+
+import io.grpc.ManagedChannel;
+import io.grpc.ManagedChannelBuilder;
+import io.grpc.netty.shaded.io.grpc.netty.NegotiationType;
+import io.grpc.netty.shaded.io.grpc.netty.NettyChannelBuilder;
+import io.grpc.netty.shaded.io.netty.handler.ssl.SslContext;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import javax.net.ssl.SSLException;
+import java.util.concurrent.TimeUnit;
+
+public class FilerGrpcClient {
+
+    private static final Logger logger = LoggerFactory.getLogger(FilerGrpcClient.class);
+
+    private final ManagedChannel channel;
+    private final SeaweedFilerGrpc.SeaweedFilerBlockingStub blockingStub;
+    private final SeaweedFilerGrpc.SeaweedFilerStub asyncStub;
+    private final SeaweedFilerGrpc.SeaweedFilerFutureStub futureStub;
+
+    static SslContext sslContext;
+
+    static {
+        try {
+            sslContext = FilerSslContext.loadSslContext();
+        } catch (SSLException e) {
+            logger.warn("failed to load ssl context", e);
+        }
+    }
+
+    public FilerGrpcClient(String host, int grpcPort) {
+        this(host, grpcPort, sslContext);
+    }
+
+    public FilerGrpcClient(String host, int grpcPort, SslContext sslContext) {
+
+        this(sslContext == null ?
+                ManagedChannelBuilder.forAddress(host, grpcPort).usePlaintext() :
+                NettyChannelBuilder.forAddress(host, grpcPort)
+                        .negotiationType(NegotiationType.TLS)
+                        .sslContext(sslContext));
+
+    }
+
+    public FilerGrpcClient(ManagedChannelBuilder<?> channelBuilder) {
+        channel = channelBuilder.build();
+        blockingStub = SeaweedFilerGrpc.newBlockingStub(channel);
+        asyncStub = SeaweedFilerGrpc.newStub(channel);
+        futureStub = SeaweedFilerGrpc.newFutureStub(channel);
+    }
+
+    public void shutdown() throws InterruptedException {
+        channel.shutdown().awaitTermination(5, TimeUnit.SECONDS);
+    }
+
+    public SeaweedFilerGrpc.SeaweedFilerBlockingStub getBlockingStub() {
+        return blockingStub;
+    }
+
+    public SeaweedFilerGrpc.SeaweedFilerStub getAsyncStub() {
+        return asyncStub;
+    }
+
+    public SeaweedFilerGrpc.SeaweedFilerFutureStub getFutureStub() {
+        return futureStub;
+    }
+
+}

other/java/client/src/main/java/seaweedfs/client/FilerSslContext.java

@@ -0,0 +1,64 @@
+package seaweedfs.client;
+
+import com.google.common.base.Strings;
+import com.moandjiezana.toml.Toml;
+import io.grpc.netty.shaded.io.grpc.netty.GrpcSslContexts;
+import io.grpc.netty.shaded.io.netty.handler.ssl.SslContext;
+import io.grpc.netty.shaded.io.netty.handler.ssl.SslContextBuilder;
+import io.grpc.netty.shaded.io.netty.handler.ssl.util.InsecureTrustManagerFactory;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import javax.net.ssl.SSLException;
+import java.io.File;
+
+public class FilerSslContext {
+
+    private static final Logger logger = LoggerFactory.getLogger(FilerSslContext.class);
+
+    public static SslContext loadSslContext() throws SSLException {
+        String securityFileName = "security.toml";
+        String home = System.getProperty("user.home");
+        File f1 = new File("./"+securityFileName);
+        File f2 = new File(home + "/.seaweedfs/"+securityFileName);
+        File f3 = new File(home + "/etc/seaweedfs/"+securityFileName);
+
+        File securityFile = f1.exists()? f1 : f2.exists() ? f2 : f3.exists()? f3 : null;
+
+        if (securityFile==null){
+            return null;
+        }
+
+        Toml toml = new Toml().read(securityFile);
+        logger.debug("reading ssl setup from {}", securityFile);
+
+        String trustCertCollectionFilePath = toml.getString("grpc.ca");
+        logger.debug("loading ca from {}", trustCertCollectionFilePath);
+        String clientCertChainFilePath = toml.getString("grpc.client.cert");
+        logger.debug("loading client ca from {}", clientCertChainFilePath);
+        String clientPrivateKeyFilePath = toml.getString("grpc.client.key");
+        logger.debug("loading client key from {}", clientPrivateKeyFilePath);
+
+        if (Strings.isNullOrEmpty(clientPrivateKeyFilePath) && Strings.isNullOrEmpty(clientPrivateKeyFilePath)){
+            return null;
+        }
+
+        // possibly fix the format https://netty.io/wiki/sslcontextbuilder-and-private-key.html
+
+        return buildSslContext(trustCertCollectionFilePath, clientCertChainFilePath, clientPrivateKeyFilePath);
+    }
+
+
+    private static SslContext buildSslContext(String trustCertCollectionFilePath,
+                                             String clientCertChainFilePath,
+                                             String clientPrivateKeyFilePath) throws SSLException {
+        SslContextBuilder builder = GrpcSslContexts.forClient();
+        if (trustCertCollectionFilePath != null) {
+            builder.trustManager(new File(trustCertCollectionFilePath));
+        }
+        if (clientCertChainFilePath != null && clientPrivateKeyFilePath != null) {
+            builder.keyManager(new File(clientCertChainFilePath), new File(clientPrivateKeyFilePath));
+        }
+        return builder.trustManager(InsecureTrustManagerFactory.INSTANCE).build();
+    }
+}

other/java/client/src/main/java/seaweedfs/client/SeaweedRead.java

@@ -0,0 +1,259 @@
+package seaweedfs.client;
+
+import org.apache.http.HttpEntity;
+import org.apache.http.HttpHeaders;
+import org.apache.http.HttpResponse;
+import org.apache.http.client.HttpClient;
+import org.apache.http.client.methods.HttpGet;
+import org.apache.http.impl.client.DefaultHttpClient;
+
+import java.io.Closeable;
+import java.io.IOException;
+import java.io.OutputStream;
+import java.nio.ByteBuffer;
+import java.util.*;
+
+public class SeaweedRead {
+
+    // private static final Logger LOG = LoggerFactory.getLogger(SeaweedRead.class);
+
+    // returns bytesRead
+    public static long read(FilerGrpcClient filerGrpcClient, List<VisibleInterval> visibleIntervals,
+                            final long position, final byte[] buffer, final int bufferOffset,
+                            final int bufferLength) throws IOException {
+
+        List<ChunkView> chunkViews = viewFromVisibles(visibleIntervals, position, bufferLength);
+
+        FilerProto.LookupVolumeRequest.Builder lookupRequest = FilerProto.LookupVolumeRequest.newBuilder();
+        for (ChunkView chunkView : chunkViews) {
+            String vid = parseVolumeId(chunkView.fileId);
+            lookupRequest.addVolumeIds(vid);
+        }
+
+        FilerProto.LookupVolumeResponse lookupResponse = filerGrpcClient
+            .getBlockingStub().lookupVolume(lookupRequest.build());
+
+        Map<String, FilerProto.Locations> vid2Locations = lookupResponse.getLocationsMapMap();
+
+        //TODO parallel this
+        long readCount = 0;
+        int startOffset = bufferOffset;
+        for (ChunkView chunkView : chunkViews) {
+            FilerProto.Locations locations = vid2Locations.get(parseVolumeId(chunkView.fileId));
+            if (locations.getLocationsCount() == 0) {
+                // log here!
+                return 0;
+            }
+
+            int len = readChunkView(position, buffer, startOffset, chunkView, locations);
+
+            readCount += len;
+            startOffset += len;
+
+        }
+
+        return readCount;
+    }
+
+    private static int readChunkView(long position, byte[] buffer, int startOffset, ChunkView chunkView, FilerProto.Locations locations) throws IOException {
+        HttpClient client = new DefaultHttpClient();
+        HttpGet request = new HttpGet(
+            String.format("http://%s/%s", locations.getLocations(0).getUrl(), chunkView.fileId));
+
+        if (!chunkView.isFullChunk) {
+            request.setHeader(HttpHeaders.ACCEPT_ENCODING, "");
+            request.setHeader(HttpHeaders.RANGE,
+                String.format("bytes=%d-%d", chunkView.offset, chunkView.offset + chunkView.size));
+        }
+
+        try {
+            HttpResponse response = client.execute(request);
+            HttpEntity entity = response.getEntity();
+
+            int len = (int) (chunkView.logicOffset - position + chunkView.size);
+            OutputStream outputStream = new ByteBufferOutputStream(ByteBuffer.wrap(buffer, startOffset, len));
+            entity.writeTo(outputStream);
+            // LOG.debug("* read chunkView:{} startOffset:{} length:{}", chunkView, startOffset, len);
+
+            return len;
+
+        } finally {
+            if (client instanceof Closeable) {
+                Closeable t = (Closeable) client;
+                t.close();
+            }
+        }
+    }
+
+    protected static List<ChunkView> viewFromVisibles(List<VisibleInterval> visibleIntervals, long offset, long size) {
+        List<ChunkView> views = new ArrayList<>();
+
+        long stop = offset + size;
+        for (VisibleInterval chunk : visibleIntervals) {
+            if (chunk.start <= offset && offset < chunk.stop && offset < stop) {
+                boolean isFullChunk = chunk.isFullChunk && chunk.start == offset && chunk.stop <= stop;
+                views.add(new ChunkView(
+                    chunk.fileId,
+                    offset - chunk.start,
+                    Math.min(chunk.stop, stop) - offset,
+                    offset,
+                    isFullChunk
+                ));
+                offset = Math.min(chunk.stop, stop);
+            }
+        }
+        return views;
+    }
+
+    public static List<VisibleInterval> nonOverlappingVisibleIntervals(List<FilerProto.FileChunk> chunkList) {
+        FilerProto.FileChunk[] chunks = chunkList.toArray(new FilerProto.FileChunk[0]);
+        Arrays.sort(chunks, new Comparator<FilerProto.FileChunk>() {
+            @Override
+            public int compare(FilerProto.FileChunk a, FilerProto.FileChunk b) {
+                return (int) (a.getMtime() - b.getMtime());
+            }
+        });
+
+        List<VisibleInterval> visibles = new ArrayList<>();
+        for (FilerProto.FileChunk chunk : chunks) {
+            List<VisibleInterval> newVisibles = new ArrayList<>();
+            visibles = mergeIntoVisibles(visibles, newVisibles, chunk);
+        }
+
+        return visibles;
+    }
+
+    private static List<VisibleInterval> mergeIntoVisibles(List<VisibleInterval> visibles,
+                                                           List<VisibleInterval> newVisibles,
+                                                           FilerProto.FileChunk chunk) {
+        VisibleInterval newV = new VisibleInterval(
+            chunk.getOffset(),
+            chunk.getOffset() + chunk.getSize(),
+            chunk.getFileId(),
+            chunk.getMtime(),
+            true
+        );
+
+        // easy cases to speed up
+        if (visibles.size() == 0) {
+            visibles.add(newV);
+            return visibles;
+        }
+        if (visibles.get(visibles.size() - 1).stop <= chunk.getOffset()) {
+            visibles.add(newV);
+            return visibles;
+        }
+
+        for (VisibleInterval v : visibles) {
+            if (v.start < chunk.getOffset() && chunk.getOffset() < v.stop) {
+                newVisibles.add(new VisibleInterval(
+                    v.start,
+                    chunk.getOffset(),
+                    v.fileId,
+                    v.modifiedTime,
+                    false
+                ));
+            }
+            long chunkStop = chunk.getOffset() + chunk.getSize();
+            if (v.start < chunkStop && chunkStop < v.stop) {
+                newVisibles.add(new VisibleInterval(
+                    chunkStop,
+                    v.stop,
+                    v.fileId,
+                    v.modifiedTime,
+                    false
+                ));
+            }
+            if (chunkStop <= v.start || v.stop <= chunk.getOffset()) {
+                newVisibles.add(v);
+            }
+        }
+        newVisibles.add(newV);
+
+        // keep everything sorted
+        for (int i = newVisibles.size() - 1; i >= 0; i--) {
+            if (i > 0 && newV.start < newVisibles.get(i - 1).start) {
+                newVisibles.set(i, newVisibles.get(i - 1));
+            } else {
+                newVisibles.set(i, newV);
+                break;
+            }
+        }
+
+        return newVisibles;
+    }
+
+    public static String parseVolumeId(String fileId) {
+        int commaIndex = fileId.lastIndexOf(',');
+        if (commaIndex > 0) {
+            return fileId.substring(0, commaIndex);
+        }
+        return fileId;
+    }
+
+    public static long totalSize(List<FilerProto.FileChunk> chunksList) {
+        long size = 0;
+        for (FilerProto.FileChunk chunk : chunksList) {
+            long t = chunk.getOffset() + chunk.getSize();
+            if (size < t) {
+                size = t;
+            }
+        }
+        return size;
+    }
+
+    public static class VisibleInterval {
+        public final long start;
+        public final long stop;
+        public final long modifiedTime;
+        public final String fileId;
+        public final boolean isFullChunk;
+
+        public VisibleInterval(long start, long stop, String fileId, long modifiedTime, boolean isFullChunk) {
+            this.start = start;
+            this.stop = stop;
+            this.modifiedTime = modifiedTime;
+            this.fileId = fileId;
+            this.isFullChunk = isFullChunk;
+        }
+
+        @Override
+        public String toString() {
+            return "VisibleInterval{" +
+                "start=" + start +
+                ", stop=" + stop +
+                ", modifiedTime=" + modifiedTime +
+                ", fileId='" + fileId + '\'' +
+                ", isFullChunk=" + isFullChunk +
+                '}';
+        }
+    }
+
+    public static class ChunkView {
+        public final String fileId;
+        public final long offset;
+        public final long size;
+        public final long logicOffset;
+        public final boolean isFullChunk;
+
+        public ChunkView(String fileId, long offset, long size, long logicOffset, boolean isFullChunk) {
+            this.fileId = fileId;
+            this.offset = offset;
+            this.size = size;
+            this.logicOffset = logicOffset;
+            this.isFullChunk = isFullChunk;
+        }
+
+        @Override
+        public String toString() {
+            return "ChunkView{" +
+                "fileId='" + fileId + '\'' +
+                ", offset=" + offset +
+                ", size=" + size +
+                ", logicOffset=" + logicOffset +
+                ", isFullChunk=" + isFullChunk +
+                '}';
+        }
+    }
+
+}

other/java/client/src/main/java/seaweedfs/client/SeaweedWrite.java

@@ -0,0 +1,95 @@
+package seaweedfs.client;
+
+import org.apache.http.HttpResponse;
+import org.apache.http.client.HttpClient;
+import org.apache.http.client.methods.HttpPost;
+import org.apache.http.entity.mime.HttpMultipartMode;
+import org.apache.http.entity.mime.MultipartEntityBuilder;
+import org.apache.http.impl.client.DefaultHttpClient;
+
+import java.io.ByteArrayInputStream;
+import java.io.Closeable;
+import java.io.IOException;
+import java.io.InputStream;
+
+public class SeaweedWrite {
+
+    public static void writeData(FilerProto.Entry.Builder entry,
+                                 final String replication,
+                                 final FilerGrpcClient filerGrpcClient,
+                                 final long offset,
+                                 final byte[] bytes,
+                                 final long bytesOffset, final long bytesLength) throws IOException {
+        FilerProto.AssignVolumeResponse response = filerGrpcClient.getBlockingStub().assignVolume(
+                FilerProto.AssignVolumeRequest.newBuilder()
+                        .setCollection("")
+                        .setReplication(replication)
+                        .setDataCenter("")
+                        .setReplication("")
+                        .setTtlSec(0)
+                        .build());
+        String fileId = response.getFileId();
+        String url = response.getUrl();
+        String auth = response.getAuth();
+        String targetUrl = String.format("http://%s/%s", url, fileId);
+
+        String etag = multipartUpload(targetUrl, auth, bytes, bytesOffset, bytesLength);
+
+        entry.addChunks(FilerProto.FileChunk.newBuilder()
+                .setFileId(fileId)
+                .setOffset(offset)
+                .setSize(bytesLength)
+                .setMtime(System.currentTimeMillis() / 10000L)
+                .setETag(etag)
+        );
+
+    }
+
+    public static void writeMeta(final FilerGrpcClient filerGrpcClient,
+                                 final String parentDirectory, final FilerProto.Entry.Builder entry) {
+        filerGrpcClient.getBlockingStub().createEntry(
+                FilerProto.CreateEntryRequest.newBuilder()
+                        .setDirectory(parentDirectory)
+                        .setEntry(entry)
+                        .build()
+        );
+    }
+
+    private static String multipartUpload(String targetUrl,
+                                          String auth,
+                                          final byte[] bytes,
+                                          final long bytesOffset, final long bytesLength) throws IOException {
+
+        HttpClient client = new DefaultHttpClient();
+
+        InputStream inputStream = new ByteArrayInputStream(bytes, (int) bytesOffset, (int) bytesLength);
+
+        HttpPost post = new HttpPost(targetUrl);
+        if (auth != null && auth.length() != 0) {
+            post.addHeader("Authorization", "BEARER " + auth);
+        }
+
+        post.setEntity(MultipartEntityBuilder.create()
+                .setMode(HttpMultipartMode.BROWSER_COMPATIBLE)
+                .addBinaryBody("upload", inputStream)
+                .build());
+
+        try {
+            HttpResponse response = client.execute(post);
+
+            String etag = response.getLastHeader("ETag").getValue();
+
+            if (etag != null && etag.startsWith("\"") && etag.endsWith("\"")) {
+                etag = etag.substring(1, etag.length() - 1);
+            }
+
+            return etag;
+        } finally {
+            if (client instanceof Closeable) {
+                Closeable t = (Closeable) client;
+                t.close();
+            }
+        }
+
+    }
+}

other/java/client/src/main/proto/filer.proto

@@ -0,0 +1,220 @@
+syntax = "proto3";
+
+package filer_pb;
+
+option java_package = "seaweedfs.client";
+option java_outer_classname = "FilerProto";
+
+//////////////////////////////////////////////////
+
+service SeaweedFiler {
+
+    rpc LookupDirectoryEntry (LookupDirectoryEntryRequest) returns (LookupDirectoryEntryResponse) {
+    }
+
+    rpc ListEntries (ListEntriesRequest) returns (stream ListEntriesResponse) {
+    }
+
+    rpc CreateEntry (CreateEntryRequest) returns (CreateEntryResponse) {
+    }
+
+    rpc UpdateEntry (UpdateEntryRequest) returns (UpdateEntryResponse) {
+    }
+
+    rpc DeleteEntry (DeleteEntryRequest) returns (DeleteEntryResponse) {
+    }
+
+    rpc AtomicRenameEntry (AtomicRenameEntryRequest) returns (AtomicRenameEntryResponse) {
+    }
+
+    rpc AssignVolume (AssignVolumeRequest) returns (AssignVolumeResponse) {
+    }
+
+    rpc LookupVolume (LookupVolumeRequest) returns (LookupVolumeResponse) {
+    }
+
+    rpc DeleteCollection (DeleteCollectionRequest) returns (DeleteCollectionResponse) {
+    }
+
+    rpc Statistics (StatisticsRequest) returns (StatisticsResponse) {
+    }
+
+    rpc GetFilerConfiguration (GetFilerConfigurationRequest) returns (GetFilerConfigurationResponse) {
+    }
+
+}
+
+//////////////////////////////////////////////////
+
+message LookupDirectoryEntryRequest {
+    string directory = 1;
+    string name = 2;
+}
+
+message LookupDirectoryEntryResponse {
+    Entry entry = 1;
+}
+
+message ListEntriesRequest {
+    string directory = 1;
+    string prefix = 2;
+    string startFromFileName = 3;
+    bool inclusiveStartFrom = 4;
+    uint32 limit = 5;
+}
+
+message ListEntriesResponse {
+    Entry entry = 1;
+}
+
+message Entry {
+    string name = 1;
+    bool is_directory = 2;
+    repeated FileChunk chunks = 3;
+    FuseAttributes attributes = 4;
+    map<string, bytes> extended = 5;
+}
+
+message FullEntry {
+    string dir = 1;
+    Entry entry = 2;
+}
+
+message EventNotification {
+    Entry old_entry = 1;
+    Entry new_entry = 2;
+    bool delete_chunks = 3;
+    string new_parent_path = 4;
+}
+
+message FileChunk {
+    string file_id = 1; // to be deprecated
+    int64 offset = 2;
+    uint64 size = 3;
+    int64 mtime = 4;
+    string e_tag = 5;
+    string source_file_id = 6; // to be deprecated
+    FileId fid = 7;
+    FileId source_fid = 8;
+}
+
+message FileId {
+    uint32 volume_id = 1;
+    uint64 file_key = 2;
+    fixed32 cookie = 3;
+}
+
+message FuseAttributes {
+    uint64 file_size = 1;
+    int64 mtime = 2; // unix time in seconds
+    uint32 file_mode = 3;
+    uint32 uid = 4;
+    uint32 gid = 5;
+    int64 crtime = 6; // unix time in seconds
+    string mime = 7;
+    string replication = 8;
+    string collection = 9;
+    int32 ttl_sec = 10;
+    string user_name = 11; // for hdfs
+    repeated string group_name = 12; // for hdfs
+    string symlink_target = 13;
+}
+
+message CreateEntryRequest {
+    string directory = 1;
+    Entry entry = 2;
+}
+
+message CreateEntryResponse {
+}
+
+message UpdateEntryRequest {
+    string directory = 1;
+    Entry entry = 2;
+}
+message UpdateEntryResponse {
+}
+
+message DeleteEntryRequest {
+    string directory = 1;
+    string name = 2;
+    // bool is_directory = 3;
+    bool is_delete_data = 4;
+    bool is_recursive = 5;
+    bool ignore_recursive_error = 6;
+}
+
+message DeleteEntryResponse {
+}
+
+message AtomicRenameEntryRequest {
+    string old_directory = 1;
+    string old_name = 2;
+    string new_directory = 3;
+    string new_name = 4;
+}
+
+message AtomicRenameEntryResponse {
+}
+
+message AssignVolumeRequest {
+    int32 count = 1;
+    string collection = 2;
+    string replication = 3;
+    int32 ttl_sec = 4;
+    string data_center = 5;
+}
+
+message AssignVolumeResponse {
+    string file_id = 1;
+    string url = 2;
+    string public_url = 3;
+    int32 count = 4;
+    string auth = 5;
+}
+
+message LookupVolumeRequest {
+    repeated string volume_ids = 1;
+}
+
+message Locations {
+    repeated Location locations = 1;
+}
+
+message Location {
+    string url = 1;
+    string public_url = 2;
+}
+message LookupVolumeResponse {
+    map<string, Locations> locations_map = 1;
+}
+
+message DeleteCollectionRequest {
+    string collection = 1;
+}
+
+message DeleteCollectionResponse {
+}
+
+message StatisticsRequest {
+    string replication = 1;
+    string collection = 2;
+    string ttl = 3;
+}
+message StatisticsResponse {
+    string replication = 1;
+    string collection = 2;
+    string ttl = 3;
+    uint64 total_size = 4;
+    uint64 used_size = 5;
+    uint64 file_count = 6;
+}
+
+message GetFilerConfigurationRequest {
+}
+message GetFilerConfigurationResponse {
+    repeated string masters = 1;
+    string replication = 2;
+    string collection = 3;
+    uint32 max_mb = 4;
+}

other/java/client/src/test/java/seaweedfs/client/SeaweedFilerTest.java

@@ -0,0 +1,23 @@
+package seaweedfs.client;
+
+import java.util.List;
+
+public class SeaweedFilerTest {
+    public static void main(String[] args){
+
+        FilerClient filerClient = new FilerClient("localhost", 18888);
+
+        List<FilerProto.Entry> entries = filerClient.listEntries("/");
+
+        for (FilerProto.Entry entry : entries) {
+            System.out.println(entry.toString());
+        }
+
+        filerClient.mkdirs("/new_folder", 0755);
+        filerClient.touch("/new_folder/new_empty_file", 0755);
+        filerClient.touch("/new_folder/new_empty_file2", 0755);
+        filerClient.rm("/new_folder/new_empty_file", false, true);
+        filerClient.rm("/new_folder", true, true);
+
+    }
+}

other/java/client/src/test/java/seaweedfs/client/SeaweedReadTest.java

@@ -0,0 +1,63 @@
+package seaweedfs.client;
+
+import org.junit.Assert;
+import org.junit.Test;
+
+import java.util.ArrayList;
+import java.util.List;
+
+public class SeaweedReadTest {
+
+    @Test
+    public void testNonOverlappingVisibleIntervals() {
+        List<FilerProto.FileChunk> chunks = new ArrayList<>();
+        chunks.add(FilerProto.FileChunk.newBuilder()
+            .setFileId("aaa")
+            .setOffset(0)
+            .setSize(100)
+            .setMtime(1000)
+            .build());
+        chunks.add(FilerProto.FileChunk.newBuilder()
+            .setFileId("bbb")
+            .setOffset(100)
+            .setSize(133)
+            .setMtime(2000)
+            .build());
+
+        List<SeaweedRead.VisibleInterval> visibleIntervals = SeaweedRead.nonOverlappingVisibleIntervals(chunks);
+        for (SeaweedRead.VisibleInterval visibleInterval : visibleIntervals) {
+            System.out.println("visible:" + visibleInterval);
+        }
+
+        Assert.assertEquals(visibleIntervals.size(), 2);
+
+        SeaweedRead.VisibleInterval visibleInterval = visibleIntervals.get(0);
+        Assert.assertEquals(visibleInterval.start, 0);
+        Assert.assertEquals(visibleInterval.stop, 100);
+        Assert.assertEquals(visibleInterval.modifiedTime, 1000);
+        Assert.assertEquals(visibleInterval.fileId, "aaa");
+
+        visibleInterval = visibleIntervals.get(1);
+        Assert.assertEquals(visibleInterval.start, 100);
+        Assert.assertEquals(visibleInterval.stop, 233);
+        Assert.assertEquals(visibleInterval.modifiedTime, 2000);
+        Assert.assertEquals(visibleInterval.fileId, "bbb");
+
+        List<SeaweedRead.ChunkView> chunkViews = SeaweedRead.viewFromVisibles(visibleIntervals, 0, 233);
+
+        SeaweedRead.ChunkView chunkView = chunkViews.get(0);
+        Assert.assertEquals(chunkView.offset, 0);
+        Assert.assertEquals(chunkView.size, 100);
+        Assert.assertEquals(chunkView.logicOffset, 0);
+        Assert.assertEquals(chunkView.fileId, "aaa");
+
+        chunkView = chunkViews.get(1);
+        Assert.assertEquals(chunkView.offset, 0);
+        Assert.assertEquals(chunkView.size, 133);
+        Assert.assertEquals(chunkView.logicOffset, 100);
+        Assert.assertEquals(chunkView.fileId, "bbb");
+
+
+    }
+
+}

other/java/hdfs2/dependency-reduced-pom.xml

@@ -0,0 +1,133 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
+  <parent>
+    <artifactId>oss-parent</artifactId>
+    <groupId>org.sonatype.oss</groupId>
+    <version>9</version>
+    <relativePath>../pom.xml/pom.xml</relativePath>
+  </parent>
+  <modelVersion>4.0.0</modelVersion>
+  <groupId>com.github.chrislusf</groupId>
+  <artifactId>seaweedfs-hadoop2-client</artifactId>
+  <version>${seaweedfs.client.version}</version>
+  <build>
+    <plugins>
+      <plugin>
+        <artifactId>maven-compiler-plugin</artifactId>
+        <configuration>
+          <source>7</source>
+          <target>7</target>
+        </configuration>
+      </plugin>
+      <plugin>
+        <artifactId>maven-shade-plugin</artifactId>
+        <version>3.2.1</version>
+        <executions>
+          <execution>
+            <phase>package</phase>
+            <goals>
+              <goal>shade</goal>
+            </goals>
+            <configuration>
+              <filters>
+                <filter>
+                  <artifact>*:*</artifact>
+                  <excludes>
+                    <exclude>META-INF/*.SF</exclude>
+                    <exclude>META-INF/*.DSA</exclude>
+                    <exclude>META-INF/*.RSA</exclude>
+                    <exclude>org/slf4j/**</exclude>
+                    <exclude>META-INF/maven/org.slf4j/**</exclude>
+                  </excludes>
+                </filter>
+              </filters>
+              <transformers>
+                <transformer />
+              </transformers>
+              <relocations>
+                <relocation>
+                  <pattern>com.google</pattern>
+                  <shadedPattern>shaded.com.google</shadedPattern>
+                </relocation>
+                <relocation>
+                  <pattern>io.grpc.internal</pattern>
+                  <shadedPattern>shaded.io.grpc.internal</shadedPattern>
+                </relocation>
+                <relocation>
+                  <pattern>org.apache.commons</pattern>
+                  <shadedPattern>shaded.org.apache.commons</shadedPattern>
+                  <excludes>
+                    <exclude>org.apache.hadoop</exclude>
+                    <exclude>org.apache.log4j</exclude>
+                  </excludes>
+                </relocation>
+                <relocation>
+                  <pattern>org.apache.http</pattern>
+                  <shadedPattern>shaded.org.apache.http</shadedPattern>
+                </relocation>
+              </relocations>
+            </configuration>
+          </execution>
+        </executions>
+      </plugin>
+      <plugin>
+        <artifactId>maven-gpg-plugin</artifactId>
+        <version>1.5</version>
+        <executions>
+          <execution>
+            <id>sign-artifacts</id>
+            <phase>verify</phase>
+            <goals>
+              <goal>sign</goal>
+            </goals>
+          </execution>
+        </executions>
+      </plugin>
+      <plugin>
+        <groupId>org.sonatype.plugins</groupId>
+        <artifactId>nexus-staging-maven-plugin</artifactId>
+        <version>1.6.7</version>
+        <extensions>true</extensions>
+        <configuration>
+          <serverId>ossrh</serverId>
+          <nexusUrl>https://oss.sonatype.org/</nexusUrl>
+          <autoReleaseAfterClose>true</autoReleaseAfterClose>
+        </configuration>
+      </plugin>
+      <plugin>
+        <artifactId>maven-source-plugin</artifactId>
+        <version>2.2.1</version>
+        <executions>
+          <execution>
+            <id>attach-sources</id>
+            <goals>
+              <goal>jar-no-fork</goal>
+            </goals>
+          </execution>
+        </executions>
+      </plugin>
+      <plugin>
+        <artifactId>maven-javadoc-plugin</artifactId>
+        <version>2.9.1</version>
+        <executions>
+          <execution>
+            <id>attach-javadocs</id>
+            <goals>
+              <goal>jar</goal>
+            </goals>
+          </execution>
+        </executions>
+      </plugin>
+    </plugins>
+  </build>
+  <distributionManagement>
+    <snapshotRepository>
+      <id>ossrh</id>
+      <url>https://oss.sonatype.org/content/repositories/snapshots</url>
+    </snapshotRepository>
+  </distributionManagement>
+  <properties>
+    <seaweedfs.client.version>1.2.4</seaweedfs.client.version>
+    <hadoop.version>2.9.2</hadoop.version>
+  </properties>
+</project>

other/java/hdfs2/pom.xml

@@ -0,0 +1,163 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project xmlns="http://maven.apache.org/POM/4.0.0"
+         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
+    <modelVersion>4.0.0</modelVersion>
+
+    <properties>
+        <seaweedfs.client.version>1.2.4</seaweedfs.client.version>
+        <hadoop.version>2.9.2</hadoop.version>
+    </properties>
+
+    <groupId>com.github.chrislusf</groupId>
+    <artifactId>seaweedfs-hadoop2-client</artifactId>
+    <version>${seaweedfs.client.version}</version>
+
+    <parent>
+        <groupId>org.sonatype.oss</groupId>
+        <artifactId>oss-parent</artifactId>
+        <version>9</version>
+    </parent>
+
+    <distributionManagement>
+        <snapshotRepository>
+            <id>ossrh</id>
+            <url>https://oss.sonatype.org/content/repositories/snapshots</url>
+        </snapshotRepository>
+    </distributionManagement>
+    <build>
+        <plugins>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-compiler-plugin</artifactId>
+                <configuration>
+                    <source>7</source>
+                    <target>7</target>
+                </configuration>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-shade-plugin</artifactId>
+                <version>3.2.1</version>
+                <executions>
+                    <execution>
+                        <phase>package</phase>
+                        <goals>
+                            <goal>shade</goal>
+                        </goals>
+                        <configuration>
+                            <filters>
+                                <filter>
+                                    <artifact>*:*</artifact>
+                                    <excludes>
+                                        <exclude>META-INF/*.SF</exclude>
+                                        <exclude>META-INF/*.DSA</exclude>
+                                        <exclude>META-INF/*.RSA</exclude>
+                                        <exclude>org/slf4j/**</exclude>
+                                        <exclude>META-INF/maven/org.slf4j/**</exclude>
+                                    </excludes>
+                                </filter>
+                            </filters>
+                            <transformers>
+                                <transformer
+                                        implementation="org.apache.maven.plugins.shade.resource.ServicesResourceTransformer"/>
+                            </transformers>
+                            <relocations>
+                                <relocation>
+                                    <pattern>com.google</pattern>
+                                    <shadedPattern>shaded.com.google</shadedPattern>
+                                </relocation>
+                                <relocation>
+                                    <pattern>io.grpc.internal</pattern>
+                                    <shadedPattern>shaded.io.grpc.internal</shadedPattern>
+                                </relocation>
+                                <relocation>
+                                    <pattern>org.apache.commons</pattern>
+                                    <shadedPattern>shaded.org.apache.commons</shadedPattern>
+                                    <excludes>
+                                        <exclude>org.apache.hadoop</exclude>
+                                        <exclude>org.apache.log4j</exclude>
+                                    </excludes>
+                                </relocation>
+                                <relocation>
+                                    <pattern>org.apache.http</pattern>
+                                    <shadedPattern>shaded.org.apache.http</shadedPattern>
+                                </relocation>
+                            </relocations>
+                        </configuration>
+                    </execution>
+                </executions>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-gpg-plugin</artifactId>
+                <version>1.5</version>
+                <executions>
+                    <execution>
+                        <id>sign-artifacts</id>
+                        <phase>verify</phase>
+                        <goals>
+                            <goal>sign</goal>
+                        </goals>
+                    </execution>
+                </executions>
+            </plugin>
+            <plugin>
+                <groupId>org.sonatype.plugins</groupId>
+                <artifactId>nexus-staging-maven-plugin</artifactId>
+                <version>1.6.7</version>
+                <extensions>true</extensions>
+                <configuration>
+                    <serverId>ossrh</serverId>
+                    <nexusUrl>https://oss.sonatype.org/</nexusUrl>
+                    <autoReleaseAfterClose>true</autoReleaseAfterClose>
+                </configuration>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-source-plugin</artifactId>
+                <version>2.2.1</version>
+                <executions>
+                    <execution>
+                        <id>attach-sources</id>
+                        <goals>
+                            <goal>jar-no-fork</goal>
+                        </goals>
+                    </execution>
+                </executions>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-javadoc-plugin</artifactId>
+                <version>2.9.1</version>
+                <executions>
+                    <execution>
+                        <id>attach-javadocs</id>
+                        <goals>
+                            <goal>jar</goal>
+                        </goals>
+                    </execution>
+                </executions>
+            </plugin>
+        </plugins>
+    </build>
+
+    <dependencies>
+        <dependency>
+            <groupId>org.apache.hadoop</groupId>
+            <artifactId>hadoop-client</artifactId>
+            <version>${hadoop.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>com.github.chrislusf</groupId>
+            <artifactId>seaweedfs-client</artifactId>
+            <version>${seaweedfs.client.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>org.apache.hadoop</groupId>
+            <artifactId>hadoop-common</artifactId>
+            <version>${hadoop.version}</version>
+        </dependency>
+    </dependencies>
+
+</project>

other/java/hdfs2/src/main/java/seaweed/hdfs/ReadBuffer.java

@@ -0,0 +1,137 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package seaweed.hdfs;
+
+import java.util.concurrent.CountDownLatch;
+
+class ReadBuffer {
+
+  private SeaweedInputStream stream;
+  private long offset;                   // offset within the file for the buffer
+  private int length;                    // actual length, set after the buffer is filles
+  private int requestedLength;           // requested length of the read
+  private byte[] buffer;                 // the buffer itself
+  private int bufferindex = -1;          // index in the buffers array in Buffer manager
+  private ReadBufferStatus status;             // status of the buffer
+  private CountDownLatch latch = null;   // signaled when the buffer is done reading, so any client
+  // waiting on this buffer gets unblocked
+
+  // fields to help with eviction logic
+  private long timeStamp = 0;  // tick at which buffer became available to read
+  private boolean isFirstByteConsumed = false;
+  private boolean isLastByteConsumed = false;
+  private boolean isAnyByteConsumed = false;
+
+  public SeaweedInputStream getStream() {
+    return stream;
+  }
+
+  public void setStream(SeaweedInputStream stream) {
+    this.stream = stream;
+  }
+
+  public long getOffset() {
+    return offset;
+  }
+
+  public void setOffset(long offset) {
+    this.offset = offset;
+  }
+
+  public int getLength() {
+    return length;
+  }
+
+  public void setLength(int length) {
+    this.length = length;
+  }
+
+  public int getRequestedLength() {
+    return requestedLength;
+  }
+
+  public void setRequestedLength(int requestedLength) {
+    this.requestedLength = requestedLength;
+  }
+
+  public byte[] getBuffer() {
+    return buffer;
+  }
+
+  public void setBuffer(byte[] buffer) {
+    this.buffer = buffer;
+  }
+
+  public int getBufferindex() {
+    return bufferindex;
+  }
+
+  public void setBufferindex(int bufferindex) {
+    this.bufferindex = bufferindex;
+  }
+
+  public ReadBufferStatus getStatus() {
+    return status;
+  }
+
+  public void setStatus(ReadBufferStatus status) {
+    this.status = status;
+  }
+
+  public CountDownLatch getLatch() {
+    return latch;
+  }
+
+  public void setLatch(CountDownLatch latch) {
+    this.latch = latch;
+  }
+
+  public long getTimeStamp() {
+    return timeStamp;
+  }
+
+  public void setTimeStamp(long timeStamp) {
+    this.timeStamp = timeStamp;
+  }
+
+  public boolean isFirstByteConsumed() {
+    return isFirstByteConsumed;
+  }
+
+  public void setFirstByteConsumed(boolean isFirstByteConsumed) {
+    this.isFirstByteConsumed = isFirstByteConsumed;
+  }
+
+  public boolean isLastByteConsumed() {
+    return isLastByteConsumed;
+  }
+
+  public void setLastByteConsumed(boolean isLastByteConsumed) {
+    this.isLastByteConsumed = isLastByteConsumed;
+  }
+
+  public boolean isAnyByteConsumed() {
+    return isAnyByteConsumed;
+  }
+
+  public void setAnyByteConsumed(boolean isAnyByteConsumed) {
+    this.isAnyByteConsumed = isAnyByteConsumed;
+  }
+
+}

other/java/hdfs2/src/main/java/seaweed/hdfs/ReadBufferManager.java

@@ -0,0 +1,394 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package seaweed.hdfs;
+
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.util.Collection;
+import java.util.LinkedList;
+import java.util.Queue;
+import java.util.Stack;
+import java.util.concurrent.CountDownLatch;
+
+/**
+ * The Read Buffer Manager for Rest AbfsClient.
+ */
+final class ReadBufferManager {
+  private static final Logger LOGGER = LoggerFactory.getLogger(ReadBufferManager.class);
+
+  private static final int NUM_BUFFERS = 16;
+  private static final int BLOCK_SIZE = 4 * 1024 * 1024;
+  private static final int NUM_THREADS = 8;
+  private static final int THRESHOLD_AGE_MILLISECONDS = 3000; // have to see if 3 seconds is a good threshold
+
+  private Thread[] threads = new Thread[NUM_THREADS];
+  private byte[][] buffers;    // array of byte[] buffers, to hold the data that is read
+  private Stack<Integer> freeList = new Stack<>();   // indices in buffers[] array that are available
+
+  private Queue<ReadBuffer> readAheadQueue = new LinkedList<>(); // queue of requests that are not picked up by any worker thread yet
+  private LinkedList<ReadBuffer> inProgressList = new LinkedList<>(); // requests being processed by worker threads
+  private LinkedList<ReadBuffer> completedReadList = new LinkedList<>(); // buffers available for reading
+  private static final ReadBufferManager BUFFER_MANAGER; // singleton, initialized in static initialization block
+
+  static {
+    BUFFER_MANAGER = new ReadBufferManager();
+    BUFFER_MANAGER.init();
+  }
+
+  static ReadBufferManager getBufferManager() {
+    return BUFFER_MANAGER;
+  }
+
+  private void init() {
+    buffers = new byte[NUM_BUFFERS][];
+    for (int i = 0; i < NUM_BUFFERS; i++) {
+      buffers[i] = new byte[BLOCK_SIZE];  // same buffers are reused. The byte array never goes back to GC
+      freeList.add(i);
+    }
+    for (int i = 0; i < NUM_THREADS; i++) {
+      Thread t = new Thread(new ReadBufferWorker(i));
+      t.setDaemon(true);
+      threads[i] = t;
+      t.setName("SeaweedFS-prefetch-" + i);
+      t.start();
+    }
+    ReadBufferWorker.UNLEASH_WORKERS.countDown();
+  }
+
+  // hide instance constructor
+  private ReadBufferManager() {
+  }
+
+
+  /*
+   *
+   *  SeaweedInputStream-facing methods
+   *
+   */
+
+
+  /**
+   * {@link SeaweedInputStream} calls this method to queue read-aheads.
+   *
+   * @param stream          The {@link SeaweedInputStream} for which to do the read-ahead
+   * @param requestedOffset The offset in the file which shoukd be read
+   * @param requestedLength The length to read
+   */
+  void queueReadAhead(final SeaweedInputStream stream, final long requestedOffset, final int requestedLength) {
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("Start Queueing readAhead for {} offset {} length {}",
+          stream.getPath(), requestedOffset, requestedLength);
+    }
+    ReadBuffer buffer;
+    synchronized (this) {
+      if (isAlreadyQueued(stream, requestedOffset)) {
+        return; // already queued, do not queue again
+      }
+      if (freeList.isEmpty() && !tryEvict()) {
+        return; // no buffers available, cannot queue anything
+      }
+
+      buffer = new ReadBuffer();
+      buffer.setStream(stream);
+      buffer.setOffset(requestedOffset);
+      buffer.setLength(0);
+      buffer.setRequestedLength(requestedLength);
+      buffer.setStatus(ReadBufferStatus.NOT_AVAILABLE);
+      buffer.setLatch(new CountDownLatch(1));
+
+      Integer bufferIndex = freeList.pop();  // will return a value, since we have checked size > 0 already
+
+      buffer.setBuffer(buffers[bufferIndex]);
+      buffer.setBufferindex(bufferIndex);
+      readAheadQueue.add(buffer);
+      notifyAll();
+    }
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("Done q-ing readAhead for file {} offset {} buffer idx {}",
+          stream.getPath(), requestedOffset, buffer.getBufferindex());
+    }
+  }
+
+
+  /**
+   * {@link SeaweedInputStream} calls this method read any bytes already available in a buffer (thereby saving a
+   * remote read). This returns the bytes if the data already exists in buffer. If there is a buffer that is reading
+   * the requested offset, then this method blocks until that read completes. If the data is queued in a read-ahead
+   * but not picked up by a worker thread yet, then it cancels that read-ahead and reports cache miss. This is because
+   * depending on worker thread availability, the read-ahead may take a while - the calling thread can do it's own
+   * read to get the data faster (copmared to the read waiting in queue for an indeterminate amount of time).
+   *
+   * @param stream   the file to read bytes for
+   * @param position the offset in the file to do a read for
+   * @param length   the length to read
+   * @param buffer   the buffer to read data into. Note that the buffer will be written into from offset 0.
+   * @return the number of bytes read
+   */
+  int getBlock(final SeaweedInputStream stream, final long position, final int length, final byte[] buffer) {
+    // not synchronized, so have to be careful with locking
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("getBlock for file {}  position {}  thread {}",
+          stream.getPath(), position, Thread.currentThread().getName());
+    }
+
+    waitForProcess(stream, position);
+
+    int bytesRead = 0;
+    synchronized (this) {
+      bytesRead = getBlockFromCompletedQueue(stream, position, length, buffer);
+    }
+    if (bytesRead > 0) {
+      if (LOGGER.isTraceEnabled()) {
+        LOGGER.trace("Done read from Cache for {} position {} length {}",
+            stream.getPath(), position, bytesRead);
+      }
+      return bytesRead;
+    }
+
+    // otherwise, just say we got nothing - calling thread can do its own read
+    return 0;
+  }
+
+  /*
+   *
+   *  Internal methods
+   *
+   */
+
+  private void waitForProcess(final SeaweedInputStream stream, final long position) {
+    ReadBuffer readBuf;
+    synchronized (this) {
+      clearFromReadAheadQueue(stream, position);
+      readBuf = getFromList(inProgressList, stream, position);
+    }
+    if (readBuf != null) {         // if in in-progress queue, then block for it
+      try {
+        if (LOGGER.isTraceEnabled()) {
+          LOGGER.trace("got a relevant read buffer for file {} offset {} buffer idx {}",
+              stream.getPath(), readBuf.getOffset(), readBuf.getBufferindex());
+        }
+        readBuf.getLatch().await();  // blocking wait on the caller stream's thread
+        // Note on correctness: readBuf gets out of inProgressList only in 1 place: after worker thread
+        // is done processing it (in doneReading). There, the latch is set after removing the buffer from
+        // inProgressList. So this latch is safe to be outside the synchronized block.
+        // Putting it in synchronized would result in a deadlock, since this thread would be holding the lock
+        // while waiting, so no one will be able to  change any state. If this becomes more complex in the future,
+        // then the latch cane be removed and replaced with wait/notify whenever inProgressList is touched.
+      } catch (InterruptedException ex) {
+        Thread.currentThread().interrupt();
+      }
+      if (LOGGER.isTraceEnabled()) {
+        LOGGER.trace("latch done for file {} buffer idx {} length {}",
+            stream.getPath(), readBuf.getBufferindex(), readBuf.getLength());
+      }
+    }
+  }
+
+  /**
+   * If any buffer in the completedlist can be reclaimed then reclaim it and return the buffer to free list.
+   * The objective is to find just one buffer - there is no advantage to evicting more than one.
+   *
+   * @return whether the eviction succeeeded - i.e., were we able to free up one buffer
+   */
+  private synchronized boolean tryEvict() {
+    ReadBuffer nodeToEvict = null;
+    if (completedReadList.size() <= 0) {
+      return false;  // there are no evict-able buffers
+    }
+
+    // first, try buffers where all bytes have been consumed (approximated as first and last bytes consumed)
+    for (ReadBuffer buf : completedReadList) {
+      if (buf.isFirstByteConsumed() && buf.isLastByteConsumed()) {
+        nodeToEvict = buf;
+        break;
+      }
+    }
+    if (nodeToEvict != null) {
+      return evict(nodeToEvict);
+    }
+
+    // next, try buffers where any bytes have been consumed (may be a bad idea? have to experiment and see)
+    for (ReadBuffer buf : completedReadList) {
+      if (buf.isAnyByteConsumed()) {
+        nodeToEvict = buf;
+        break;
+      }
+    }
+
+    if (nodeToEvict != null) {
+      return evict(nodeToEvict);
+    }
+
+    // next, try any old nodes that have not been consumed
+    long earliestBirthday = Long.MAX_VALUE;
+    for (ReadBuffer buf : completedReadList) {
+      if (buf.getTimeStamp() < earliestBirthday) {
+        nodeToEvict = buf;
+        earliestBirthday = buf.getTimeStamp();
+      }
+    }
+    if ((currentTimeMillis() - earliestBirthday > THRESHOLD_AGE_MILLISECONDS) && (nodeToEvict != null)) {
+      return evict(nodeToEvict);
+    }
+
+    // nothing can be evicted
+    return false;
+  }
+
+  private boolean evict(final ReadBuffer buf) {
+    freeList.push(buf.getBufferindex());
+    completedReadList.remove(buf);
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("Evicting buffer idx {}; was used for file {} offset {} length {}",
+          buf.getBufferindex(), buf.getStream().getPath(), buf.getOffset(), buf.getLength());
+    }
+    return true;
+  }
+
+  private boolean isAlreadyQueued(final SeaweedInputStream stream, final long requestedOffset) {
+    // returns true if any part of the buffer is already queued
+    return (isInList(readAheadQueue, stream, requestedOffset)
+        || isInList(inProgressList, stream, requestedOffset)
+        || isInList(completedReadList, stream, requestedOffset));
+  }
+
+  private boolean isInList(final Collection<ReadBuffer> list, final SeaweedInputStream stream, final long requestedOffset) {
+    return (getFromList(list, stream, requestedOffset) != null);
+  }
+
+  private ReadBuffer getFromList(final Collection<ReadBuffer> list, final SeaweedInputStream stream, final long requestedOffset) {
+    for (ReadBuffer buffer : list) {
+      if (buffer.getStream() == stream) {
+        if (buffer.getStatus() == ReadBufferStatus.AVAILABLE
+            && requestedOffset >= buffer.getOffset()
+            && requestedOffset < buffer.getOffset() + buffer.getLength()) {
+          return buffer;
+        } else if (requestedOffset >= buffer.getOffset()
+            && requestedOffset < buffer.getOffset() + buffer.getRequestedLength()) {
+          return buffer;
+        }
+      }
+    }
+    return null;
+  }
+
+  private void clearFromReadAheadQueue(final SeaweedInputStream stream, final long requestedOffset) {
+    ReadBuffer buffer = getFromList(readAheadQueue, stream, requestedOffset);
+    if (buffer != null) {
+      readAheadQueue.remove(buffer);
+      notifyAll();   // lock is held in calling method
+      freeList.push(buffer.getBufferindex());
+    }
+  }
+
+  private int getBlockFromCompletedQueue(final SeaweedInputStream stream, final long position, final int length,
+                                         final byte[] buffer) {
+    ReadBuffer buf = getFromList(completedReadList, stream, position);
+    if (buf == null || position >= buf.getOffset() + buf.getLength()) {
+      return 0;
+    }
+    int cursor = (int) (position - buf.getOffset());
+    int availableLengthInBuffer = buf.getLength() - cursor;
+    int lengthToCopy = Math.min(length, availableLengthInBuffer);
+    System.arraycopy(buf.getBuffer(), cursor, buffer, 0, lengthToCopy);
+    if (cursor == 0) {
+      buf.setFirstByteConsumed(true);
+    }
+    if (cursor + lengthToCopy == buf.getLength()) {
+      buf.setLastByteConsumed(true);
+    }
+    buf.setAnyByteConsumed(true);
+    return lengthToCopy;
+  }
+
+  /*
+   *
+   *  ReadBufferWorker-thread-facing methods
+   *
+   */
+
+  /**
+   * ReadBufferWorker thread calls this to get the next buffer that it should work on.
+   *
+   * @return {@link ReadBuffer}
+   * @throws InterruptedException if thread is interrupted
+   */
+  ReadBuffer getNextBlockToRead() throws InterruptedException {
+    ReadBuffer buffer = null;
+    synchronized (this) {
+      //buffer = readAheadQueue.take();  // blocking method
+      while (readAheadQueue.size() == 0) {
+        wait();
+      }
+      buffer = readAheadQueue.remove();
+      notifyAll();
+      if (buffer == null) {
+        return null;            // should never happen
+      }
+      buffer.setStatus(ReadBufferStatus.READING_IN_PROGRESS);
+      inProgressList.add(buffer);
+    }
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("ReadBufferWorker picked file {} for offset {}",
+          buffer.getStream().getPath(), buffer.getOffset());
+    }
+    return buffer;
+  }
+
+  /**
+   * ReadBufferWorker thread calls this method to post completion.
+   *
+   * @param buffer            the buffer whose read was completed
+   * @param result            the {@link ReadBufferStatus} after the read operation in the worker thread
+   * @param bytesActuallyRead the number of bytes that the worker thread was actually able to read
+   */
+  void doneReading(final ReadBuffer buffer, final ReadBufferStatus result, final int bytesActuallyRead) {
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("ReadBufferWorker completed file {} for offset {} bytes {}",
+          buffer.getStream().getPath(),  buffer.getOffset(), bytesActuallyRead);
+    }
+    synchronized (this) {
+      inProgressList.remove(buffer);
+      if (result == ReadBufferStatus.AVAILABLE && bytesActuallyRead > 0) {
+        buffer.setStatus(ReadBufferStatus.AVAILABLE);
+        buffer.setTimeStamp(currentTimeMillis());
+        buffer.setLength(bytesActuallyRead);
+        completedReadList.add(buffer);
+      } else {
+        freeList.push(buffer.getBufferindex());
+        // buffer should go out of scope after the end of the calling method in ReadBufferWorker, and eligible for GC
+      }
+    }
+    //outside the synchronized, since anyone receiving a wake-up from the latch must see safe-published results
+    buffer.getLatch().countDown(); // wake up waiting threads (if any)
+  }
+
+  /**
+   * Similar to System.currentTimeMillis, except implemented with System.nanoTime().
+   * System.currentTimeMillis can go backwards when system clock is changed (e.g., with NTP time synchronization),
+   * making it unsuitable for measuring time intervals. nanotime is strictly monotonically increasing per CPU core.
+   * Note: it is not monotonic across Sockets, and even within a CPU, its only the
+   * more recent parts which share a clock across all cores.
+   *
+   * @return current time in milliseconds
+   */
+  private long currentTimeMillis() {
+    return System.nanoTime() / 1000 / 1000;
+  }
+}

other/java/hdfs2/src/main/java/seaweed/hdfs/ReadBufferStatus.java

@@ -0,0 +1,29 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package seaweed.hdfs;
+
+/**
+ * The ReadBufferStatus for Rest AbfsClient
+ */
+public enum ReadBufferStatus {
+  NOT_AVAILABLE,  // buffers sitting in readaheadqueue have this stats
+  READING_IN_PROGRESS,  // reading is in progress on this buffer. Buffer should be in inProgressList
+  AVAILABLE,  // data is available in buffer. It should be in completedList
+  READ_FAILED  // read completed, but failed.
+}

other/java/hdfs2/src/main/java/seaweed/hdfs/ReadBufferWorker.java

@@ -0,0 +1,70 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package seaweed.hdfs;
+
+import java.util.concurrent.CountDownLatch;
+
+class ReadBufferWorker implements Runnable {
+
+  protected static final CountDownLatch UNLEASH_WORKERS = new CountDownLatch(1);
+  private int id;
+
+  ReadBufferWorker(final int id) {
+    this.id = id;
+  }
+
+  /**
+   * return the ID of ReadBufferWorker.
+   */
+  public int getId() {
+    return this.id;
+  }
+
+  /**
+   * Waits until a buffer becomes available in ReadAheadQueue.
+   * Once a buffer becomes available, reads the file specified in it and then posts results back to buffer manager.
+   * Rinse and repeat. Forever.
+   */
+  public void run() {
+    try {
+      UNLEASH_WORKERS.await();
+    } catch (InterruptedException ex) {
+      Thread.currentThread().interrupt();
+    }
+    ReadBufferManager bufferManager = ReadBufferManager.getBufferManager();
+    ReadBuffer buffer;
+    while (true) {
+      try {
+        buffer = bufferManager.getNextBlockToRead();   // blocks, until a buffer is available for this thread
+      } catch (InterruptedException ex) {
+        Thread.currentThread().interrupt();
+        return;
+      }
+      if (buffer != null) {
+        try {
+          // do the actual read, from the file.
+          int bytesRead = buffer.getStream().readRemote(buffer.getOffset(), buffer.getBuffer(), 0, buffer.getRequestedLength());
+          bufferManager.doneReading(buffer, ReadBufferStatus.AVAILABLE, bytesRead);  // post result back to ReadBufferManager
+        } catch (Exception ex) {
+          bufferManager.doneReading(buffer, ReadBufferStatus.READ_FAILED, 0);
+        }
+      }
+    }
+  }
+}

other/java/hdfs2/src/main/java/seaweed/hdfs/SeaweedFileSystem.java

@@ -0,0 +1,620 @@
+package seaweed.hdfs;
+
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.*;
+import org.apache.hadoop.fs.permission.AclEntry;
+import org.apache.hadoop.fs.permission.AclStatus;
+import org.apache.hadoop.fs.permission.FsPermission;
+import org.apache.hadoop.security.AccessControlException;
+import org.apache.hadoop.security.UserGroupInformation;
+import org.apache.hadoop.util.Progressable;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.io.FileNotFoundException;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.OutputStream;
+import java.net.URI;
+import java.util.EnumSet;
+import java.util.List;
+import java.util.Map;
+
+import static org.apache.hadoop.fs.CommonConfigurationKeysPublic.IO_FILE_BUFFER_SIZE_KEY;
+
+public class SeaweedFileSystem extends org.apache.hadoop.fs.FileSystem {
+
+    public static final int FS_SEAWEED_DEFAULT_PORT = 8888;
+    public static final String FS_SEAWEED_FILER_HOST = "fs.seaweed.filer.host";
+    public static final String FS_SEAWEED_FILER_PORT = "fs.seaweed.filer.port";
+
+    private static final Logger LOG = LoggerFactory.getLogger(SeaweedFileSystem.class);
+    private static int BUFFER_SIZE = 16 * 1024 * 1024;
+
+    private URI uri;
+    private Path workingDirectory = new Path("/");
+    private SeaweedFileSystemStore seaweedFileSystemStore;
+
+    public URI getUri() {
+        return uri;
+    }
+
+    public String getScheme() {
+        return "seaweedfs";
+    }
+
+    @Override
+    public void initialize(URI uri, Configuration conf) throws IOException { // get
+        super.initialize(uri, conf);
+
+        // get host information from uri (overrides info in conf)
+        String host = uri.getHost();
+        host = (host == null) ? conf.get(FS_SEAWEED_FILER_HOST, "localhost") : host;
+        if (host == null) {
+            throw new IOException("Invalid host specified");
+        }
+        conf.set(FS_SEAWEED_FILER_HOST, host);
+
+        // get port information from uri, (overrides info in conf)
+        int port = uri.getPort();
+        port = (port == -1) ? FS_SEAWEED_DEFAULT_PORT : port;
+        conf.setInt(FS_SEAWEED_FILER_PORT, port);
+
+        conf.setInt(IO_FILE_BUFFER_SIZE_KEY, BUFFER_SIZE);
+
+        setConf(conf);
+        this.uri = uri;
+
+        seaweedFileSystemStore = new SeaweedFileSystemStore(host, port);
+
+    }
+
+    @Override
+    public FSDataInputStream open(Path path, int bufferSize) throws IOException {
+
+        LOG.debug("open path: {} bufferSize:{}", path, bufferSize);
+
+        path = qualify(path);
+
+        try {
+            InputStream inputStream = seaweedFileSystemStore.openFileForRead(path, statistics, bufferSize);
+            return new FSDataInputStream(inputStream);
+        } catch (Exception ex) {
+            LOG.warn("open path: {} bufferSize:{}", path, bufferSize, ex);
+            return null;
+        }
+    }
+
+    @Override
+    public FSDataOutputStream create(Path path, FsPermission permission, final boolean overwrite, final int bufferSize,
+                                     final short replication, final long blockSize, final Progressable progress) throws IOException {
+
+        LOG.debug("create path: {} bufferSize:{} blockSize:{}", path, bufferSize, blockSize);
+
+        path = qualify(path);
+
+        try {
+            String replicaPlacement = String.format("%03d", replication - 1);
+            OutputStream outputStream = seaweedFileSystemStore.createFile(path, overwrite, permission, bufferSize, replicaPlacement);
+            return new FSDataOutputStream(outputStream, statistics);
+        } catch (Exception ex) {
+            LOG.warn("create path: {} bufferSize:{} blockSize:{}", path, bufferSize, blockSize, ex);
+            return null;
+        }
+    }
+
+    /**
+     * {@inheritDoc}
+     * @throws FileNotFoundException if the parent directory is not present -or
+     * is not a directory.
+     */
+    @Override
+    public FSDataOutputStream createNonRecursive(Path path,
+                                                 FsPermission permission,
+                                                 EnumSet<CreateFlag> flags,
+                                                 int bufferSize,
+                                                 short replication,
+                                                 long blockSize,
+                                                 Progressable progress) throws IOException {
+        Path parent = path.getParent();
+        if (parent != null) {
+            // expect this to raise an exception if there is no parent
+            if (!getFileStatus(parent).isDirectory()) {
+                throw new FileAlreadyExistsException("Not a directory: " + parent);
+            }
+        }
+        return create(path, permission,
+                flags.contains(CreateFlag.OVERWRITE), bufferSize,
+                replication, blockSize, progress);
+    }
+
+    @Override
+    public FSDataOutputStream append(Path path, int bufferSize, Progressable progressable) throws IOException {
+
+        LOG.debug("append path: {} bufferSize:{}", path, bufferSize);
+
+        path = qualify(path);
+        try {
+            OutputStream outputStream = seaweedFileSystemStore.createFile(path, false, null, bufferSize, "");
+            return new FSDataOutputStream(outputStream, statistics);
+        } catch (Exception ex) {
+            LOG.warn("append path: {} bufferSize:{}", path, bufferSize, ex);
+            return null;
+        }
+    }
+
+    @Override
+    public boolean rename(Path src, Path dst) {
+
+        LOG.debug("rename path: {} => {}", src, dst);
+
+        if (src.isRoot()) {
+            return false;
+        }
+
+        if (src.equals(dst)) {
+            return true;
+        }
+        FileStatus dstFileStatus = getFileStatus(dst);
+
+        String sourceFileName = src.getName();
+        Path adjustedDst = dst;
+
+        if (dstFileStatus != null) {
+            if (!dstFileStatus.isDirectory()) {
+                return false;
+            }
+            adjustedDst = new Path(dst, sourceFileName);
+        }
+
+        Path qualifiedSrcPath = qualify(src);
+        Path qualifiedDstPath = qualify(adjustedDst);
+
+        seaweedFileSystemStore.rename(qualifiedSrcPath, qualifiedDstPath);
+        return true;
+    }
+
+    @Override
+    public boolean delete(Path path, boolean recursive) {
+
+        LOG.debug("delete path: {} recursive:{}", path, recursive);
+
+        path = qualify(path);
+
+        FileStatus fileStatus = getFileStatus(path);
+
+        if (fileStatus == null) {
+            return true;
+        }
+
+        return seaweedFileSystemStore.deleteEntries(path, fileStatus.isDirectory(), recursive);
+
+    }
+
+    @Override
+    public FileStatus[] listStatus(Path path) throws IOException {
+
+        LOG.debug("listStatus path: {}", path);
+
+        path = qualify(path);
+
+        return seaweedFileSystemStore.listEntries(path);
+    }
+
+    @Override
+    public Path getWorkingDirectory() {
+        return workingDirectory;
+    }
+
+    @Override
+    public void setWorkingDirectory(Path path) {
+        if (path.isAbsolute()) {
+            workingDirectory = path;
+        } else {
+            workingDirectory = new Path(workingDirectory, path);
+        }
+    }
+
+    @Override
+    public boolean mkdirs(Path path, FsPermission fsPermission) throws IOException {
+
+        LOG.debug("mkdirs path: {}", path);
+
+        path = qualify(path);
+
+        FileStatus fileStatus = getFileStatus(path);
+
+        if (fileStatus == null) {
+
+            UserGroupInformation currentUser = UserGroupInformation.getCurrentUser();
+            return seaweedFileSystemStore.createDirectory(path, currentUser,
+                    fsPermission == null ? FsPermission.getDirDefault() : fsPermission,
+                    FsPermission.getUMask(getConf()));
+
+        }
+
+        if (fileStatus.isDirectory()) {
+            return true;
+        } else {
+            throw new FileAlreadyExistsException("Path is a file: " + path);
+        }
+    }
+
+    @Override
+    public FileStatus getFileStatus(Path path) {
+
+        LOG.debug("getFileStatus path: {}", path);
+
+        path = qualify(path);
+
+        return seaweedFileSystemStore.getFileStatus(path);
+    }
+
+    /**
+     * Set owner of a path (i.e. a file or a directory).
+     * The parameters owner and group cannot both be null.
+     *
+     * @param path  The path
+     * @param owner If it is null, the original username remains unchanged.
+     * @param group If it is null, the original groupname remains unchanged.
+     */
+    @Override
+    public void setOwner(Path path, final String owner, final String group)
+            throws IOException {
+        LOG.debug("setOwner path: {}", path);
+        path = qualify(path);
+
+        seaweedFileSystemStore.setOwner(path, owner, group);
+    }
+
+
+    /**
+     * Set permission of a path.
+     *
+     * @param path       The path
+     * @param permission Access permission
+     */
+    @Override
+    public void setPermission(Path path, final FsPermission permission) throws IOException {
+        LOG.debug("setPermission path: {}", path);
+
+        if (permission == null) {
+            throw new IllegalArgumentException("The permission can't be null");
+        }
+
+        path = qualify(path);
+
+        seaweedFileSystemStore.setPermission(path, permission);
+    }
+
+    Path qualify(Path path) {
+        return path.makeQualified(uri, workingDirectory);
+    }
+
+    /**
+     * Concat existing files together.
+     *
+     * @param trg   the path to the target destination.
+     * @param psrcs the paths to the sources to use for the concatenation.
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default).
+     */
+    @Override
+    public void concat(final Path trg, final Path[] psrcs) throws IOException {
+        throw new UnsupportedOperationException("Not implemented by the " +
+                getClass().getSimpleName() + " FileSystem implementation");
+    }
+
+    /**
+     * Truncate the file in the indicated path to the indicated size.
+     * <ul>
+     * <li>Fails if path is a directory.</li>
+     * <li>Fails if path does not exist.</li>
+     * <li>Fails if path is not closed.</li>
+     * <li>Fails if new size is greater than current size.</li>
+     * </ul>
+     *
+     * @param f         The path to the file to be truncated
+     * @param newLength The size the file is to be truncated to
+     * @return <code>true</code> if the file has been truncated to the desired
+     * <code>newLength</code> and is immediately available to be reused for
+     * write operations such as <code>append</code>, or
+     * <code>false</code> if a background process of adjusting the length of
+     * the last block has been started, and clients should wait for it to
+     * complete before proceeding with further file updates.
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default).
+     */
+    @Override
+    public boolean truncate(Path f, long newLength) throws IOException {
+        throw new UnsupportedOperationException("Not implemented by the " +
+                getClass().getSimpleName() + " FileSystem implementation");
+    }
+
+    @Override
+    public void createSymlink(final Path target, final Path link,
+                              final boolean createParent) throws AccessControlException,
+            FileAlreadyExistsException, FileNotFoundException,
+            ParentNotDirectoryException, UnsupportedFileSystemException,
+            IOException {
+        // Supporting filesystems should override this method
+        throw new UnsupportedOperationException(
+                "Filesystem does not support symlinks!");
+    }
+
+    public boolean supportsSymlinks() {
+        return false;
+    }
+
+    /**
+     * Create a snapshot.
+     *
+     * @param path         The directory where snapshots will be taken.
+     * @param snapshotName The name of the snapshot
+     * @return the snapshot path.
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     */
+    @Override
+    public Path createSnapshot(Path path, String snapshotName)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support createSnapshot");
+    }
+
+    /**
+     * Rename a snapshot.
+     *
+     * @param path            The directory path where the snapshot was taken
+     * @param snapshotOldName Old name of the snapshot
+     * @param snapshotNewName New name of the snapshot
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void renameSnapshot(Path path, String snapshotOldName,
+                               String snapshotNewName) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support renameSnapshot");
+    }
+
+    /**
+     * Delete a snapshot of a directory.
+     *
+     * @param path         The directory that the to-be-deleted snapshot belongs to
+     * @param snapshotName The name of the snapshot
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void deleteSnapshot(Path path, String snapshotName)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support deleteSnapshot");
+    }
+
+    /**
+     * Modifies ACL entries of files and directories.  This method can add new ACL
+     * entries or modify the permissions on existing ACL entries.  All existing
+     * ACL entries that are not specified in this call are retained without
+     * changes.  (Modifications are merged into the current ACL.)
+     *
+     * @param path    Path to modify
+     * @param aclSpec List&lt;AclEntry&gt; describing modifications
+     * @throws IOException                   if an ACL could not be modified
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void modifyAclEntries(Path path, List<AclEntry> aclSpec)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support modifyAclEntries");
+    }
+
+    /**
+     * Removes ACL entries from files and directories.  Other ACL entries are
+     * retained.
+     *
+     * @param path    Path to modify
+     * @param aclSpec List describing entries to remove
+     * @throws IOException                   if an ACL could not be modified
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void removeAclEntries(Path path, List<AclEntry> aclSpec)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support removeAclEntries");
+    }
+
+    /**
+     * Removes all default ACL entries from files and directories.
+     *
+     * @param path Path to modify
+     * @throws IOException                   if an ACL could not be modified
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void removeDefaultAcl(Path path)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support removeDefaultAcl");
+    }
+
+    /**
+     * Removes all but the base ACL entries of files and directories.  The entries
+     * for user, group, and others are retained for compatibility with permission
+     * bits.
+     *
+     * @param path Path to modify
+     * @throws IOException                   if an ACL could not be removed
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void removeAcl(Path path)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support removeAcl");
+    }
+
+    /**
+     * Fully replaces ACL of files and directories, discarding all existing
+     * entries.
+     *
+     * @param path    Path to modify
+     * @param aclSpec List describing modifications, which must include entries
+     *                for user, group, and others for compatibility with permission bits.
+     * @throws IOException                   if an ACL could not be modified
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void setAcl(Path path, List<AclEntry> aclSpec) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support setAcl");
+    }
+
+    /**
+     * Gets the ACL of a file or directory.
+     *
+     * @param path Path to get
+     * @return AclStatus describing the ACL of the file or directory
+     * @throws IOException                   if an ACL could not be read
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public AclStatus getAclStatus(Path path) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support getAclStatus");
+    }
+
+    /**
+     * Set an xattr of a file or directory.
+     * The name must be prefixed with the namespace followed by ".". For example,
+     * "user.attr".
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path  Path to modify
+     * @param name  xattr name.
+     * @param value xattr value.
+     * @param flag  xattr set flag
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void setXAttr(Path path, String name, byte[] value,
+                         EnumSet<XAttrSetFlag> flag) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support setXAttr");
+    }
+
+    /**
+     * Get an xattr name and value for a file or directory.
+     * The name must be prefixed with the namespace followed by ".". For example,
+     * "user.attr".
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path Path to get extended attribute
+     * @param name xattr name.
+     * @return byte[] xattr value.
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public byte[] getXAttr(Path path, String name) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support getXAttr");
+    }
+
+    /**
+     * Get all of the xattr name/value pairs for a file or directory.
+     * Only those xattrs which the logged-in user has permissions to view
+     * are returned.
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path Path to get extended attributes
+     * @return Map describing the XAttrs of the file or directory
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public Map<String, byte[]> getXAttrs(Path path) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support getXAttrs");
+    }
+
+    /**
+     * Get all of the xattrs name/value pairs for a file or directory.
+     * Only those xattrs which the logged-in user has permissions to view
+     * are returned.
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path  Path to get extended attributes
+     * @param names XAttr names.
+     * @return Map describing the XAttrs of the file or directory
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public Map<String, byte[]> getXAttrs(Path path, List<String> names)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support getXAttrs");
+    }
+
+    /**
+     * Get all of the xattr names for a file or directory.
+     * Only those xattr names which the logged-in user has permissions to view
+     * are returned.
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path Path to get extended attributes
+     * @return List{@literal <String>} of the XAttr names of the file or directory
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public List<String> listXAttrs(Path path) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support listXAttrs");
+    }
+
+    /**
+     * Remove an xattr of a file or directory.
+     * The name must be prefixed with the namespace followed by ".". For example,
+     * "user.attr".
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path Path to remove extended attribute
+     * @param name xattr name
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void removeXAttr(Path path, String name) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support removeXAttr");
+    }
+
+}

other/java/hdfs2/src/main/java/seaweed/hdfs/SeaweedFileSystemStore.java

@@ -0,0 +1,270 @@
+package seaweed.hdfs;
+
+import org.apache.hadoop.fs.FileStatus;
+import org.apache.hadoop.fs.FileSystem;
+import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.fs.permission.FsPermission;
+import org.apache.hadoop.security.UserGroupInformation;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+import seaweedfs.client.FilerClient;
+import seaweedfs.client.FilerGrpcClient;
+import seaweedfs.client.FilerProto;
+import seaweedfs.client.SeaweedRead;
+
+import java.io.FileNotFoundException;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.OutputStream;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.List;
+
+public class SeaweedFileSystemStore {
+
+    private static final Logger LOG = LoggerFactory.getLogger(SeaweedFileSystemStore.class);
+
+    private FilerGrpcClient filerGrpcClient;
+    private FilerClient filerClient;
+
+    public SeaweedFileSystemStore(String host, int port) {
+        int grpcPort = 10000 + port;
+        filerGrpcClient = new FilerGrpcClient(host, grpcPort);
+        filerClient = new FilerClient(filerGrpcClient);
+    }
+
+    public static String getParentDirectory(Path path) {
+        return path.isRoot() ? "/" : path.getParent().toUri().getPath();
+    }
+
+    static int permissionToMode(FsPermission permission, boolean isDirectory) {
+        int p = permission.toShort();
+        if (isDirectory) {
+            p = p | 1 << 31;
+        }
+        return p;
+    }
+
+    public boolean createDirectory(final Path path, UserGroupInformation currentUser,
+                                   final FsPermission permission, final FsPermission umask) {
+
+        LOG.debug("createDirectory path: {} permission: {} umask: {}",
+            path,
+            permission,
+            umask);
+
+        return filerClient.mkdirs(
+            path.toUri().getPath(),
+            permissionToMode(permission, true),
+            currentUser.getUserName(),
+            currentUser.getGroupNames()
+        );
+    }
+
+    public FileStatus[] listEntries(final Path path) {
+        LOG.debug("listEntries path: {}", path);
+
+        List<FileStatus> fileStatuses = new ArrayList<FileStatus>();
+
+        List<FilerProto.Entry> entries = filerClient.listEntries(path.toUri().getPath());
+
+        for (FilerProto.Entry entry : entries) {
+
+            FileStatus fileStatus = doGetFileStatus(new Path(path, entry.getName()), entry);
+
+            fileStatuses.add(fileStatus);
+        }
+        return fileStatuses.toArray(new FileStatus[0]);
+    }
+
+    public FileStatus getFileStatus(final Path path) {
+
+        FilerProto.Entry entry = lookupEntry(path);
+        if (entry == null) {
+            return null;
+        }
+        LOG.debug("doGetFileStatus path:{} entry:{}", path, entry);
+
+        FileStatus fileStatus = doGetFileStatus(path, entry);
+        return fileStatus;
+    }
+
+    public boolean deleteEntries(final Path path, boolean isDirectory, boolean recursive) {
+        LOG.debug("deleteEntries path: {} isDirectory {} recursive: {}",
+            path,
+            String.valueOf(isDirectory),
+            String.valueOf(recursive));
+
+        if (path.isRoot()) {
+            return true;
+        }
+
+        if (recursive && isDirectory) {
+            List<FilerProto.Entry> entries = filerClient.listEntries(path.toUri().getPath());
+            for (FilerProto.Entry entry : entries) {
+                deleteEntries(new Path(path, entry.getName()), entry.getIsDirectory(), true);
+            }
+        }
+
+        return filerClient.deleteEntry(getParentDirectory(path), path.getName(), true, recursive, true);
+    }
+
+    private FileStatus doGetFileStatus(Path path, FilerProto.Entry entry) {
+        FilerProto.FuseAttributes attributes = entry.getAttributes();
+        long length = SeaweedRead.totalSize(entry.getChunksList());
+        boolean isDir = entry.getIsDirectory();
+        int block_replication = 1;
+        int blocksize = 512;
+        long modification_time = attributes.getMtime() * 1000; // milliseconds
+        long access_time = 0;
+        FsPermission permission = FsPermission.createImmutable((short) attributes.getFileMode());
+        String owner = attributes.getUserName();
+        String group = attributes.getGroupNameCount() > 0 ? attributes.getGroupName(0) : "";
+        return new FileStatus(length, isDir, block_replication, blocksize,
+            modification_time, access_time, permission, owner, group, null, path);
+    }
+
+    private FilerProto.Entry lookupEntry(Path path) {
+
+        return filerClient.lookupEntry(getParentDirectory(path), path.getName());
+
+    }
+
+    public void rename(Path source, Path destination) {
+
+        LOG.debug("rename source: {} destination:{}", source, destination);
+
+        if (source.isRoot()) {
+            return;
+        }
+        LOG.info("rename source: {} destination:{}", source, destination);
+        FilerProto.Entry entry = lookupEntry(source);
+        if (entry == null) {
+            LOG.warn("rename non-existing source: {}", source);
+            return;
+        }
+        filerClient.mv(source.toUri().getPath(), destination.toUri().getPath());
+    }
+
+    public OutputStream createFile(final Path path,
+                                   final boolean overwrite,
+                                   FsPermission permission,
+                                   int bufferSize,
+                                   String replication) throws IOException {
+
+        permission = permission == null ? FsPermission.getFileDefault() : permission;
+
+        LOG.debug("createFile path: {} overwrite: {} permission: {}",
+            path,
+            overwrite,
+            permission.toString());
+
+        UserGroupInformation userGroupInformation = UserGroupInformation.getCurrentUser();
+        long now = System.currentTimeMillis() / 1000L;
+
+        FilerProto.Entry.Builder entry = null;
+        long writePosition = 0;
+        if (!overwrite) {
+            FilerProto.Entry existingEntry = lookupEntry(path);
+            LOG.debug("createFile merged entry path:{} existingEntry:{}", path, existingEntry);
+            if (existingEntry != null) {
+                entry = FilerProto.Entry.newBuilder();
+                entry.mergeFrom(existingEntry);
+                entry.getAttributesBuilder().setMtime(now);
+                LOG.debug("createFile merged entry path:{} entry:{} from:{}", path, entry, existingEntry);
+                writePosition = SeaweedRead.totalSize(existingEntry.getChunksList());
+                replication = existingEntry.getAttributes().getReplication();
+            }
+        }
+        if (entry == null) {
+            entry = FilerProto.Entry.newBuilder()
+                .setName(path.getName())
+                .setIsDirectory(false)
+                .setAttributes(FilerProto.FuseAttributes.newBuilder()
+                    .setFileMode(permissionToMode(permission, false))
+                    .setReplication(replication)
+                    .setCrtime(now)
+                    .setMtime(now)
+                    .setUserName(userGroupInformation.getUserName())
+                    .clearGroupName()
+                    .addAllGroupName(Arrays.asList(userGroupInformation.getGroupNames()))
+                );
+        }
+
+        return new SeaweedOutputStream(filerGrpcClient, path, entry, writePosition, bufferSize, replication);
+
+    }
+
+    public InputStream openFileForRead(final Path path, FileSystem.Statistics statistics,
+                                       int bufferSize) throws IOException {
+
+        LOG.debug("openFileForRead path:{} bufferSize:{}", path, bufferSize);
+
+        int readAheadQueueDepth = 2;
+        FilerProto.Entry entry = lookupEntry(path);
+
+        if (entry == null) {
+            throw new FileNotFoundException("read non-exist file " + path);
+        }
+
+        return new SeaweedInputStream(filerGrpcClient,
+            statistics,
+            path.toUri().getPath(),
+            entry,
+            bufferSize,
+            readAheadQueueDepth);
+    }
+
+    public void setOwner(Path path, String owner, String group) {
+
+        LOG.debug("setOwner path:{} owner:{} group:{}", path, owner, group);
+
+        FilerProto.Entry entry = lookupEntry(path);
+        if (entry == null) {
+            LOG.debug("setOwner path:{} entry:{}", path, entry);
+            return;
+        }
+
+        FilerProto.Entry.Builder entryBuilder = entry.toBuilder();
+        FilerProto.FuseAttributes.Builder attributesBuilder = entry.getAttributes().toBuilder();
+
+        if (owner != null) {
+            attributesBuilder.setUserName(owner);
+        }
+        if (group != null) {
+            attributesBuilder.clearGroupName();
+            attributesBuilder.addGroupName(group);
+        }
+
+        entryBuilder.setAttributes(attributesBuilder);
+
+        LOG.debug("setOwner path:{} entry:{}", path, entryBuilder);
+
+        filerClient.updateEntry(getParentDirectory(path), entryBuilder.build());
+
+    }
+
+    public void setPermission(Path path, FsPermission permission) {
+
+        LOG.debug("setPermission path:{} permission:{}", path, permission);
+
+        FilerProto.Entry entry = lookupEntry(path);
+        if (entry == null) {
+            LOG.debug("setPermission path:{} entry:{}", path, entry);
+            return;
+        }
+
+        FilerProto.Entry.Builder entryBuilder = entry.toBuilder();
+        FilerProto.FuseAttributes.Builder attributesBuilder = entry.getAttributes().toBuilder();
+
+        attributesBuilder.setFileMode(permissionToMode(permission, entry.getIsDirectory()));
+
+        entryBuilder.setAttributes(attributesBuilder);
+
+        LOG.debug("setPermission path:{} entry:{}", path, entryBuilder);
+
+        filerClient.updateEntry(getParentDirectory(path), entryBuilder.build());
+
+    }
+
+}

other/java/hdfs2/src/main/java/seaweed/hdfs/SeaweedInputStream.java

@@ -0,0 +1,371 @@
+package seaweed.hdfs;
+
+// based on org.apache.hadoop.fs.azurebfs.services.AbfsInputStream
+
+import com.google.common.base.Preconditions;
+import org.apache.hadoop.fs.FSExceptionMessages;
+import org.apache.hadoop.fs.FSInputStream;
+import org.apache.hadoop.fs.FileSystem.Statistics;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+import seaweedfs.client.FilerGrpcClient;
+import seaweedfs.client.FilerProto;
+import seaweedfs.client.SeaweedRead;
+
+import java.io.EOFException;
+import java.io.IOException;
+import java.util.List;
+
+public class SeaweedInputStream extends FSInputStream {
+
+    private static final Logger LOG = LoggerFactory.getLogger(SeaweedInputStream.class);
+
+    private final FilerGrpcClient filerGrpcClient;
+    private final Statistics statistics;
+    private final String path;
+    private final FilerProto.Entry entry;
+    private final List<SeaweedRead.VisibleInterval> visibleIntervalList;
+    private final long contentLength;
+    private final int bufferSize; // default buffer size
+    private final int readAheadQueueDepth;         // initialized in constructor
+    private final boolean readAheadEnabled; // whether enable readAhead;
+
+    private byte[] buffer = null;            // will be initialized on first use
+
+    private long fCursor = 0;  // cursor of buffer within file - offset of next byte to read from remote server
+    private long fCursorAfterLastRead = -1;
+    private int bCursor = 0;   // cursor of read within buffer - offset of next byte to be returned from buffer
+    private int limit = 0;     // offset of next byte to be read into buffer from service (i.e., upper marker+1
+    //                                                      of valid bytes in buffer)
+    private boolean closed = false;
+
+    public SeaweedInputStream(
+        final FilerGrpcClient filerGrpcClient,
+        final Statistics statistics,
+        final String path,
+        final FilerProto.Entry entry,
+        final int bufferSize,
+        final int readAheadQueueDepth) {
+        this.filerGrpcClient = filerGrpcClient;
+        this.statistics = statistics;
+        this.path = path;
+        this.entry = entry;
+        this.contentLength = SeaweedRead.totalSize(entry.getChunksList());
+        this.bufferSize = bufferSize;
+        this.readAheadQueueDepth = (readAheadQueueDepth >= 0) ? readAheadQueueDepth : Runtime.getRuntime().availableProcessors();
+        this.readAheadEnabled = true;
+
+        this.visibleIntervalList = SeaweedRead.nonOverlappingVisibleIntervals(entry.getChunksList());
+
+        LOG.debug("new path:{} entry:{} visibleIntervalList:{}", path, entry, visibleIntervalList);
+
+    }
+
+    public String getPath() {
+        return path;
+    }
+
+    @Override
+    public int read() throws IOException {
+        byte[] b = new byte[1];
+        int numberOfBytesRead = read(b, 0, 1);
+        if (numberOfBytesRead < 0) {
+            return -1;
+        } else {
+            return (b[0] & 0xFF);
+        }
+    }
+
+    @Override
+    public synchronized int read(final byte[] b, final int off, final int len) throws IOException {
+        int currentOff = off;
+        int currentLen = len;
+        int lastReadBytes;
+        int totalReadBytes = 0;
+        do {
+            lastReadBytes = readOneBlock(b, currentOff, currentLen);
+            if (lastReadBytes > 0) {
+                currentOff += lastReadBytes;
+                currentLen -= lastReadBytes;
+                totalReadBytes += lastReadBytes;
+            }
+            if (currentLen <= 0 || currentLen > b.length - currentOff) {
+                break;
+            }
+        } while (lastReadBytes > 0);
+        return totalReadBytes > 0 ? totalReadBytes : lastReadBytes;
+    }
+
+    private int readOneBlock(final byte[] b, final int off, final int len) throws IOException {
+        if (closed) {
+            throw new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+
+        Preconditions.checkNotNull(b);
+
+        if (len == 0) {
+            return 0;
+        }
+
+        if (this.available() == 0) {
+            return -1;
+        }
+
+        if (off < 0 || len < 0 || len > b.length - off) {
+            throw new IndexOutOfBoundsException();
+        }
+
+        //If buffer is empty, then fill the buffer.
+        if (bCursor == limit) {
+            //If EOF, then return -1
+            if (fCursor >= contentLength) {
+                return -1;
+            }
+
+            long bytesRead = 0;
+            //reset buffer to initial state - i.e., throw away existing data
+            bCursor = 0;
+            limit = 0;
+            if (buffer == null) {
+                buffer = new byte[bufferSize];
+            }
+
+            // Enable readAhead when reading sequentially
+            if (-1 == fCursorAfterLastRead || fCursorAfterLastRead == fCursor || b.length >= bufferSize) {
+                bytesRead = readInternal(fCursor, buffer, 0, bufferSize, false);
+            } else {
+                bytesRead = readInternal(fCursor, buffer, 0, b.length, true);
+            }
+
+            if (bytesRead == -1) {
+                return -1;
+            }
+
+            limit += bytesRead;
+            fCursor += bytesRead;
+            fCursorAfterLastRead = fCursor;
+        }
+
+        //If there is anything in the buffer, then return lesser of (requested bytes) and (bytes in buffer)
+        //(bytes returned may be less than requested)
+        int bytesRemaining = limit - bCursor;
+        int bytesToRead = Math.min(len, bytesRemaining);
+        System.arraycopy(buffer, bCursor, b, off, bytesToRead);
+        bCursor += bytesToRead;
+        if (statistics != null) {
+            statistics.incrementBytesRead(bytesToRead);
+        }
+        return bytesToRead;
+    }
+
+
+    private int readInternal(final long position, final byte[] b, final int offset, final int length,
+                             final boolean bypassReadAhead) throws IOException {
+        if (readAheadEnabled && !bypassReadAhead) {
+            // try reading from read-ahead
+            if (offset != 0) {
+                throw new IllegalArgumentException("readahead buffers cannot have non-zero buffer offsets");
+            }
+            int receivedBytes;
+
+            // queue read-aheads
+            int numReadAheads = this.readAheadQueueDepth;
+            long nextSize;
+            long nextOffset = position;
+            while (numReadAheads > 0 && nextOffset < contentLength) {
+                nextSize = Math.min((long) bufferSize, contentLength - nextOffset);
+                ReadBufferManager.getBufferManager().queueReadAhead(this, nextOffset, (int) nextSize);
+                nextOffset = nextOffset + nextSize;
+                numReadAheads--;
+            }
+
+            // try reading from buffers first
+            receivedBytes = ReadBufferManager.getBufferManager().getBlock(this, position, length, b);
+            if (receivedBytes > 0) {
+                return receivedBytes;
+            }
+
+            // got nothing from read-ahead, do our own read now
+            receivedBytes = readRemote(position, b, offset, length);
+            return receivedBytes;
+        } else {
+            return readRemote(position, b, offset, length);
+        }
+    }
+
+    int readRemote(long position, byte[] b, int offset, int length) throws IOException {
+        if (position < 0) {
+            throw new IllegalArgumentException("attempting to read from negative offset");
+        }
+        if (position >= contentLength) {
+            return -1;  // Hadoop prefers -1 to EOFException
+        }
+        if (b == null) {
+            throw new IllegalArgumentException("null byte array passed in to read() method");
+        }
+        if (offset >= b.length) {
+            throw new IllegalArgumentException("offset greater than length of array");
+        }
+        if (length < 0) {
+            throw new IllegalArgumentException("requested read length is less than zero");
+        }
+        if (length > (b.length - offset)) {
+            throw new IllegalArgumentException("requested read length is more than will fit after requested offset in buffer");
+        }
+
+        long bytesRead = SeaweedRead.read(filerGrpcClient, visibleIntervalList, position, b, offset, length);
+        if (bytesRead > Integer.MAX_VALUE) {
+            throw new IOException("Unexpected Content-Length");
+        }
+        return (int) bytesRead;
+    }
+
+    /**
+     * Seek to given position in stream.
+     *
+     * @param n position to seek to
+     * @throws IOException  if there is an error
+     * @throws EOFException if attempting to seek past end of file
+     */
+    @Override
+    public synchronized void seek(long n) throws IOException {
+        if (closed) {
+            throw new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+        if (n < 0) {
+            throw new EOFException(FSExceptionMessages.NEGATIVE_SEEK);
+        }
+        if (n > contentLength) {
+            throw new EOFException(FSExceptionMessages.CANNOT_SEEK_PAST_EOF);
+        }
+
+        if (n >= fCursor - limit && n <= fCursor) { // within buffer
+            bCursor = (int) (n - (fCursor - limit));
+            return;
+        }
+
+        // next read will read from here
+        fCursor = n;
+
+        //invalidate buffer
+        limit = 0;
+        bCursor = 0;
+    }
+
+    @Override
+    public synchronized long skip(long n) throws IOException {
+        if (closed) {
+            throw new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+        long currentPos = getPos();
+        if (currentPos == contentLength) {
+            if (n > 0) {
+                throw new EOFException(FSExceptionMessages.CANNOT_SEEK_PAST_EOF);
+            }
+        }
+        long newPos = currentPos + n;
+        if (newPos < 0) {
+            newPos = 0;
+            n = newPos - currentPos;
+        }
+        if (newPos > contentLength) {
+            newPos = contentLength;
+            n = newPos - currentPos;
+        }
+        seek(newPos);
+        return n;
+    }
+
+    /**
+     * Return the size of the remaining available bytes
+     * if the size is less than or equal to {@link Integer#MAX_VALUE},
+     * otherwise, return {@link Integer#MAX_VALUE}.
+     * <p>
+     * This is to match the behavior of DFSInputStream.available(),
+     * which some clients may rely on (HBase write-ahead log reading in
+     * particular).
+     */
+    @Override
+    public synchronized int available() throws IOException {
+        if (closed) {
+            throw new IOException(
+                FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+        final long remaining = this.contentLength - this.getPos();
+        return remaining <= Integer.MAX_VALUE
+            ? (int) remaining : Integer.MAX_VALUE;
+    }
+
+    /**
+     * Returns the length of the file that this stream refers to. Note that the length returned is the length
+     * as of the time the Stream was opened. Specifically, if there have been subsequent appends to the file,
+     * they wont be reflected in the returned length.
+     *
+     * @return length of the file.
+     * @throws IOException if the stream is closed
+     */
+    public long length() throws IOException {
+        if (closed) {
+            throw new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+        return contentLength;
+    }
+
+    /**
+     * Return the current offset from the start of the file
+     *
+     * @throws IOException throws {@link IOException} if there is an error
+     */
+    @Override
+    public synchronized long getPos() throws IOException {
+        if (closed) {
+            throw new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+        return fCursor - limit + bCursor;
+    }
+
+    /**
+     * Seeks a different copy of the data.  Returns true if
+     * found a new source, false otherwise.
+     *
+     * @throws IOException throws {@link IOException} if there is an error
+     */
+    @Override
+    public boolean seekToNewSource(long l) throws IOException {
+        return false;
+    }
+
+    @Override
+    public synchronized void close() throws IOException {
+        closed = true;
+        buffer = null; // de-reference the buffer so it can be GC'ed sooner
+    }
+
+    /**
+     * Not supported by this stream. Throws {@link UnsupportedOperationException}
+     *
+     * @param readlimit ignored
+     */
+    @Override
+    public synchronized void mark(int readlimit) {
+        throw new UnsupportedOperationException("mark()/reset() not supported on this stream");
+    }
+
+    /**
+     * Not supported by this stream. Throws {@link UnsupportedOperationException}
+     */
+    @Override
+    public synchronized void reset() throws IOException {
+        throw new UnsupportedOperationException("mark()/reset() not supported on this stream");
+    }
+
+    /**
+     * gets whether mark and reset are supported by {@code ADLFileInputStream}. Always returns false.
+     *
+     * @return always {@code false}
+     */
+    @Override
+    public boolean markSupported() {
+        return false;
+    }
+}

other/java/hdfs2/src/main/java/seaweed/hdfs/SeaweedOutputStream.java

@@ -0,0 +1,283 @@
+package seaweed.hdfs;
+
+// adapted from org.apache.hadoop.fs.azurebfs.services.AbfsOutputStream
+
+import com.google.common.base.Preconditions;
+import org.apache.hadoop.fs.FSExceptionMessages;
+import org.apache.hadoop.fs.Path;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+import seaweedfs.client.FilerGrpcClient;
+import seaweedfs.client.FilerProto;
+import seaweedfs.client.SeaweedWrite;
+
+import java.io.IOException;
+import java.io.InterruptedIOException;
+import java.io.OutputStream;
+import java.util.concurrent.*;
+
+import static seaweed.hdfs.SeaweedFileSystemStore.getParentDirectory;
+
+public class SeaweedOutputStream extends OutputStream {
+
+    private static final Logger LOG = LoggerFactory.getLogger(SeaweedOutputStream.class);
+
+    private final FilerGrpcClient filerGrpcClient;
+    private final Path path;
+    private final int bufferSize;
+    private final int maxConcurrentRequestCount;
+    private final ThreadPoolExecutor threadExecutor;
+    private final ExecutorCompletionService<Void> completionService;
+    private FilerProto.Entry.Builder entry;
+    private long position;
+    private boolean closed;
+    private boolean supportFlush = true;
+    private volatile IOException lastError;
+    private long lastFlushOffset;
+    private long lastTotalAppendOffset = 0;
+    private byte[] buffer;
+    private int bufferIndex;
+    private ConcurrentLinkedDeque<WriteOperation> writeOperations;
+    private String replication = "000";
+
+    public SeaweedOutputStream(FilerGrpcClient filerGrpcClient, final Path path, FilerProto.Entry.Builder entry,
+                               final long position, final int bufferSize, final String replication) {
+        this.filerGrpcClient = filerGrpcClient;
+        this.replication = replication;
+        this.path = path;
+        this.position = position;
+        this.closed = false;
+        this.lastError = null;
+        this.lastFlushOffset = 0;
+        this.bufferSize = bufferSize;
+        this.buffer = new byte[bufferSize];
+        this.bufferIndex = 0;
+        this.writeOperations = new ConcurrentLinkedDeque<>();
+
+        this.maxConcurrentRequestCount = 4 * Runtime.getRuntime().availableProcessors();
+
+        this.threadExecutor
+            = new ThreadPoolExecutor(maxConcurrentRequestCount,
+            maxConcurrentRequestCount,
+            10L,
+            TimeUnit.SECONDS,
+            new LinkedBlockingQueue<Runnable>());
+        this.completionService = new ExecutorCompletionService<>(this.threadExecutor);
+
+        this.entry = entry;
+
+    }
+
+    private synchronized void flushWrittenBytesToServiceInternal(final long offset) throws IOException {
+
+        LOG.debug("SeaweedWrite.writeMeta path: {} entry:{}", path, entry);
+
+        try {
+            SeaweedWrite.writeMeta(filerGrpcClient, getParentDirectory(path), entry);
+        } catch (Exception ex) {
+            throw new IOException(ex);
+        }
+        this.lastFlushOffset = offset;
+    }
+
+    @Override
+    public void write(final int byteVal) throws IOException {
+        write(new byte[]{(byte) (byteVal & 0xFF)});
+    }
+
+    @Override
+    public synchronized void write(final byte[] data, final int off, final int length)
+        throws IOException {
+        maybeThrowLastError();
+
+        Preconditions.checkArgument(data != null, "null data");
+
+        if (off < 0 || length < 0 || length > data.length - off) {
+            throw new IndexOutOfBoundsException();
+        }
+
+        int currentOffset = off;
+        int writableBytes = bufferSize - bufferIndex;
+        int numberOfBytesToWrite = length;
+
+        while (numberOfBytesToWrite > 0) {
+            if (writableBytes <= numberOfBytesToWrite) {
+                System.arraycopy(data, currentOffset, buffer, bufferIndex, writableBytes);
+                bufferIndex += writableBytes;
+                writeCurrentBufferToService();
+                currentOffset += writableBytes;
+                numberOfBytesToWrite = numberOfBytesToWrite - writableBytes;
+            } else {
+                System.arraycopy(data, currentOffset, buffer, bufferIndex, numberOfBytesToWrite);
+                bufferIndex += numberOfBytesToWrite;
+                numberOfBytesToWrite = 0;
+            }
+
+            writableBytes = bufferSize - bufferIndex;
+        }
+    }
+
+    /**
+     * Flushes this output stream and forces any buffered output bytes to be
+     * written out. If any data remains in the payload it is committed to the
+     * service. Data is queued for writing and forced out to the service
+     * before the call returns.
+     */
+    @Override
+    public void flush() throws IOException {
+        if (supportFlush) {
+            flushInternalAsync();
+        }
+    }
+
+    /**
+     * Force all data in the output stream to be written to Azure storage.
+     * Wait to return until this is complete. Close the access to the stream and
+     * shutdown the upload thread pool.
+     * If the blob was created, its lease will be released.
+     * Any error encountered caught in threads and stored will be rethrown here
+     * after cleanup.
+     */
+    @Override
+    public synchronized void close() throws IOException {
+        if (closed) {
+            return;
+        }
+
+        LOG.debug("close path: {}", path);
+        try {
+            flushInternal();
+            threadExecutor.shutdown();
+        } finally {
+            lastError = new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+            buffer = null;
+            bufferIndex = 0;
+            closed = true;
+            writeOperations.clear();
+            if (!threadExecutor.isShutdown()) {
+                threadExecutor.shutdownNow();
+            }
+        }
+    }
+
+    private synchronized void writeCurrentBufferToService() throws IOException {
+        if (bufferIndex == 0) {
+            return;
+        }
+
+        final byte[] bytes = buffer;
+        final int bytesLength = bufferIndex;
+
+        buffer = new byte[bufferSize];
+        bufferIndex = 0;
+        final long offset = position;
+        position += bytesLength;
+
+        if (threadExecutor.getQueue().size() >= maxConcurrentRequestCount * 2) {
+            waitForTaskToComplete();
+        }
+
+        final Future<Void> job = completionService.submit(new Callable<Void>() {
+            @Override
+            public Void call() throws Exception {
+                // originally: client.append(path, offset, bytes, 0, bytesLength);
+                SeaweedWrite.writeData(entry, replication, filerGrpcClient, offset, bytes, 0, bytesLength);
+                return null;
+            }
+        });
+
+        writeOperations.add(new WriteOperation(job, offset, bytesLength));
+
+        // Try to shrink the queue
+        shrinkWriteOperationQueue();
+    }
+
+    private void waitForTaskToComplete() throws IOException {
+        boolean completed;
+        for (completed = false; completionService.poll() != null; completed = true) {
+            // keep polling until there is no data
+        }
+
+        if (!completed) {
+            try {
+                completionService.take();
+            } catch (InterruptedException e) {
+                lastError = (IOException) new InterruptedIOException(e.toString()).initCause(e);
+                throw lastError;
+            }
+        }
+    }
+
+    private void maybeThrowLastError() throws IOException {
+        if (lastError != null) {
+            throw lastError;
+        }
+    }
+
+    /**
+     * Try to remove the completed write operations from the beginning of write
+     * operation FIFO queue.
+     */
+    private synchronized void shrinkWriteOperationQueue() throws IOException {
+        try {
+            while (writeOperations.peek() != null && writeOperations.peek().task.isDone()) {
+                writeOperations.peek().task.get();
+                lastTotalAppendOffset += writeOperations.peek().length;
+                writeOperations.remove();
+            }
+        } catch (Exception e) {
+            lastError = new IOException(e);
+            throw lastError;
+        }
+    }
+
+    private synchronized void flushInternal() throws IOException {
+        maybeThrowLastError();
+        writeCurrentBufferToService();
+        flushWrittenBytesToService();
+    }
+
+    private synchronized void flushInternalAsync() throws IOException {
+        maybeThrowLastError();
+        writeCurrentBufferToService();
+        flushWrittenBytesToServiceAsync();
+    }
+
+    private synchronized void flushWrittenBytesToService() throws IOException {
+        for (WriteOperation writeOperation : writeOperations) {
+            try {
+                writeOperation.task.get();
+            } catch (Exception ex) {
+                lastError = new IOException(ex);
+                throw lastError;
+            }
+        }
+        LOG.debug("flushWrittenBytesToService: {} position:{}", path, position);
+        flushWrittenBytesToServiceInternal(position);
+    }
+
+    private synchronized void flushWrittenBytesToServiceAsync() throws IOException {
+        shrinkWriteOperationQueue();
+
+        if (this.lastTotalAppendOffset > this.lastFlushOffset) {
+            this.flushWrittenBytesToServiceInternal(this.lastTotalAppendOffset);
+        }
+    }
+
+    private static class WriteOperation {
+        private final Future<Void> task;
+        private final long startOffset;
+        private final long length;
+
+        WriteOperation(final Future<Void> task, final long startOffset, final long length) {
+            Preconditions.checkNotNull(task, "task");
+            Preconditions.checkArgument(startOffset >= 0, "startOffset");
+            Preconditions.checkArgument(length >= 0, "length");
+
+            this.task = task;
+            this.startOffset = startOffset;
+            this.length = length;
+        }
+    }
+
+}

other/java/hdfs3/dependency-reduced-pom.xml

@@ -0,0 +1,133 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
+  <parent>
+    <artifactId>oss-parent</artifactId>
+    <groupId>org.sonatype.oss</groupId>
+    <version>9</version>
+    <relativePath>../pom.xml/pom.xml</relativePath>
+  </parent>
+  <modelVersion>4.0.0</modelVersion>
+  <groupId>com.github.chrislusf</groupId>
+  <artifactId>seaweedfs-hadoop3-client</artifactId>
+  <version>${seaweedfs.client.version}</version>
+  <build>
+    <plugins>
+      <plugin>
+        <artifactId>maven-compiler-plugin</artifactId>
+        <configuration>
+          <source>7</source>
+          <target>7</target>
+        </configuration>
+      </plugin>
+      <plugin>
+        <artifactId>maven-shade-plugin</artifactId>
+        <version>3.2.1</version>
+        <executions>
+          <execution>
+            <phase>package</phase>
+            <goals>
+              <goal>shade</goal>
+            </goals>
+            <configuration>
+              <filters>
+                <filter>
+                  <artifact>*:*</artifact>
+                  <excludes>
+                    <exclude>META-INF/*.SF</exclude>
+                    <exclude>META-INF/*.DSA</exclude>
+                    <exclude>META-INF/*.RSA</exclude>
+                    <exclude>org/slf4j/**</exclude>
+                    <exclude>META-INF/maven/org.slf4j/**</exclude>
+                  </excludes>
+                </filter>
+              </filters>
+              <transformers>
+                <transformer />
+              </transformers>
+              <relocations>
+                <relocation>
+                  <pattern>com.google</pattern>
+                  <shadedPattern>shaded.com.google</shadedPattern>
+                </relocation>
+                <relocation>
+                  <pattern>io.grpc.internal</pattern>
+                  <shadedPattern>shaded.io.grpc.internal</shadedPattern>
+                </relocation>
+                <relocation>
+                  <pattern>org.apache.commons</pattern>
+                  <shadedPattern>shaded.org.apache.commons</shadedPattern>
+                  <excludes>
+                    <exclude>org.apache.hadoop</exclude>
+                    <exclude>org.apache.log4j</exclude>
+                  </excludes>
+                </relocation>
+                <relocation>
+                  <pattern>org.apache.http</pattern>
+                  <shadedPattern>shaded.org.apache.http</shadedPattern>
+                </relocation>
+              </relocations>
+            </configuration>
+          </execution>
+        </executions>
+      </plugin>
+      <plugin>
+        <artifactId>maven-gpg-plugin</artifactId>
+        <version>1.5</version>
+        <executions>
+          <execution>
+            <id>sign-artifacts</id>
+            <phase>verify</phase>
+            <goals>
+              <goal>sign</goal>
+            </goals>
+          </execution>
+        </executions>
+      </plugin>
+      <plugin>
+        <groupId>org.sonatype.plugins</groupId>
+        <artifactId>nexus-staging-maven-plugin</artifactId>
+        <version>1.6.7</version>
+        <extensions>true</extensions>
+        <configuration>
+          <serverId>ossrh</serverId>
+          <nexusUrl>https://oss.sonatype.org/</nexusUrl>
+          <autoReleaseAfterClose>true</autoReleaseAfterClose>
+        </configuration>
+      </plugin>
+      <plugin>
+        <artifactId>maven-source-plugin</artifactId>
+        <version>2.2.1</version>
+        <executions>
+          <execution>
+            <id>attach-sources</id>
+            <goals>
+              <goal>jar-no-fork</goal>
+            </goals>
+          </execution>
+        </executions>
+      </plugin>
+      <plugin>
+        <artifactId>maven-javadoc-plugin</artifactId>
+        <version>2.9.1</version>
+        <executions>
+          <execution>
+            <id>attach-javadocs</id>
+            <goals>
+              <goal>jar</goal>
+            </goals>
+          </execution>
+        </executions>
+      </plugin>
+    </plugins>
+  </build>
+  <distributionManagement>
+    <snapshotRepository>
+      <id>ossrh</id>
+      <url>https://oss.sonatype.org/content/repositories/snapshots</url>
+    </snapshotRepository>
+  </distributionManagement>
+  <properties>
+    <seaweedfs.client.version>1.2.4</seaweedfs.client.version>
+    <hadoop.version>3.1.1</hadoop.version>
+  </properties>
+</project>

other/java/hdfs3/pom.xml

@@ -0,0 +1,163 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project xmlns="http://maven.apache.org/POM/4.0.0"
+         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
+    <modelVersion>4.0.0</modelVersion>
+
+    <properties>
+        <seaweedfs.client.version>1.2.4</seaweedfs.client.version>
+        <hadoop.version>3.1.1</hadoop.version>
+    </properties>
+
+    <groupId>com.github.chrislusf</groupId>
+    <artifactId>seaweedfs-hadoop3-client</artifactId>
+    <version>${seaweedfs.client.version}</version>
+
+    <parent>
+        <groupId>org.sonatype.oss</groupId>
+        <artifactId>oss-parent</artifactId>
+        <version>9</version>
+    </parent>
+
+    <distributionManagement>
+        <snapshotRepository>
+            <id>ossrh</id>
+            <url>https://oss.sonatype.org/content/repositories/snapshots</url>
+        </snapshotRepository>
+    </distributionManagement>
+    <build>
+        <plugins>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-compiler-plugin</artifactId>
+                <configuration>
+                    <source>7</source>
+                    <target>7</target>
+                </configuration>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-shade-plugin</artifactId>
+                <version>3.2.1</version>
+                <executions>
+                    <execution>
+                        <phase>package</phase>
+                        <goals>
+                            <goal>shade</goal>
+                        </goals>
+                        <configuration>
+                            <filters>
+                                <filter>
+                                    <artifact>*:*</artifact>
+                                    <excludes>
+                                        <exclude>META-INF/*.SF</exclude>
+                                        <exclude>META-INF/*.DSA</exclude>
+                                        <exclude>META-INF/*.RSA</exclude>
+                                        <exclude>org/slf4j/**</exclude>
+                                        <exclude>META-INF/maven/org.slf4j/**</exclude>
+                                    </excludes>
+                                </filter>
+                            </filters>
+                            <transformers>
+                                <transformer
+                                        implementation="org.apache.maven.plugins.shade.resource.ServicesResourceTransformer"/>
+                            </transformers>
+                            <relocations>
+                                <relocation>
+                                    <pattern>com.google</pattern>
+                                    <shadedPattern>shaded.com.google</shadedPattern>
+                                </relocation>
+                                <relocation>
+                                    <pattern>io.grpc.internal</pattern>
+                                    <shadedPattern>shaded.io.grpc.internal</shadedPattern>
+                                </relocation>
+                                <relocation>
+                                    <pattern>org.apache.commons</pattern>
+                                    <shadedPattern>shaded.org.apache.commons</shadedPattern>
+                                    <excludes>
+                                        <exclude>org.apache.hadoop</exclude>
+                                        <exclude>org.apache.log4j</exclude>
+                                    </excludes>
+                                </relocation>
+                                <relocation>
+                                    <pattern>org.apache.http</pattern>
+                                    <shadedPattern>shaded.org.apache.http</shadedPattern>
+                                </relocation>
+                            </relocations>
+                        </configuration>
+                    </execution>
+                </executions>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-gpg-plugin</artifactId>
+                <version>1.5</version>
+                <executions>
+                    <execution>
+                        <id>sign-artifacts</id>
+                        <phase>verify</phase>
+                        <goals>
+                            <goal>sign</goal>
+                        </goals>
+                    </execution>
+                </executions>
+            </plugin>
+            <plugin>
+                <groupId>org.sonatype.plugins</groupId>
+                <artifactId>nexus-staging-maven-plugin</artifactId>
+                <version>1.6.7</version>
+                <extensions>true</extensions>
+                <configuration>
+                    <serverId>ossrh</serverId>
+                    <nexusUrl>https://oss.sonatype.org/</nexusUrl>
+                    <autoReleaseAfterClose>true</autoReleaseAfterClose>
+                </configuration>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-source-plugin</artifactId>
+                <version>2.2.1</version>
+                <executions>
+                    <execution>
+                        <id>attach-sources</id>
+                        <goals>
+                            <goal>jar-no-fork</goal>
+                        </goals>
+                    </execution>
+                </executions>
+            </plugin>
+            <plugin>
+                <groupId>org.apache.maven.plugins</groupId>
+                <artifactId>maven-javadoc-plugin</artifactId>
+                <version>2.9.1</version>
+                <executions>
+                    <execution>
+                        <id>attach-javadocs</id>
+                        <goals>
+                            <goal>jar</goal>
+                        </goals>
+                    </execution>
+                </executions>
+            </plugin>
+        </plugins>
+    </build>
+
+    <dependencies>
+        <dependency>
+            <groupId>org.apache.hadoop</groupId>
+            <artifactId>hadoop-client</artifactId>
+            <version>${hadoop.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>com.github.chrislusf</groupId>
+            <artifactId>seaweedfs-client</artifactId>
+            <version>${seaweedfs.client.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>org.apache.hadoop</groupId>
+            <artifactId>hadoop-common</artifactId>
+            <version>${hadoop.version}</version>
+        </dependency>
+    </dependencies>
+
+</project>

other/java/hdfs3/src/main/java/seaweed/hdfs/ReadBuffer.java

@@ -0,0 +1,137 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package seaweed.hdfs;
+
+import java.util.concurrent.CountDownLatch;
+
+class ReadBuffer {
+
+  private SeaweedInputStream stream;
+  private long offset;                   // offset within the file for the buffer
+  private int length;                    // actual length, set after the buffer is filles
+  private int requestedLength;           // requested length of the read
+  private byte[] buffer;                 // the buffer itself
+  private int bufferindex = -1;          // index in the buffers array in Buffer manager
+  private ReadBufferStatus status;             // status of the buffer
+  private CountDownLatch latch = null;   // signaled when the buffer is done reading, so any client
+  // waiting on this buffer gets unblocked
+
+  // fields to help with eviction logic
+  private long timeStamp = 0;  // tick at which buffer became available to read
+  private boolean isFirstByteConsumed = false;
+  private boolean isLastByteConsumed = false;
+  private boolean isAnyByteConsumed = false;
+
+  public SeaweedInputStream getStream() {
+    return stream;
+  }
+
+  public void setStream(SeaweedInputStream stream) {
+    this.stream = stream;
+  }
+
+  public long getOffset() {
+    return offset;
+  }
+
+  public void setOffset(long offset) {
+    this.offset = offset;
+  }
+
+  public int getLength() {
+    return length;
+  }
+
+  public void setLength(int length) {
+    this.length = length;
+  }
+
+  public int getRequestedLength() {
+    return requestedLength;
+  }
+
+  public void setRequestedLength(int requestedLength) {
+    this.requestedLength = requestedLength;
+  }
+
+  public byte[] getBuffer() {
+    return buffer;
+  }
+
+  public void setBuffer(byte[] buffer) {
+    this.buffer = buffer;
+  }
+
+  public int getBufferindex() {
+    return bufferindex;
+  }
+
+  public void setBufferindex(int bufferindex) {
+    this.bufferindex = bufferindex;
+  }
+
+  public ReadBufferStatus getStatus() {
+    return status;
+  }
+
+  public void setStatus(ReadBufferStatus status) {
+    this.status = status;
+  }
+
+  public CountDownLatch getLatch() {
+    return latch;
+  }
+
+  public void setLatch(CountDownLatch latch) {
+    this.latch = latch;
+  }
+
+  public long getTimeStamp() {
+    return timeStamp;
+  }
+
+  public void setTimeStamp(long timeStamp) {
+    this.timeStamp = timeStamp;
+  }
+
+  public boolean isFirstByteConsumed() {
+    return isFirstByteConsumed;
+  }
+
+  public void setFirstByteConsumed(boolean isFirstByteConsumed) {
+    this.isFirstByteConsumed = isFirstByteConsumed;
+  }
+
+  public boolean isLastByteConsumed() {
+    return isLastByteConsumed;
+  }
+
+  public void setLastByteConsumed(boolean isLastByteConsumed) {
+    this.isLastByteConsumed = isLastByteConsumed;
+  }
+
+  public boolean isAnyByteConsumed() {
+    return isAnyByteConsumed;
+  }
+
+  public void setAnyByteConsumed(boolean isAnyByteConsumed) {
+    this.isAnyByteConsumed = isAnyByteConsumed;
+  }
+
+}

other/java/hdfs3/src/main/java/seaweed/hdfs/ReadBufferManager.java

@@ -0,0 +1,394 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package seaweed.hdfs;
+
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.util.Collection;
+import java.util.LinkedList;
+import java.util.Queue;
+import java.util.Stack;
+import java.util.concurrent.CountDownLatch;
+
+/**
+ * The Read Buffer Manager for Rest AbfsClient.
+ */
+final class ReadBufferManager {
+  private static final Logger LOGGER = LoggerFactory.getLogger(ReadBufferManager.class);
+
+  private static final int NUM_BUFFERS = 16;
+  private static final int BLOCK_SIZE = 4 * 1024 * 1024;
+  private static final int NUM_THREADS = 8;
+  private static final int THRESHOLD_AGE_MILLISECONDS = 3000; // have to see if 3 seconds is a good threshold
+
+  private Thread[] threads = new Thread[NUM_THREADS];
+  private byte[][] buffers;    // array of byte[] buffers, to hold the data that is read
+  private Stack<Integer> freeList = new Stack<>();   // indices in buffers[] array that are available
+
+  private Queue<ReadBuffer> readAheadQueue = new LinkedList<>(); // queue of requests that are not picked up by any worker thread yet
+  private LinkedList<ReadBuffer> inProgressList = new LinkedList<>(); // requests being processed by worker threads
+  private LinkedList<ReadBuffer> completedReadList = new LinkedList<>(); // buffers available for reading
+  private static final ReadBufferManager BUFFER_MANAGER; // singleton, initialized in static initialization block
+
+  static {
+    BUFFER_MANAGER = new ReadBufferManager();
+    BUFFER_MANAGER.init();
+  }
+
+  static ReadBufferManager getBufferManager() {
+    return BUFFER_MANAGER;
+  }
+
+  private void init() {
+    buffers = new byte[NUM_BUFFERS][];
+    for (int i = 0; i < NUM_BUFFERS; i++) {
+      buffers[i] = new byte[BLOCK_SIZE];  // same buffers are reused. The byte array never goes back to GC
+      freeList.add(i);
+    }
+    for (int i = 0; i < NUM_THREADS; i++) {
+      Thread t = new Thread(new ReadBufferWorker(i));
+      t.setDaemon(true);
+      threads[i] = t;
+      t.setName("SeaweedFS-prefetch-" + i);
+      t.start();
+    }
+    ReadBufferWorker.UNLEASH_WORKERS.countDown();
+  }
+
+  // hide instance constructor
+  private ReadBufferManager() {
+  }
+
+
+  /*
+   *
+   *  SeaweedInputStream-facing methods
+   *
+   */
+
+
+  /**
+   * {@link SeaweedInputStream} calls this method to queue read-aheads.
+   *
+   * @param stream          The {@link SeaweedInputStream} for which to do the read-ahead
+   * @param requestedOffset The offset in the file which shoukd be read
+   * @param requestedLength The length to read
+   */
+  void queueReadAhead(final SeaweedInputStream stream, final long requestedOffset, final int requestedLength) {
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("Start Queueing readAhead for {} offset {} length {}",
+          stream.getPath(), requestedOffset, requestedLength);
+    }
+    ReadBuffer buffer;
+    synchronized (this) {
+      if (isAlreadyQueued(stream, requestedOffset)) {
+        return; // already queued, do not queue again
+      }
+      if (freeList.isEmpty() && !tryEvict()) {
+        return; // no buffers available, cannot queue anything
+      }
+
+      buffer = new ReadBuffer();
+      buffer.setStream(stream);
+      buffer.setOffset(requestedOffset);
+      buffer.setLength(0);
+      buffer.setRequestedLength(requestedLength);
+      buffer.setStatus(ReadBufferStatus.NOT_AVAILABLE);
+      buffer.setLatch(new CountDownLatch(1));
+
+      Integer bufferIndex = freeList.pop();  // will return a value, since we have checked size > 0 already
+
+      buffer.setBuffer(buffers[bufferIndex]);
+      buffer.setBufferindex(bufferIndex);
+      readAheadQueue.add(buffer);
+      notifyAll();
+    }
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("Done q-ing readAhead for file {} offset {} buffer idx {}",
+          stream.getPath(), requestedOffset, buffer.getBufferindex());
+    }
+  }
+
+
+  /**
+   * {@link SeaweedInputStream} calls this method read any bytes already available in a buffer (thereby saving a
+   * remote read). This returns the bytes if the data already exists in buffer. If there is a buffer that is reading
+   * the requested offset, then this method blocks until that read completes. If the data is queued in a read-ahead
+   * but not picked up by a worker thread yet, then it cancels that read-ahead and reports cache miss. This is because
+   * depending on worker thread availability, the read-ahead may take a while - the calling thread can do it's own
+   * read to get the data faster (copmared to the read waiting in queue for an indeterminate amount of time).
+   *
+   * @param stream   the file to read bytes for
+   * @param position the offset in the file to do a read for
+   * @param length   the length to read
+   * @param buffer   the buffer to read data into. Note that the buffer will be written into from offset 0.
+   * @return the number of bytes read
+   */
+  int getBlock(final SeaweedInputStream stream, final long position, final int length, final byte[] buffer) {
+    // not synchronized, so have to be careful with locking
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("getBlock for file {}  position {}  thread {}",
+          stream.getPath(), position, Thread.currentThread().getName());
+    }
+
+    waitForProcess(stream, position);
+
+    int bytesRead = 0;
+    synchronized (this) {
+      bytesRead = getBlockFromCompletedQueue(stream, position, length, buffer);
+    }
+    if (bytesRead > 0) {
+      if (LOGGER.isTraceEnabled()) {
+        LOGGER.trace("Done read from Cache for {} position {} length {}",
+            stream.getPath(), position, bytesRead);
+      }
+      return bytesRead;
+    }
+
+    // otherwise, just say we got nothing - calling thread can do its own read
+    return 0;
+  }
+
+  /*
+   *
+   *  Internal methods
+   *
+   */
+
+  private void waitForProcess(final SeaweedInputStream stream, final long position) {
+    ReadBuffer readBuf;
+    synchronized (this) {
+      clearFromReadAheadQueue(stream, position);
+      readBuf = getFromList(inProgressList, stream, position);
+    }
+    if (readBuf != null) {         // if in in-progress queue, then block for it
+      try {
+        if (LOGGER.isTraceEnabled()) {
+          LOGGER.trace("got a relevant read buffer for file {} offset {} buffer idx {}",
+              stream.getPath(), readBuf.getOffset(), readBuf.getBufferindex());
+        }
+        readBuf.getLatch().await();  // blocking wait on the caller stream's thread
+        // Note on correctness: readBuf gets out of inProgressList only in 1 place: after worker thread
+        // is done processing it (in doneReading). There, the latch is set after removing the buffer from
+        // inProgressList. So this latch is safe to be outside the synchronized block.
+        // Putting it in synchronized would result in a deadlock, since this thread would be holding the lock
+        // while waiting, so no one will be able to  change any state. If this becomes more complex in the future,
+        // then the latch cane be removed and replaced with wait/notify whenever inProgressList is touched.
+      } catch (InterruptedException ex) {
+        Thread.currentThread().interrupt();
+      }
+      if (LOGGER.isTraceEnabled()) {
+        LOGGER.trace("latch done for file {} buffer idx {} length {}",
+            stream.getPath(), readBuf.getBufferindex(), readBuf.getLength());
+      }
+    }
+  }
+
+  /**
+   * If any buffer in the completedlist can be reclaimed then reclaim it and return the buffer to free list.
+   * The objective is to find just one buffer - there is no advantage to evicting more than one.
+   *
+   * @return whether the eviction succeeeded - i.e., were we able to free up one buffer
+   */
+  private synchronized boolean tryEvict() {
+    ReadBuffer nodeToEvict = null;
+    if (completedReadList.size() <= 0) {
+      return false;  // there are no evict-able buffers
+    }
+
+    // first, try buffers where all bytes have been consumed (approximated as first and last bytes consumed)
+    for (ReadBuffer buf : completedReadList) {
+      if (buf.isFirstByteConsumed() && buf.isLastByteConsumed()) {
+        nodeToEvict = buf;
+        break;
+      }
+    }
+    if (nodeToEvict != null) {
+      return evict(nodeToEvict);
+    }
+
+    // next, try buffers where any bytes have been consumed (may be a bad idea? have to experiment and see)
+    for (ReadBuffer buf : completedReadList) {
+      if (buf.isAnyByteConsumed()) {
+        nodeToEvict = buf;
+        break;
+      }
+    }
+
+    if (nodeToEvict != null) {
+      return evict(nodeToEvict);
+    }
+
+    // next, try any old nodes that have not been consumed
+    long earliestBirthday = Long.MAX_VALUE;
+    for (ReadBuffer buf : completedReadList) {
+      if (buf.getTimeStamp() < earliestBirthday) {
+        nodeToEvict = buf;
+        earliestBirthday = buf.getTimeStamp();
+      }
+    }
+    if ((currentTimeMillis() - earliestBirthday > THRESHOLD_AGE_MILLISECONDS) && (nodeToEvict != null)) {
+      return evict(nodeToEvict);
+    }
+
+    // nothing can be evicted
+    return false;
+  }
+
+  private boolean evict(final ReadBuffer buf) {
+    freeList.push(buf.getBufferindex());
+    completedReadList.remove(buf);
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("Evicting buffer idx {}; was used for file {} offset {} length {}",
+          buf.getBufferindex(), buf.getStream().getPath(), buf.getOffset(), buf.getLength());
+    }
+    return true;
+  }
+
+  private boolean isAlreadyQueued(final SeaweedInputStream stream, final long requestedOffset) {
+    // returns true if any part of the buffer is already queued
+    return (isInList(readAheadQueue, stream, requestedOffset)
+        || isInList(inProgressList, stream, requestedOffset)
+        || isInList(completedReadList, stream, requestedOffset));
+  }
+
+  private boolean isInList(final Collection<ReadBuffer> list, final SeaweedInputStream stream, final long requestedOffset) {
+    return (getFromList(list, stream, requestedOffset) != null);
+  }
+
+  private ReadBuffer getFromList(final Collection<ReadBuffer> list, final SeaweedInputStream stream, final long requestedOffset) {
+    for (ReadBuffer buffer : list) {
+      if (buffer.getStream() == stream) {
+        if (buffer.getStatus() == ReadBufferStatus.AVAILABLE
+            && requestedOffset >= buffer.getOffset()
+            && requestedOffset < buffer.getOffset() + buffer.getLength()) {
+          return buffer;
+        } else if (requestedOffset >= buffer.getOffset()
+            && requestedOffset < buffer.getOffset() + buffer.getRequestedLength()) {
+          return buffer;
+        }
+      }
+    }
+    return null;
+  }
+
+  private void clearFromReadAheadQueue(final SeaweedInputStream stream, final long requestedOffset) {
+    ReadBuffer buffer = getFromList(readAheadQueue, stream, requestedOffset);
+    if (buffer != null) {
+      readAheadQueue.remove(buffer);
+      notifyAll();   // lock is held in calling method
+      freeList.push(buffer.getBufferindex());
+    }
+  }
+
+  private int getBlockFromCompletedQueue(final SeaweedInputStream stream, final long position, final int length,
+                                         final byte[] buffer) {
+    ReadBuffer buf = getFromList(completedReadList, stream, position);
+    if (buf == null || position >= buf.getOffset() + buf.getLength()) {
+      return 0;
+    }
+    int cursor = (int) (position - buf.getOffset());
+    int availableLengthInBuffer = buf.getLength() - cursor;
+    int lengthToCopy = Math.min(length, availableLengthInBuffer);
+    System.arraycopy(buf.getBuffer(), cursor, buffer, 0, lengthToCopy);
+    if (cursor == 0) {
+      buf.setFirstByteConsumed(true);
+    }
+    if (cursor + lengthToCopy == buf.getLength()) {
+      buf.setLastByteConsumed(true);
+    }
+    buf.setAnyByteConsumed(true);
+    return lengthToCopy;
+  }
+
+  /*
+   *
+   *  ReadBufferWorker-thread-facing methods
+   *
+   */
+
+  /**
+   * ReadBufferWorker thread calls this to get the next buffer that it should work on.
+   *
+   * @return {@link ReadBuffer}
+   * @throws InterruptedException if thread is interrupted
+   */
+  ReadBuffer getNextBlockToRead() throws InterruptedException {
+    ReadBuffer buffer = null;
+    synchronized (this) {
+      //buffer = readAheadQueue.take();  // blocking method
+      while (readAheadQueue.size() == 0) {
+        wait();
+      }
+      buffer = readAheadQueue.remove();
+      notifyAll();
+      if (buffer == null) {
+        return null;            // should never happen
+      }
+      buffer.setStatus(ReadBufferStatus.READING_IN_PROGRESS);
+      inProgressList.add(buffer);
+    }
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("ReadBufferWorker picked file {} for offset {}",
+          buffer.getStream().getPath(), buffer.getOffset());
+    }
+    return buffer;
+  }
+
+  /**
+   * ReadBufferWorker thread calls this method to post completion.
+   *
+   * @param buffer            the buffer whose read was completed
+   * @param result            the {@link ReadBufferStatus} after the read operation in the worker thread
+   * @param bytesActuallyRead the number of bytes that the worker thread was actually able to read
+   */
+  void doneReading(final ReadBuffer buffer, final ReadBufferStatus result, final int bytesActuallyRead) {
+    if (LOGGER.isTraceEnabled()) {
+      LOGGER.trace("ReadBufferWorker completed file {} for offset {} bytes {}",
+          buffer.getStream().getPath(),  buffer.getOffset(), bytesActuallyRead);
+    }
+    synchronized (this) {
+      inProgressList.remove(buffer);
+      if (result == ReadBufferStatus.AVAILABLE && bytesActuallyRead > 0) {
+        buffer.setStatus(ReadBufferStatus.AVAILABLE);
+        buffer.setTimeStamp(currentTimeMillis());
+        buffer.setLength(bytesActuallyRead);
+        completedReadList.add(buffer);
+      } else {
+        freeList.push(buffer.getBufferindex());
+        // buffer should go out of scope after the end of the calling method in ReadBufferWorker, and eligible for GC
+      }
+    }
+    //outside the synchronized, since anyone receiving a wake-up from the latch must see safe-published results
+    buffer.getLatch().countDown(); // wake up waiting threads (if any)
+  }
+
+  /**
+   * Similar to System.currentTimeMillis, except implemented with System.nanoTime().
+   * System.currentTimeMillis can go backwards when system clock is changed (e.g., with NTP time synchronization),
+   * making it unsuitable for measuring time intervals. nanotime is strictly monotonically increasing per CPU core.
+   * Note: it is not monotonic across Sockets, and even within a CPU, its only the
+   * more recent parts which share a clock across all cores.
+   *
+   * @return current time in milliseconds
+   */
+  private long currentTimeMillis() {
+    return System.nanoTime() / 1000 / 1000;
+  }
+}

other/java/hdfs3/src/main/java/seaweed/hdfs/ReadBufferStatus.java

@@ -0,0 +1,29 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package seaweed.hdfs;
+
+/**
+ * The ReadBufferStatus for Rest AbfsClient
+ */
+public enum ReadBufferStatus {
+  NOT_AVAILABLE,  // buffers sitting in readaheadqueue have this stats
+  READING_IN_PROGRESS,  // reading is in progress on this buffer. Buffer should be in inProgressList
+  AVAILABLE,  // data is available in buffer. It should be in completedList
+  READ_FAILED  // read completed, but failed.
+}

other/java/hdfs3/src/main/java/seaweed/hdfs/ReadBufferWorker.java

@@ -0,0 +1,70 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package seaweed.hdfs;
+
+import java.util.concurrent.CountDownLatch;
+
+class ReadBufferWorker implements Runnable {
+
+  protected static final CountDownLatch UNLEASH_WORKERS = new CountDownLatch(1);
+  private int id;
+
+  ReadBufferWorker(final int id) {
+    this.id = id;
+  }
+
+  /**
+   * return the ID of ReadBufferWorker.
+   */
+  public int getId() {
+    return this.id;
+  }
+
+  /**
+   * Waits until a buffer becomes available in ReadAheadQueue.
+   * Once a buffer becomes available, reads the file specified in it and then posts results back to buffer manager.
+   * Rinse and repeat. Forever.
+   */
+  public void run() {
+    try {
+      UNLEASH_WORKERS.await();
+    } catch (InterruptedException ex) {
+      Thread.currentThread().interrupt();
+    }
+    ReadBufferManager bufferManager = ReadBufferManager.getBufferManager();
+    ReadBuffer buffer;
+    while (true) {
+      try {
+        buffer = bufferManager.getNextBlockToRead();   // blocks, until a buffer is available for this thread
+      } catch (InterruptedException ex) {
+        Thread.currentThread().interrupt();
+        return;
+      }
+      if (buffer != null) {
+        try {
+          // do the actual read, from the file.
+          int bytesRead = buffer.getStream().readRemote(buffer.getOffset(), buffer.getBuffer(), 0, buffer.getRequestedLength());
+          bufferManager.doneReading(buffer, ReadBufferStatus.AVAILABLE, bytesRead);  // post result back to ReadBufferManager
+        } catch (Exception ex) {
+          bufferManager.doneReading(buffer, ReadBufferStatus.READ_FAILED, 0);
+        }
+      }
+    }
+  }
+}

other/java/hdfs3/src/main/java/seaweed/hdfs/SeaweedFileSystem.java

@@ -0,0 +1,620 @@
+package seaweed.hdfs;
+
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.*;
+import org.apache.hadoop.fs.permission.AclEntry;
+import org.apache.hadoop.fs.permission.AclStatus;
+import org.apache.hadoop.fs.permission.FsPermission;
+import org.apache.hadoop.security.AccessControlException;
+import org.apache.hadoop.security.UserGroupInformation;
+import org.apache.hadoop.util.Progressable;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.io.FileNotFoundException;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.OutputStream;
+import java.net.URI;
+import java.util.EnumSet;
+import java.util.List;
+import java.util.Map;
+
+import static org.apache.hadoop.fs.CommonConfigurationKeysPublic.IO_FILE_BUFFER_SIZE_KEY;
+
+public class SeaweedFileSystem extends FileSystem {
+
+    public static final int FS_SEAWEED_DEFAULT_PORT = 8888;
+    public static final String FS_SEAWEED_FILER_HOST = "fs.seaweed.filer.host";
+    public static final String FS_SEAWEED_FILER_PORT = "fs.seaweed.filer.port";
+
+    private static final Logger LOG = LoggerFactory.getLogger(SeaweedFileSystem.class);
+    private static int BUFFER_SIZE = 16 * 1024 * 1024;
+
+    private URI uri;
+    private Path workingDirectory = new Path("/");
+    private SeaweedFileSystemStore seaweedFileSystemStore;
+
+    public URI getUri() {
+        return uri;
+    }
+
+    public String getScheme() {
+        return "seaweedfs";
+    }
+
+    @Override
+    public void initialize(URI uri, Configuration conf) throws IOException { // get
+        super.initialize(uri, conf);
+
+        // get host information from uri (overrides info in conf)
+        String host = uri.getHost();
+        host = (host == null) ? conf.get(FS_SEAWEED_FILER_HOST, "localhost") : host;
+        if (host == null) {
+            throw new IOException("Invalid host specified");
+        }
+        conf.set(FS_SEAWEED_FILER_HOST, host);
+
+        // get port information from uri, (overrides info in conf)
+        int port = uri.getPort();
+        port = (port == -1) ? FS_SEAWEED_DEFAULT_PORT : port;
+        conf.setInt(FS_SEAWEED_FILER_PORT, port);
+
+        conf.setInt(IO_FILE_BUFFER_SIZE_KEY, BUFFER_SIZE);
+
+        setConf(conf);
+        this.uri = uri;
+
+        seaweedFileSystemStore = new SeaweedFileSystemStore(host, port);
+
+    }
+
+    @Override
+    public FSDataInputStream open(Path path, int bufferSize) throws IOException {
+
+        LOG.debug("open path: {} bufferSize:{}", path, bufferSize);
+
+        path = qualify(path);
+
+        try {
+            InputStream inputStream = seaweedFileSystemStore.openFileForRead(path, statistics, bufferSize);
+            return new FSDataInputStream(inputStream);
+        } catch (Exception ex) {
+            LOG.warn("open path: {} bufferSize:{}", path, bufferSize, ex);
+            return null;
+        }
+    }
+
+    @Override
+    public FSDataOutputStream create(Path path, FsPermission permission, final boolean overwrite, final int bufferSize,
+                                     final short replication, final long blockSize, final Progressable progress) throws IOException {
+
+        LOG.debug("create path: {} bufferSize:{} blockSize:{}", path, bufferSize, blockSize);
+
+        path = qualify(path);
+
+        try {
+            String replicaPlacement = String.format("%03d", replication - 1);
+            OutputStream outputStream = seaweedFileSystemStore.createFile(path, overwrite, permission, bufferSize, replicaPlacement);
+            return new FSDataOutputStream(outputStream, statistics);
+        } catch (Exception ex) {
+            LOG.warn("create path: {} bufferSize:{} blockSize:{}", path, bufferSize, blockSize, ex);
+            return null;
+        }
+    }
+
+    /**
+     * {@inheritDoc}
+     * @throws FileNotFoundException if the parent directory is not present -or
+     * is not a directory.
+     */
+    @Override
+    public FSDataOutputStream createNonRecursive(Path path,
+                                                 FsPermission permission,
+                                                 EnumSet<CreateFlag> flags,
+                                                 int bufferSize,
+                                                 short replication,
+                                                 long blockSize,
+                                                 Progressable progress) throws IOException {
+        Path parent = path.getParent();
+        if (parent != null) {
+            // expect this to raise an exception if there is no parent
+            if (!getFileStatus(parent).isDirectory()) {
+                throw new FileAlreadyExistsException("Not a directory: " + parent);
+            }
+        }
+        return create(path, permission,
+                flags.contains(CreateFlag.OVERWRITE), bufferSize,
+                replication, blockSize, progress);
+    }
+
+    @Override
+    public FSDataOutputStream append(Path path, int bufferSize, Progressable progressable) throws IOException {
+
+        LOG.debug("append path: {} bufferSize:{}", path, bufferSize);
+
+        path = qualify(path);
+        try {
+            OutputStream outputStream = seaweedFileSystemStore.createFile(path, false, null, bufferSize, "");
+            return new FSDataOutputStream(outputStream, statistics);
+        } catch (Exception ex) {
+            LOG.warn("append path: {} bufferSize:{}", path, bufferSize, ex);
+            return null;
+        }
+    }
+
+    @Override
+    public boolean rename(Path src, Path dst) {
+
+        LOG.debug("rename path: {} => {}", src, dst);
+
+        if (src.isRoot()) {
+            return false;
+        }
+
+        if (src.equals(dst)) {
+            return true;
+        }
+        FileStatus dstFileStatus = getFileStatus(dst);
+
+        String sourceFileName = src.getName();
+        Path adjustedDst = dst;
+
+        if (dstFileStatus != null) {
+            if (!dstFileStatus.isDirectory()) {
+                return false;
+            }
+            adjustedDst = new Path(dst, sourceFileName);
+        }
+
+        Path qualifiedSrcPath = qualify(src);
+        Path qualifiedDstPath = qualify(adjustedDst);
+
+        seaweedFileSystemStore.rename(qualifiedSrcPath, qualifiedDstPath);
+        return true;
+    }
+
+    @Override
+    public boolean delete(Path path, boolean recursive) {
+
+        LOG.debug("delete path: {} recursive:{}", path, recursive);
+
+        path = qualify(path);
+
+        FileStatus fileStatus = getFileStatus(path);
+
+        if (fileStatus == null) {
+            return true;
+        }
+
+        return seaweedFileSystemStore.deleteEntries(path, fileStatus.isDirectory(), recursive);
+
+    }
+
+    @Override
+    public FileStatus[] listStatus(Path path) throws IOException {
+
+        LOG.debug("listStatus path: {}", path);
+
+        path = qualify(path);
+
+        return seaweedFileSystemStore.listEntries(path);
+    }
+
+    @Override
+    public Path getWorkingDirectory() {
+        return workingDirectory;
+    }
+
+    @Override
+    public void setWorkingDirectory(Path path) {
+        if (path.isAbsolute()) {
+            workingDirectory = path;
+        } else {
+            workingDirectory = new Path(workingDirectory, path);
+        }
+    }
+
+    @Override
+    public boolean mkdirs(Path path, FsPermission fsPermission) throws IOException {
+
+        LOG.debug("mkdirs path: {}", path);
+
+        path = qualify(path);
+
+        FileStatus fileStatus = getFileStatus(path);
+
+        if (fileStatus == null) {
+
+            UserGroupInformation currentUser = UserGroupInformation.getCurrentUser();
+            return seaweedFileSystemStore.createDirectory(path, currentUser,
+                    fsPermission == null ? FsPermission.getDirDefault() : fsPermission,
+                    FsPermission.getUMask(getConf()));
+
+        }
+
+        if (fileStatus.isDirectory()) {
+            return true;
+        } else {
+            throw new FileAlreadyExistsException("Path is a file: " + path);
+        }
+    }
+
+    @Override
+    public FileStatus getFileStatus(Path path) {
+
+        LOG.debug("getFileStatus path: {}", path);
+
+        path = qualify(path);
+
+        return seaweedFileSystemStore.getFileStatus(path);
+    }
+
+    /**
+     * Set owner of a path (i.e. a file or a directory).
+     * The parameters owner and group cannot both be null.
+     *
+     * @param path  The path
+     * @param owner If it is null, the original username remains unchanged.
+     * @param group If it is null, the original groupname remains unchanged.
+     */
+    @Override
+    public void setOwner(Path path, final String owner, final String group)
+            throws IOException {
+        LOG.debug("setOwner path: {}", path);
+        path = qualify(path);
+
+        seaweedFileSystemStore.setOwner(path, owner, group);
+    }
+
+
+    /**
+     * Set permission of a path.
+     *
+     * @param path       The path
+     * @param permission Access permission
+     */
+    @Override
+    public void setPermission(Path path, final FsPermission permission) throws IOException {
+        LOG.debug("setPermission path: {}", path);
+
+        if (permission == null) {
+            throw new IllegalArgumentException("The permission can't be null");
+        }
+
+        path = qualify(path);
+
+        seaweedFileSystemStore.setPermission(path, permission);
+    }
+
+    Path qualify(Path path) {
+        return path.makeQualified(uri, workingDirectory);
+    }
+
+    /**
+     * Concat existing files together.
+     *
+     * @param trg   the path to the target destination.
+     * @param psrcs the paths to the sources to use for the concatenation.
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default).
+     */
+    @Override
+    public void concat(final Path trg, final Path[] psrcs) throws IOException {
+        throw new UnsupportedOperationException("Not implemented by the " +
+                getClass().getSimpleName() + " FileSystem implementation");
+    }
+
+    /**
+     * Truncate the file in the indicated path to the indicated size.
+     * <ul>
+     * <li>Fails if path is a directory.</li>
+     * <li>Fails if path does not exist.</li>
+     * <li>Fails if path is not closed.</li>
+     * <li>Fails if new size is greater than current size.</li>
+     * </ul>
+     *
+     * @param f         The path to the file to be truncated
+     * @param newLength The size the file is to be truncated to
+     * @return <code>true</code> if the file has been truncated to the desired
+     * <code>newLength</code> and is immediately available to be reused for
+     * write operations such as <code>append</code>, or
+     * <code>false</code> if a background process of adjusting the length of
+     * the last block has been started, and clients should wait for it to
+     * complete before proceeding with further file updates.
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default).
+     */
+    @Override
+    public boolean truncate(Path f, long newLength) throws IOException {
+        throw new UnsupportedOperationException("Not implemented by the " +
+                getClass().getSimpleName() + " FileSystem implementation");
+    }
+
+    @Override
+    public void createSymlink(final Path target, final Path link,
+                              final boolean createParent) throws AccessControlException,
+            FileAlreadyExistsException, FileNotFoundException,
+            ParentNotDirectoryException, UnsupportedFileSystemException,
+            IOException {
+        // Supporting filesystems should override this method
+        throw new UnsupportedOperationException(
+                "Filesystem does not support symlinks!");
+    }
+
+    public boolean supportsSymlinks() {
+        return false;
+    }
+
+    /**
+     * Create a snapshot.
+     *
+     * @param path         The directory where snapshots will be taken.
+     * @param snapshotName The name of the snapshot
+     * @return the snapshot path.
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     */
+    @Override
+    public Path createSnapshot(Path path, String snapshotName)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support createSnapshot");
+    }
+
+    /**
+     * Rename a snapshot.
+     *
+     * @param path            The directory path where the snapshot was taken
+     * @param snapshotOldName Old name of the snapshot
+     * @param snapshotNewName New name of the snapshot
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void renameSnapshot(Path path, String snapshotOldName,
+                               String snapshotNewName) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support renameSnapshot");
+    }
+
+    /**
+     * Delete a snapshot of a directory.
+     *
+     * @param path         The directory that the to-be-deleted snapshot belongs to
+     * @param snapshotName The name of the snapshot
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void deleteSnapshot(Path path, String snapshotName)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support deleteSnapshot");
+    }
+
+    /**
+     * Modifies ACL entries of files and directories.  This method can add new ACL
+     * entries or modify the permissions on existing ACL entries.  All existing
+     * ACL entries that are not specified in this call are retained without
+     * changes.  (Modifications are merged into the current ACL.)
+     *
+     * @param path    Path to modify
+     * @param aclSpec List&lt;AclEntry&gt; describing modifications
+     * @throws IOException                   if an ACL could not be modified
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void modifyAclEntries(Path path, List<AclEntry> aclSpec)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support modifyAclEntries");
+    }
+
+    /**
+     * Removes ACL entries from files and directories.  Other ACL entries are
+     * retained.
+     *
+     * @param path    Path to modify
+     * @param aclSpec List describing entries to remove
+     * @throws IOException                   if an ACL could not be modified
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void removeAclEntries(Path path, List<AclEntry> aclSpec)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support removeAclEntries");
+    }
+
+    /**
+     * Removes all default ACL entries from files and directories.
+     *
+     * @param path Path to modify
+     * @throws IOException                   if an ACL could not be modified
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void removeDefaultAcl(Path path)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support removeDefaultAcl");
+    }
+
+    /**
+     * Removes all but the base ACL entries of files and directories.  The entries
+     * for user, group, and others are retained for compatibility with permission
+     * bits.
+     *
+     * @param path Path to modify
+     * @throws IOException                   if an ACL could not be removed
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void removeAcl(Path path)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support removeAcl");
+    }
+
+    /**
+     * Fully replaces ACL of files and directories, discarding all existing
+     * entries.
+     *
+     * @param path    Path to modify
+     * @param aclSpec List describing modifications, which must include entries
+     *                for user, group, and others for compatibility with permission bits.
+     * @throws IOException                   if an ACL could not be modified
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void setAcl(Path path, List<AclEntry> aclSpec) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support setAcl");
+    }
+
+    /**
+     * Gets the ACL of a file or directory.
+     *
+     * @param path Path to get
+     * @return AclStatus describing the ACL of the file or directory
+     * @throws IOException                   if an ACL could not be read
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public AclStatus getAclStatus(Path path) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support getAclStatus");
+    }
+
+    /**
+     * Set an xattr of a file or directory.
+     * The name must be prefixed with the namespace followed by ".". For example,
+     * "user.attr".
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path  Path to modify
+     * @param name  xattr name.
+     * @param value xattr value.
+     * @param flag  xattr set flag
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void setXAttr(Path path, String name, byte[] value,
+                         EnumSet<XAttrSetFlag> flag) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support setXAttr");
+    }
+
+    /**
+     * Get an xattr name and value for a file or directory.
+     * The name must be prefixed with the namespace followed by ".". For example,
+     * "user.attr".
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path Path to get extended attribute
+     * @param name xattr name.
+     * @return byte[] xattr value.
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public byte[] getXAttr(Path path, String name) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support getXAttr");
+    }
+
+    /**
+     * Get all of the xattr name/value pairs for a file or directory.
+     * Only those xattrs which the logged-in user has permissions to view
+     * are returned.
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path Path to get extended attributes
+     * @return Map describing the XAttrs of the file or directory
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public Map<String, byte[]> getXAttrs(Path path) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support getXAttrs");
+    }
+
+    /**
+     * Get all of the xattrs name/value pairs for a file or directory.
+     * Only those xattrs which the logged-in user has permissions to view
+     * are returned.
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path  Path to get extended attributes
+     * @param names XAttr names.
+     * @return Map describing the XAttrs of the file or directory
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public Map<String, byte[]> getXAttrs(Path path, List<String> names)
+            throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support getXAttrs");
+    }
+
+    /**
+     * Get all of the xattr names for a file or directory.
+     * Only those xattr names which the logged-in user has permissions to view
+     * are returned.
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path Path to get extended attributes
+     * @return List{@literal <String>} of the XAttr names of the file or directory
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public List<String> listXAttrs(Path path) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support listXAttrs");
+    }
+
+    /**
+     * Remove an xattr of a file or directory.
+     * The name must be prefixed with the namespace followed by ".". For example,
+     * "user.attr".
+     * <p>
+     * Refer to the HDFS extended attributes user documentation for details.
+     *
+     * @param path Path to remove extended attribute
+     * @param name xattr name
+     * @throws IOException                   IO failure
+     * @throws UnsupportedOperationException if the operation is unsupported
+     *                                       (default outcome).
+     */
+    @Override
+    public void removeXAttr(Path path, String name) throws IOException {
+        throw new UnsupportedOperationException(getClass().getSimpleName()
+                + " doesn't support removeXAttr");
+    }
+
+}

other/java/hdfs3/src/main/java/seaweed/hdfs/SeaweedFileSystemStore.java

@@ -0,0 +1,270 @@
+package seaweed.hdfs;
+
+import org.apache.hadoop.fs.FileStatus;
+import org.apache.hadoop.fs.FileSystem;
+import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.fs.permission.FsPermission;
+import org.apache.hadoop.security.UserGroupInformation;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+import seaweedfs.client.FilerClient;
+import seaweedfs.client.FilerGrpcClient;
+import seaweedfs.client.FilerProto;
+import seaweedfs.client.SeaweedRead;
+
+import java.io.FileNotFoundException;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.OutputStream;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.List;
+
+public class SeaweedFileSystemStore {
+
+    private static final Logger LOG = LoggerFactory.getLogger(SeaweedFileSystemStore.class);
+
+    private FilerGrpcClient filerGrpcClient;
+    private FilerClient filerClient;
+
+    public SeaweedFileSystemStore(String host, int port) {
+        int grpcPort = 10000 + port;
+        filerGrpcClient = new FilerGrpcClient(host, grpcPort);
+        filerClient = new FilerClient(filerGrpcClient);
+    }
+
+    public static String getParentDirectory(Path path) {
+        return path.isRoot() ? "/" : path.getParent().toUri().getPath();
+    }
+
+    static int permissionToMode(FsPermission permission, boolean isDirectory) {
+        int p = permission.toShort();
+        if (isDirectory) {
+            p = p | 1 << 31;
+        }
+        return p;
+    }
+
+    public boolean createDirectory(final Path path, UserGroupInformation currentUser,
+                                   final FsPermission permission, final FsPermission umask) {
+
+        LOG.debug("createDirectory path: {} permission: {} umask: {}",
+            path,
+            permission,
+            umask);
+
+        return filerClient.mkdirs(
+            path.toUri().getPath(),
+            permissionToMode(permission, true),
+            currentUser.getUserName(),
+            currentUser.getGroupNames()
+        );
+    }
+
+    public FileStatus[] listEntries(final Path path) {
+        LOG.debug("listEntries path: {}", path);
+
+        List<FileStatus> fileStatuses = new ArrayList<FileStatus>();
+
+        List<FilerProto.Entry> entries = filerClient.listEntries(path.toUri().getPath());
+
+        for (FilerProto.Entry entry : entries) {
+
+            FileStatus fileStatus = doGetFileStatus(new Path(path, entry.getName()), entry);
+
+            fileStatuses.add(fileStatus);
+        }
+        return fileStatuses.toArray(new FileStatus[0]);
+    }
+
+    public FileStatus getFileStatus(final Path path) {
+
+        FilerProto.Entry entry = lookupEntry(path);
+        if (entry == null) {
+            return null;
+        }
+        LOG.debug("doGetFileStatus path:{} entry:{}", path, entry);
+
+        FileStatus fileStatus = doGetFileStatus(path, entry);
+        return fileStatus;
+    }
+
+    public boolean deleteEntries(final Path path, boolean isDirectory, boolean recursive) {
+        LOG.debug("deleteEntries path: {} isDirectory {} recursive: {}",
+            path,
+            String.valueOf(isDirectory),
+            String.valueOf(recursive));
+
+        if (path.isRoot()) {
+            return true;
+        }
+
+        if (recursive && isDirectory) {
+            List<FilerProto.Entry> entries = filerClient.listEntries(path.toUri().getPath());
+            for (FilerProto.Entry entry : entries) {
+                deleteEntries(new Path(path, entry.getName()), entry.getIsDirectory(), true);
+            }
+        }
+
+        return filerClient.deleteEntry(getParentDirectory(path), path.getName(), true, recursive, true);
+    }
+
+    private FileStatus doGetFileStatus(Path path, FilerProto.Entry entry) {
+        FilerProto.FuseAttributes attributes = entry.getAttributes();
+        long length = SeaweedRead.totalSize(entry.getChunksList());
+        boolean isDir = entry.getIsDirectory();
+        int block_replication = 1;
+        int blocksize = 512;
+        long modification_time = attributes.getMtime() * 1000; // milliseconds
+        long access_time = 0;
+        FsPermission permission = FsPermission.createImmutable((short) attributes.getFileMode());
+        String owner = attributes.getUserName();
+        String group = attributes.getGroupNameCount() > 0 ? attributes.getGroupName(0) : "";
+        return new FileStatus(length, isDir, block_replication, blocksize,
+            modification_time, access_time, permission, owner, group, null, path);
+    }
+
+    private FilerProto.Entry lookupEntry(Path path) {
+
+        return filerClient.lookupEntry(getParentDirectory(path), path.getName());
+
+    }
+
+    public void rename(Path source, Path destination) {
+
+        LOG.debug("rename source: {} destination:{}", source, destination);
+
+        if (source.isRoot()) {
+            return;
+        }
+        LOG.info("rename source: {} destination:{}", source, destination);
+        FilerProto.Entry entry = lookupEntry(source);
+        if (entry == null) {
+            LOG.warn("rename non-existing source: {}", source);
+            return;
+        }
+        filerClient.mv(source.toUri().getPath(), destination.toUri().getPath());
+    }
+
+    public OutputStream createFile(final Path path,
+                                   final boolean overwrite,
+                                   FsPermission permission,
+                                   int bufferSize,
+                                   String replication) throws IOException {
+
+        permission = permission == null ? FsPermission.getFileDefault() : permission;
+
+        LOG.debug("createFile path: {} overwrite: {} permission: {}",
+            path,
+            overwrite,
+            permission.toString());
+
+        UserGroupInformation userGroupInformation = UserGroupInformation.getCurrentUser();
+        long now = System.currentTimeMillis() / 1000L;
+
+        FilerProto.Entry.Builder entry = null;
+        long writePosition = 0;
+        if (!overwrite) {
+            FilerProto.Entry existingEntry = lookupEntry(path);
+            LOG.debug("createFile merged entry path:{} existingEntry:{}", path, existingEntry);
+            if (existingEntry != null) {
+                entry = FilerProto.Entry.newBuilder();
+                entry.mergeFrom(existingEntry);
+                entry.getAttributesBuilder().setMtime(now);
+                LOG.debug("createFile merged entry path:{} entry:{} from:{}", path, entry, existingEntry);
+                writePosition = SeaweedRead.totalSize(existingEntry.getChunksList());
+                replication = existingEntry.getAttributes().getReplication();
+            }
+        }
+        if (entry == null) {
+            entry = FilerProto.Entry.newBuilder()
+                .setName(path.getName())
+                .setIsDirectory(false)
+                .setAttributes(FilerProto.FuseAttributes.newBuilder()
+                    .setFileMode(permissionToMode(permission, false))
+                    .setReplication(replication)
+                    .setCrtime(now)
+                    .setMtime(now)
+                    .setUserName(userGroupInformation.getUserName())
+                    .clearGroupName()
+                    .addAllGroupName(Arrays.asList(userGroupInformation.getGroupNames()))
+                );
+        }
+
+        return new SeaweedOutputStream(filerGrpcClient, path, entry, writePosition, bufferSize, replication);
+
+    }
+
+    public InputStream openFileForRead(final Path path, FileSystem.Statistics statistics,
+                                       int bufferSize) throws IOException {
+
+        LOG.debug("openFileForRead path:{} bufferSize:{}", path, bufferSize);
+
+        int readAheadQueueDepth = 2;
+        FilerProto.Entry entry = lookupEntry(path);
+
+        if (entry == null) {
+            throw new FileNotFoundException("read non-exist file " + path);
+        }
+
+        return new SeaweedInputStream(filerGrpcClient,
+            statistics,
+            path.toUri().getPath(),
+            entry,
+            bufferSize,
+            readAheadQueueDepth);
+    }
+
+    public void setOwner(Path path, String owner, String group) {
+
+        LOG.debug("setOwner path:{} owner:{} group:{}", path, owner, group);
+
+        FilerProto.Entry entry = lookupEntry(path);
+        if (entry == null) {
+            LOG.debug("setOwner path:{} entry:{}", path, entry);
+            return;
+        }
+
+        FilerProto.Entry.Builder entryBuilder = entry.toBuilder();
+        FilerProto.FuseAttributes.Builder attributesBuilder = entry.getAttributes().toBuilder();
+
+        if (owner != null) {
+            attributesBuilder.setUserName(owner);
+        }
+        if (group != null) {
+            attributesBuilder.clearGroupName();
+            attributesBuilder.addGroupName(group);
+        }
+
+        entryBuilder.setAttributes(attributesBuilder);
+
+        LOG.debug("setOwner path:{} entry:{}", path, entryBuilder);
+
+        filerClient.updateEntry(getParentDirectory(path), entryBuilder.build());
+
+    }
+
+    public void setPermission(Path path, FsPermission permission) {
+
+        LOG.debug("setPermission path:{} permission:{}", path, permission);
+
+        FilerProto.Entry entry = lookupEntry(path);
+        if (entry == null) {
+            LOG.debug("setPermission path:{} entry:{}", path, entry);
+            return;
+        }
+
+        FilerProto.Entry.Builder entryBuilder = entry.toBuilder();
+        FilerProto.FuseAttributes.Builder attributesBuilder = entry.getAttributes().toBuilder();
+
+        attributesBuilder.setFileMode(permissionToMode(permission, entry.getIsDirectory()));
+
+        entryBuilder.setAttributes(attributesBuilder);
+
+        LOG.debug("setPermission path:{} entry:{}", path, entryBuilder);
+
+        filerClient.updateEntry(getParentDirectory(path), entryBuilder.build());
+
+    }
+
+}

other/java/hdfs3/src/main/java/seaweed/hdfs/SeaweedInputStream.java

@@ -0,0 +1,371 @@
+package seaweed.hdfs;
+
+// based on org.apache.hadoop.fs.azurebfs.services.AbfsInputStream
+
+import com.google.common.base.Preconditions;
+import org.apache.hadoop.fs.FSExceptionMessages;
+import org.apache.hadoop.fs.FSInputStream;
+import org.apache.hadoop.fs.FileSystem.Statistics;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+import seaweedfs.client.FilerGrpcClient;
+import seaweedfs.client.FilerProto;
+import seaweedfs.client.SeaweedRead;
+
+import java.io.EOFException;
+import java.io.IOException;
+import java.util.List;
+
+public class SeaweedInputStream extends FSInputStream {
+
+    private static final Logger LOG = LoggerFactory.getLogger(SeaweedInputStream.class);
+
+    private final FilerGrpcClient filerGrpcClient;
+    private final Statistics statistics;
+    private final String path;
+    private final FilerProto.Entry entry;
+    private final List<SeaweedRead.VisibleInterval> visibleIntervalList;
+    private final long contentLength;
+    private final int bufferSize; // default buffer size
+    private final int readAheadQueueDepth;         // initialized in constructor
+    private final boolean readAheadEnabled; // whether enable readAhead;
+
+    private byte[] buffer = null;            // will be initialized on first use
+
+    private long fCursor = 0;  // cursor of buffer within file - offset of next byte to read from remote server
+    private long fCursorAfterLastRead = -1;
+    private int bCursor = 0;   // cursor of read within buffer - offset of next byte to be returned from buffer
+    private int limit = 0;     // offset of next byte to be read into buffer from service (i.e., upper marker+1
+    //                                                      of valid bytes in buffer)
+    private boolean closed = false;
+
+    public SeaweedInputStream(
+        final FilerGrpcClient filerGrpcClient,
+        final Statistics statistics,
+        final String path,
+        final FilerProto.Entry entry,
+        final int bufferSize,
+        final int readAheadQueueDepth) {
+        this.filerGrpcClient = filerGrpcClient;
+        this.statistics = statistics;
+        this.path = path;
+        this.entry = entry;
+        this.contentLength = SeaweedRead.totalSize(entry.getChunksList());
+        this.bufferSize = bufferSize;
+        this.readAheadQueueDepth = (readAheadQueueDepth >= 0) ? readAheadQueueDepth : Runtime.getRuntime().availableProcessors();
+        this.readAheadEnabled = true;
+
+        this.visibleIntervalList = SeaweedRead.nonOverlappingVisibleIntervals(entry.getChunksList());
+
+        LOG.debug("new path:{} entry:{} visibleIntervalList:{}", path, entry, visibleIntervalList);
+
+    }
+
+    public String getPath() {
+        return path;
+    }
+
+    @Override
+    public int read() throws IOException {
+        byte[] b = new byte[1];
+        int numberOfBytesRead = read(b, 0, 1);
+        if (numberOfBytesRead < 0) {
+            return -1;
+        } else {
+            return (b[0] & 0xFF);
+        }
+    }
+
+    @Override
+    public synchronized int read(final byte[] b, final int off, final int len) throws IOException {
+        int currentOff = off;
+        int currentLen = len;
+        int lastReadBytes;
+        int totalReadBytes = 0;
+        do {
+            lastReadBytes = readOneBlock(b, currentOff, currentLen);
+            if (lastReadBytes > 0) {
+                currentOff += lastReadBytes;
+                currentLen -= lastReadBytes;
+                totalReadBytes += lastReadBytes;
+            }
+            if (currentLen <= 0 || currentLen > b.length - currentOff) {
+                break;
+            }
+        } while (lastReadBytes > 0);
+        return totalReadBytes > 0 ? totalReadBytes : lastReadBytes;
+    }
+
+    private int readOneBlock(final byte[] b, final int off, final int len) throws IOException {
+        if (closed) {
+            throw new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+
+        Preconditions.checkNotNull(b);
+
+        if (len == 0) {
+            return 0;
+        }
+
+        if (this.available() == 0) {
+            return -1;
+        }
+
+        if (off < 0 || len < 0 || len > b.length - off) {
+            throw new IndexOutOfBoundsException();
+        }
+
+        //If buffer is empty, then fill the buffer.
+        if (bCursor == limit) {
+            //If EOF, then return -1
+            if (fCursor >= contentLength) {
+                return -1;
+            }
+
+            long bytesRead = 0;
+            //reset buffer to initial state - i.e., throw away existing data
+            bCursor = 0;
+            limit = 0;
+            if (buffer == null) {
+                buffer = new byte[bufferSize];
+            }
+
+            // Enable readAhead when reading sequentially
+            if (-1 == fCursorAfterLastRead || fCursorAfterLastRead == fCursor || b.length >= bufferSize) {
+                bytesRead = readInternal(fCursor, buffer, 0, bufferSize, false);
+            } else {
+                bytesRead = readInternal(fCursor, buffer, 0, b.length, true);
+            }
+
+            if (bytesRead == -1) {
+                return -1;
+            }
+
+            limit += bytesRead;
+            fCursor += bytesRead;
+            fCursorAfterLastRead = fCursor;
+        }
+
+        //If there is anything in the buffer, then return lesser of (requested bytes) and (bytes in buffer)
+        //(bytes returned may be less than requested)
+        int bytesRemaining = limit - bCursor;
+        int bytesToRead = Math.min(len, bytesRemaining);
+        System.arraycopy(buffer, bCursor, b, off, bytesToRead);
+        bCursor += bytesToRead;
+        if (statistics != null) {
+            statistics.incrementBytesRead(bytesToRead);
+        }
+        return bytesToRead;
+    }
+
+
+    private int readInternal(final long position, final byte[] b, final int offset, final int length,
+                             final boolean bypassReadAhead) throws IOException {
+        if (readAheadEnabled && !bypassReadAhead) {
+            // try reading from read-ahead
+            if (offset != 0) {
+                throw new IllegalArgumentException("readahead buffers cannot have non-zero buffer offsets");
+            }
+            int receivedBytes;
+
+            // queue read-aheads
+            int numReadAheads = this.readAheadQueueDepth;
+            long nextSize;
+            long nextOffset = position;
+            while (numReadAheads > 0 && nextOffset < contentLength) {
+                nextSize = Math.min((long) bufferSize, contentLength - nextOffset);
+                ReadBufferManager.getBufferManager().queueReadAhead(this, nextOffset, (int) nextSize);
+                nextOffset = nextOffset + nextSize;
+                numReadAheads--;
+            }
+
+            // try reading from buffers first
+            receivedBytes = ReadBufferManager.getBufferManager().getBlock(this, position, length, b);
+            if (receivedBytes > 0) {
+                return receivedBytes;
+            }
+
+            // got nothing from read-ahead, do our own read now
+            receivedBytes = readRemote(position, b, offset, length);
+            return receivedBytes;
+        } else {
+            return readRemote(position, b, offset, length);
+        }
+    }
+
+    int readRemote(long position, byte[] b, int offset, int length) throws IOException {
+        if (position < 0) {
+            throw new IllegalArgumentException("attempting to read from negative offset");
+        }
+        if (position >= contentLength) {
+            return -1;  // Hadoop prefers -1 to EOFException
+        }
+        if (b == null) {
+            throw new IllegalArgumentException("null byte array passed in to read() method");
+        }
+        if (offset >= b.length) {
+            throw new IllegalArgumentException("offset greater than length of array");
+        }
+        if (length < 0) {
+            throw new IllegalArgumentException("requested read length is less than zero");
+        }
+        if (length > (b.length - offset)) {
+            throw new IllegalArgumentException("requested read length is more than will fit after requested offset in buffer");
+        }
+
+        long bytesRead = SeaweedRead.read(filerGrpcClient, visibleIntervalList, position, b, offset, length);
+        if (bytesRead > Integer.MAX_VALUE) {
+            throw new IOException("Unexpected Content-Length");
+        }
+        return (int) bytesRead;
+    }
+
+    /**
+     * Seek to given position in stream.
+     *
+     * @param n position to seek to
+     * @throws IOException  if there is an error
+     * @throws EOFException if attempting to seek past end of file
+     */
+    @Override
+    public synchronized void seek(long n) throws IOException {
+        if (closed) {
+            throw new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+        if (n < 0) {
+            throw new EOFException(FSExceptionMessages.NEGATIVE_SEEK);
+        }
+        if (n > contentLength) {
+            throw new EOFException(FSExceptionMessages.CANNOT_SEEK_PAST_EOF);
+        }
+
+        if (n >= fCursor - limit && n <= fCursor) { // within buffer
+            bCursor = (int) (n - (fCursor - limit));
+            return;
+        }
+
+        // next read will read from here
+        fCursor = n;
+
+        //invalidate buffer
+        limit = 0;
+        bCursor = 0;
+    }
+
+    @Override
+    public synchronized long skip(long n) throws IOException {
+        if (closed) {
+            throw new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+        long currentPos = getPos();
+        if (currentPos == contentLength) {
+            if (n > 0) {
+                throw new EOFException(FSExceptionMessages.CANNOT_SEEK_PAST_EOF);
+            }
+        }
+        long newPos = currentPos + n;
+        if (newPos < 0) {
+            newPos = 0;
+            n = newPos - currentPos;
+        }
+        if (newPos > contentLength) {
+            newPos = contentLength;
+            n = newPos - currentPos;
+        }
+        seek(newPos);
+        return n;
+    }
+
+    /**
+     * Return the size of the remaining available bytes
+     * if the size is less than or equal to {@link Integer#MAX_VALUE},
+     * otherwise, return {@link Integer#MAX_VALUE}.
+     * <p>
+     * This is to match the behavior of DFSInputStream.available(),
+     * which some clients may rely on (HBase write-ahead log reading in
+     * particular).
+     */
+    @Override
+    public synchronized int available() throws IOException {
+        if (closed) {
+            throw new IOException(
+                FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+        final long remaining = this.contentLength - this.getPos();
+        return remaining <= Integer.MAX_VALUE
+            ? (int) remaining : Integer.MAX_VALUE;
+    }
+
+    /**
+     * Returns the length of the file that this stream refers to. Note that the length returned is the length
+     * as of the time the Stream was opened. Specifically, if there have been subsequent appends to the file,
+     * they wont be reflected in the returned length.
+     *
+     * @return length of the file.
+     * @throws IOException if the stream is closed
+     */
+    public long length() throws IOException {
+        if (closed) {
+            throw new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+        return contentLength;
+    }
+
+    /**
+     * Return the current offset from the start of the file
+     *
+     * @throws IOException throws {@link IOException} if there is an error
+     */
+    @Override
+    public synchronized long getPos() throws IOException {
+        if (closed) {
+            throw new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+        }
+        return fCursor - limit + bCursor;
+    }
+
+    /**
+     * Seeks a different copy of the data.  Returns true if
+     * found a new source, false otherwise.
+     *
+     * @throws IOException throws {@link IOException} if there is an error
+     */
+    @Override
+    public boolean seekToNewSource(long l) throws IOException {
+        return false;
+    }
+
+    @Override
+    public synchronized void close() throws IOException {
+        closed = true;
+        buffer = null; // de-reference the buffer so it can be GC'ed sooner
+    }
+
+    /**
+     * Not supported by this stream. Throws {@link UnsupportedOperationException}
+     *
+     * @param readlimit ignored
+     */
+    @Override
+    public synchronized void mark(int readlimit) {
+        throw new UnsupportedOperationException("mark()/reset() not supported on this stream");
+    }
+
+    /**
+     * Not supported by this stream. Throws {@link UnsupportedOperationException}
+     */
+    @Override
+    public synchronized void reset() throws IOException {
+        throw new UnsupportedOperationException("mark()/reset() not supported on this stream");
+    }
+
+    /**
+     * gets whether mark and reset are supported by {@code ADLFileInputStream}. Always returns false.
+     *
+     * @return always {@code false}
+     */
+    @Override
+    public boolean markSupported() {
+        return false;
+    }
+}

other/java/hdfs3/src/main/java/seaweed/hdfs/SeaweedOutputStream.java

@@ -0,0 +1,335 @@
+package seaweed.hdfs;
+
+// adapted from org.apache.hadoop.fs.azurebfs.services.AbfsOutputStream
+
+import com.google.common.base.Preconditions;
+import org.apache.hadoop.fs.FSExceptionMessages;
+import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.fs.StreamCapabilities;
+import org.apache.hadoop.fs.Syncable;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+import seaweedfs.client.FilerGrpcClient;
+import seaweedfs.client.FilerProto;
+import seaweedfs.client.SeaweedWrite;
+
+import java.io.IOException;
+import java.io.InterruptedIOException;
+import java.io.OutputStream;
+import java.util.Locale;
+import java.util.concurrent.Callable;
+import java.util.concurrent.ConcurrentLinkedDeque;
+import java.util.concurrent.ExecutorCompletionService;
+import java.util.concurrent.Future;
+import java.util.concurrent.LinkedBlockingQueue;
+import java.util.concurrent.ThreadPoolExecutor;
+import java.util.concurrent.TimeUnit;
+
+import static seaweed.hdfs.SeaweedFileSystemStore.getParentDirectory;
+
+public class SeaweedOutputStream extends OutputStream implements Syncable, StreamCapabilities {
+
+    private static final Logger LOG = LoggerFactory.getLogger(SeaweedOutputStream.class);
+
+    private final FilerGrpcClient filerGrpcClient;
+    private final Path path;
+    private final int bufferSize;
+    private final int maxConcurrentRequestCount;
+    private final ThreadPoolExecutor threadExecutor;
+    private final ExecutorCompletionService<Void> completionService;
+    private FilerProto.Entry.Builder entry;
+    private long position;
+    private boolean closed;
+    private boolean supportFlush = true;
+    private volatile IOException lastError;
+    private long lastFlushOffset;
+    private long lastTotalAppendOffset = 0;
+    private byte[] buffer;
+    private int bufferIndex;
+    private ConcurrentLinkedDeque<WriteOperation> writeOperations;
+    private String replication = "000";
+
+    public SeaweedOutputStream(FilerGrpcClient filerGrpcClient, final Path path, FilerProto.Entry.Builder entry,
+                               final long position, final int bufferSize, final String replication) {
+        this.filerGrpcClient = filerGrpcClient;
+        this.replication = replication;
+        this.path = path;
+        this.position = position;
+        this.closed = false;
+        this.lastError = null;
+        this.lastFlushOffset = 0;
+        this.bufferSize = bufferSize;
+        this.buffer = new byte[bufferSize];
+        this.bufferIndex = 0;
+        this.writeOperations = new ConcurrentLinkedDeque<>();
+
+        this.maxConcurrentRequestCount = 4 * Runtime.getRuntime().availableProcessors();
+
+        this.threadExecutor
+            = new ThreadPoolExecutor(maxConcurrentRequestCount,
+            maxConcurrentRequestCount,
+            10L,
+            TimeUnit.SECONDS,
+            new LinkedBlockingQueue<Runnable>());
+        this.completionService = new ExecutorCompletionService<>(this.threadExecutor);
+
+        this.entry = entry;
+
+    }
+
+    private synchronized void flushWrittenBytesToServiceInternal(final long offset) throws IOException {
+
+        LOG.debug("SeaweedWrite.writeMeta path: {} entry:{}", path, entry);
+
+        try {
+            SeaweedWrite.writeMeta(filerGrpcClient, getParentDirectory(path), entry);
+        } catch (Exception ex) {
+            throw new IOException(ex);
+        }
+        this.lastFlushOffset = offset;
+    }
+
+    @Override
+    public void write(final int byteVal) throws IOException {
+        write(new byte[]{(byte) (byteVal & 0xFF)});
+    }
+
+    @Override
+    public synchronized void write(final byte[] data, final int off, final int length)
+        throws IOException {
+        maybeThrowLastError();
+
+        Preconditions.checkArgument(data != null, "null data");
+
+        if (off < 0 || length < 0 || length > data.length - off) {
+            throw new IndexOutOfBoundsException();
+        }
+
+        int currentOffset = off;
+        int writableBytes = bufferSize - bufferIndex;
+        int numberOfBytesToWrite = length;
+
+        while (numberOfBytesToWrite > 0) {
+            if (writableBytes <= numberOfBytesToWrite) {
+                System.arraycopy(data, currentOffset, buffer, bufferIndex, writableBytes);
+                bufferIndex += writableBytes;
+                writeCurrentBufferToService();
+                currentOffset += writableBytes;
+                numberOfBytesToWrite = numberOfBytesToWrite - writableBytes;
+            } else {
+                System.arraycopy(data, currentOffset, buffer, bufferIndex, numberOfBytesToWrite);
+                bufferIndex += numberOfBytesToWrite;
+                numberOfBytesToWrite = 0;
+            }
+
+            writableBytes = bufferSize - bufferIndex;
+        }
+    }
+
+    /**
+     * Flushes this output stream and forces any buffered output bytes to be
+     * written out. If any data remains in the payload it is committed to the
+     * service. Data is queued for writing and forced out to the service
+     * before the call returns.
+     */
+    @Override
+    public void flush() throws IOException {
+        if (supportFlush) {
+            flushInternalAsync();
+        }
+    }
+
+    /**
+     * Similar to posix fsync, flush out the data in client's user buffer
+     * all the way to the disk device (but the disk may have it in its cache).
+     *
+     * @throws IOException if error occurs
+     */
+    @Override
+    public void hsync() throws IOException {
+        if (supportFlush) {
+            flushInternal();
+        }
+    }
+
+    /**
+     * Flush out the data in client's user buffer. After the return of
+     * this call, new readers will see the data.
+     *
+     * @throws IOException if any error occurs
+     */
+    @Override
+    public void hflush() throws IOException {
+        if (supportFlush) {
+            flushInternal();
+        }
+    }
+
+    /**
+     * Query the stream for a specific capability.
+     *
+     * @param capability string to query the stream support for.
+     * @return true for hsync and hflush.
+     */
+    @Override
+    public boolean hasCapability(String capability) {
+        switch (capability.toLowerCase(Locale.ENGLISH)) {
+            case StreamCapabilities.HSYNC:
+            case StreamCapabilities.HFLUSH:
+                return supportFlush;
+            default:
+                return false;
+        }
+    }
+
+    /**
+     * Force all data in the output stream to be written to Azure storage.
+     * Wait to return until this is complete. Close the access to the stream and
+     * shutdown the upload thread pool.
+     * If the blob was created, its lease will be released.
+     * Any error encountered caught in threads and stored will be rethrown here
+     * after cleanup.
+     */
+    @Override
+    public synchronized void close() throws IOException {
+        if (closed) {
+            return;
+        }
+
+        LOG.debug("close path: {}", path);
+        try {
+            flushInternal();
+            threadExecutor.shutdown();
+        } finally {
+            lastError = new IOException(FSExceptionMessages.STREAM_IS_CLOSED);
+            buffer = null;
+            bufferIndex = 0;
+            closed = true;
+            writeOperations.clear();
+            if (!threadExecutor.isShutdown()) {
+                threadExecutor.shutdownNow();
+            }
+        }
+    }
+
+    private synchronized void writeCurrentBufferToService() throws IOException {
+        if (bufferIndex == 0) {
+            return;
+        }
+
+        final byte[] bytes = buffer;
+        final int bytesLength = bufferIndex;
+
+        buffer = new byte[bufferSize];
+        bufferIndex = 0;
+        final long offset = position;
+        position += bytesLength;
+
+        if (threadExecutor.getQueue().size() >= maxConcurrentRequestCount * 2) {
+            waitForTaskToComplete();
+        }
+
+        final Future<Void> job = completionService.submit(new Callable<Void>() {
+            @Override
+            public Void call() throws Exception {
+                // originally: client.append(path, offset, bytes, 0, bytesLength);
+                SeaweedWrite.writeData(entry, replication, filerGrpcClient, offset, bytes, 0, bytesLength);
+                return null;
+            }
+        });
+
+        writeOperations.add(new WriteOperation(job, offset, bytesLength));
+
+        // Try to shrink the queue
+        shrinkWriteOperationQueue();
+    }
+
+    private void waitForTaskToComplete() throws IOException {
+        boolean completed;
+        for (completed = false; completionService.poll() != null; completed = true) {
+            // keep polling until there is no data
+        }
+
+        if (!completed) {
+            try {
+                completionService.take();
+            } catch (InterruptedException e) {
+                lastError = (IOException) new InterruptedIOException(e.toString()).initCause(e);
+                throw lastError;
+            }
+        }
+    }
+
+    private void maybeThrowLastError() throws IOException {
+        if (lastError != null) {
+            throw lastError;
+        }
+    }
+
+    /**
+     * Try to remove the completed write operations from the beginning of write
+     * operation FIFO queue.
+     */
+    private synchronized void shrinkWriteOperationQueue() throws IOException {
+        try {
+            while (writeOperations.peek() != null && writeOperations.peek().task.isDone()) {
+                writeOperations.peek().task.get();
+                lastTotalAppendOffset += writeOperations.peek().length;
+                writeOperations.remove();
+            }
+        } catch (Exception e) {
+            lastError = new IOException(e);
+            throw lastError;
+        }
+    }
+
+    private synchronized void flushInternal() throws IOException {
+        maybeThrowLastError();
+        writeCurrentBufferToService();
+        flushWrittenBytesToService();
+    }
+
+    private synchronized void flushInternalAsync() throws IOException {
+        maybeThrowLastError();
+        writeCurrentBufferToService();
+        flushWrittenBytesToServiceAsync();
+    }
+
+    private synchronized void flushWrittenBytesToService() throws IOException {
+        for (WriteOperation writeOperation : writeOperations) {
+            try {
+                writeOperation.task.get();
+            } catch (Exception ex) {
+                lastError = new IOException(ex);
+                throw lastError;
+            }
+        }
+        LOG.debug("flushWrittenBytesToService: {} position:{}", path, position);
+        flushWrittenBytesToServiceInternal(position);
+    }
+
+    private synchronized void flushWrittenBytesToServiceAsync() throws IOException {
+        shrinkWriteOperationQueue();
+
+        if (this.lastTotalAppendOffset > this.lastFlushOffset) {
+            this.flushWrittenBytesToServiceInternal(this.lastTotalAppendOffset);
+        }
+    }
+
+    private static class WriteOperation {
+        private final Future<Void> task;
+        private final long startOffset;
+        private final long length;
+
+        WriteOperation(final Future<Void> task, final long startOffset, final long length) {
+            Preconditions.checkNotNull(task, "task");
+            Preconditions.checkArgument(startOffset >= 0, "startOffset");
+            Preconditions.checkArgument(length >= 0, "length");
+
+            this.task = task;
+            this.startOffset = startOffset;
+            this.length = length;
+        }
+    }
+
+}

unmaintained/change_superblock/change_superblock.go

@@ -8,7 +8,9 @@ import (
 	"strconv"
 
 	"github.com/chrislusf/seaweedfs/weed/glog"
-	"github.com/chrislusf/seaweedfs/weed/storage"
+	"github.com/chrislusf/seaweedfs/weed/storage/backend"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle"
+	"github.com/chrislusf/seaweedfs/weed/storage/super_block"
 )
 
 var (
@@ -46,9 +48,10 @@ func main() {
 	if err != nil {
 		glog.Fatalf("Open Volume Data File [ERROR]: %v", err)
 	}
-	defer datFile.Close()
+	datBackend := backend.NewDiskFile(datFile)
+	defer datBackend.Close()
 
-	superBlock, err := storage.ReadSuperBlock(datFile)
+	superBlock, err := super_block.ReadSuperBlock(datBackend)
 
 	if err != nil {
 		glog.Fatalf("cannot parse existing super block: %v", err)
@@ -60,7 +63,7 @@ func main() {
 	hasChange := false
 
 	if *targetReplica != "" {
-		replica, err := storage.NewReplicaPlacementFromString(*targetReplica)
+		replica, err := super_block.NewReplicaPlacementFromString(*targetReplica)
 
 		if err != nil {
 			glog.Fatalf("cannot parse target replica %s: %v", *targetReplica, err)
@@ -73,7 +76,7 @@ func main() {
 	}
 
 	if *targetTTL != "" {
-		ttl, err := storage.ReadTTL(*targetTTL)
+		ttl, err := needle.ReadTTL(*targetTTL)
 
 		if err != nil {
 			glog.Fatalf("cannot parse target ttl %s: %v", *targetTTL, err)

unmaintained/compact_leveldb/compact_leveldb.go

@@ -0,0 +1,35 @@
+package main
+
+import (
+	"flag"
+	"log"
+
+	"github.com/syndtr/goleveldb/leveldb"
+	"github.com/syndtr/goleveldb/leveldb/opt"
+	"github.com/syndtr/goleveldb/leveldb/util"
+)
+
+var (
+	dir = flag.String("dir", ".", "data directory to store leveldb files")
+)
+
+func main() {
+
+	flag.Parse()
+
+	opts := &opt.Options{
+		BlockCacheCapacity:            32 * 1024 * 1024, // default value is 8MiB
+		WriteBuffer:                   16 * 1024 * 1024, // default value is 4MiB
+		CompactionTableSizeMultiplier: 10,
+		OpenFilesCacheCapacity:        -1,
+	}
+
+	db, err := leveldb.OpenFile(*dir, opts)
+	if err != nil {
+		log.Fatal(err)
+	}
+	defer db.Close()
+	if err := db.CompactRange(util.Range{}); err != nil {
+		log.Fatal(err)
+	}
+}

unmaintained/fix_dat/fix_dat.go

@@ -9,9 +9,11 @@ import (
 	"strconv"
 
 	"github.com/chrislusf/seaweedfs/weed/glog"
-	"github.com/chrislusf/seaweedfs/weed/storage"
-	"github.com/chrislusf/seaweedfs/weed/util"
+	"github.com/chrislusf/seaweedfs/weed/storage/backend"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle"
+	"github.com/chrislusf/seaweedfs/weed/storage/super_block"
 	"github.com/chrislusf/seaweedfs/weed/storage/types"
+	"github.com/chrislusf/seaweedfs/weed/util"
 )
 
 var (
@@ -43,11 +45,13 @@ func main() {
 		glog.Fatalf("Read Volume Index %v", err)
 	}
 	defer indexFile.Close()
-	datFile, err := os.OpenFile(path.Join(*fixVolumePath, fileName+".dat"), os.O_RDONLY, 0644)
+	datFileName := path.Join(*fixVolumePath, fileName+".dat")
+	datFile, err := os.OpenFile(datFileName, os.O_RDONLY, 0644)
 	if err != nil {
 		glog.Fatalf("Read Volume Data %v", err)
 	}
-	defer datFile.Close()
+	datBackend := backend.NewDiskFile(datFile)
+	defer datBackend.Close()
 
 	newDatFile, err := os.Create(path.Join(*fixVolumePath, fileName+".dat_fixed"))
 	if err != nil {
@@ -55,21 +59,21 @@ func main() {
 	}
 	defer newDatFile.Close()
 
-	superBlock, err := storage.ReadSuperBlock(datFile)
+	superBlock, err := super_block.ReadSuperBlock(datBackend)
 	if err != nil {
 		glog.Fatalf("Read Volume Data superblock %v", err)
 	}
 	newDatFile.Write(superBlock.Bytes())
 
-	iterateEntries(datFile, indexFile, func(n *storage.Needle, offset int64) {
+	iterateEntries(datBackend, indexFile, func(n *needle.Needle, offset int64) {
 		fmt.Printf("needle id=%v name=%s size=%d dataSize=%d\n", n.Id, string(n.Name), n.Size, n.DataSize)
-		s, _, e := n.Append(newDatFile, superBlock.Version())
+		_, s, _, e := n.Append(datBackend, superBlock.Version)
 		fmt.Printf("size %d error %v\n", s, e)
 	})
 
 }
 
-func iterateEntries(datFile, idxFile *os.File, visitNeedle func(n *storage.Needle, offset int64)) {
+func iterateEntries(datBackend backend.BackendStorageFile, idxFile *os.File, visitNeedle func(n *needle.Needle, offset int64)) {
 	// start to read index file
 	var readerOffset int64
 	bytes := make([]byte, 16)
@@ -77,14 +81,14 @@ func iterateEntries(datFile, idxFile *os.File, visitNeedle func(n *storage.Needl
 	readerOffset += int64(count)
 
 	// start to read dat file
-	superblock, err := storage.ReadSuperBlock(datFile)
+	superBlock, err := super_block.ReadSuperBlock(datBackend)
 	if err != nil {
 		fmt.Printf("cannot read dat file super block: %v", err)
 		return
 	}
-	offset := int64(superblock.BlockSize())
-	version := superblock.Version()
-	n, rest, err := storage.ReadNeedleHeader(datFile, version, offset)
+	offset := int64(superBlock.BlockSize())
+	version := superBlock.Version
+	n, _, rest, err := needle.ReadNeedleHeader(datBackend, version, offset)
 	if err != nil {
 		fmt.Printf("cannot read needle header: %v", err)
 		return
@@ -106,7 +110,7 @@ func iterateEntries(datFile, idxFile *os.File, visitNeedle func(n *storage.Needl
 
 		fmt.Printf("key: %d offsetFromIndex %d n.Size %d sizeFromIndex:%d\n", key, offsetFromIndex, n.Size, sizeFromIndex)
 
-		rest = storage.NeedleBodyLength(sizeFromIndex, version)
+		rest = needle.NeedleBodyLength(sizeFromIndex, version)
 
 		func() {
 			defer func() {
@@ -114,7 +118,7 @@ func iterateEntries(datFile, idxFile *os.File, visitNeedle func(n *storage.Needl
 					fmt.Println("Recovered in f", r)
 				}
 			}()
-			if err = n.ReadNeedleBody(datFile, version, offset+int64(types.NeedleEntrySize), rest); err != nil {
+			if _, err = n.ReadNeedleBody(datBackend, version, offset+int64(types.NeedleHeaderSize), rest); err != nil {
 				fmt.Printf("cannot read needle body: offset %d body %d %v\n", offset, rest, err)
 			}
 		}()
@@ -124,9 +128,9 @@ func iterateEntries(datFile, idxFile *os.File, visitNeedle func(n *storage.Needl
 		}
 		visitNeedle(n, offset)
 
-		offset += types.NeedleEntrySize + rest
+		offset += types.NeedleHeaderSize + rest
 		//fmt.Printf("==> new entry offset %d\n", offset)
-		if n, rest, err = storage.ReadNeedleHeader(datFile, version, offset); err != nil {
+		if n, _, rest, err = needle.ReadNeedleHeader(datBackend, version, offset); err != nil {
 			if err == io.EOF {
 				return
 			}

unmaintained/load_test/load_test_leveldb/load_test_leveldb.go

@@ -0,0 +1,155 @@
+package load_test_leveldb
+
+import (
+	"crypto/md5"
+	"flag"
+	"fmt"
+	"io"
+	"log"
+	"math/rand"
+	"os"
+	"sync"
+	"time"
+
+	"github.com/syndtr/goleveldb/leveldb"
+	"github.com/syndtr/goleveldb/leveldb/opt"
+)
+
+var (
+	dir     = flag.String("dir", "./t", "directory to store level db files")
+	useHash = flag.Bool("isHash", false, "hash the path as the key")
+	dbCount = flag.Int("dbCount", 1, "the number of leveldb")
+)
+
+func main() {
+
+	flag.Parse()
+
+	totalTenants := 300
+	totalYears := 3
+
+	opts := &opt.Options{
+		BlockCacheCapacity:            32 * 1024 * 1024, // default value is 8MiB
+		WriteBuffer:                   16 * 1024 * 1024, // default value is 4MiB
+		CompactionTableSizeMultiplier: 4,
+	}
+
+	var dbs []*leveldb.DB
+	var chans []chan string
+	for d := 0; d < *dbCount; d++ {
+		dbFolder := fmt.Sprintf("%s/%02d", *dir, d)
+		os.MkdirAll(dbFolder, 0755)
+		db, err := leveldb.OpenFile(dbFolder, opts)
+		if err != nil {
+			log.Printf("filer store open dir %s: %v", *dir, err)
+			return
+		}
+		dbs = append(dbs, db)
+		chans = append(chans, make(chan string, 1024))
+	}
+
+	var wg sync.WaitGroup
+	for d := 0; d < *dbCount; d++ {
+		wg.Add(1)
+		go func(d int) {
+			defer wg.Done()
+
+			ch := chans[d]
+			db := dbs[d]
+
+			for p := range ch {
+				if *useHash {
+					insertAsHash(db, p)
+				} else {
+					insertAsFullPath(db, p)
+				}
+			}
+		}(d)
+	}
+
+	counter := int64(0)
+	lastResetTime := time.Now()
+
+	r := rand.New(rand.NewSource(35))
+
+	for y := 0; y < totalYears; y++ {
+		for m := 0; m < 12; m++ {
+			for d := 0; d < 31; d++ {
+				for h := 0; h < 24; h++ {
+					for min := 0; min < 60; min++ {
+						for i := 0; i < totalTenants; i++ {
+							p := fmt.Sprintf("tenent%03d/%4d/%02d/%02d/%02d/%02d", i, 2015+y, 1+m, 1+d, h, min)
+
+							x := r.Intn(*dbCount)
+
+							chans[x] <- p
+
+							counter++
+						}
+
+						t := time.Now()
+						if lastResetTime.Add(time.Second).Before(t) {
+							p := fmt.Sprintf("%4d/%02d/%02d/%02d/%02d", 2015+y, 1+m, 1+d, h, min)
+							fmt.Printf("%s = %4d put/sec\n", p, counter)
+							counter = 0
+							lastResetTime = t
+						}
+					}
+				}
+			}
+		}
+	}
+
+	for d := 0; d < *dbCount; d++ {
+		close(chans[d])
+	}
+
+	wg.Wait()
+
+}
+
+func insertAsFullPath(db *leveldb.DB, p string) {
+	_, getErr := db.Get([]byte(p), nil)
+	if getErr == leveldb.ErrNotFound {
+		putErr := db.Put([]byte(p), []byte(p), nil)
+		if putErr != nil {
+			log.Printf("failed to put %s", p)
+		}
+	}
+}
+
+func insertAsHash(db *leveldb.DB, p string) {
+	key := fmt.Sprintf("%d:%s", hashToLong(p), p)
+	_, getErr := db.Get([]byte(key), nil)
+	if getErr == leveldb.ErrNotFound {
+		putErr := db.Put([]byte(key), []byte(p), nil)
+		if putErr != nil {
+			log.Printf("failed to put %s", p)
+		}
+	}
+}
+
+func hashToLong(dir string) (v int64) {
+	h := md5.New()
+	io.WriteString(h, dir)
+
+	b := h.Sum(nil)
+
+	v += int64(b[0])
+	v <<= 8
+	v += int64(b[1])
+	v <<= 8
+	v += int64(b[2])
+	v <<= 8
+	v += int64(b[3])
+	v <<= 8
+	v += int64(b[4])
+	v <<= 8
+	v += int64(b[5])
+	v <<= 8
+	v += int64(b[6])
+	v <<= 8
+	v += int64(b[7])
+
+	return
+}

unmaintained/remove_duplicate_fids/remove_duplicate_fids.go

@@ -0,0 +1,95 @@
+package main
+
+import (
+	"flag"
+	"fmt"
+	"os"
+	"path/filepath"
+
+	"github.com/chrislusf/seaweedfs/weed/glog"
+	"github.com/chrislusf/seaweedfs/weed/storage"
+	"github.com/chrislusf/seaweedfs/weed/storage/backend"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle"
+	"github.com/chrislusf/seaweedfs/weed/storage/super_block"
+)
+
+var (
+	volumePath       = flag.String("dir", "/tmp", "data directory to store files")
+	volumeCollection = flag.String("collection", "", "the volume collection name")
+	volumeId         = flag.Int("volumeId", -1, "a volume id. The volume should already exist in the dir. The volume index file should not exist.")
+)
+
+func Checksum(n *needle.Needle) string {
+	return fmt.Sprintf("%s%x", n.Id, n.Cookie)
+}
+
+type VolumeFileScanner4SeeDat struct {
+	version needle.Version
+	block   super_block.SuperBlock
+
+	dir        string
+	hashes     map[string]bool
+	dat        *os.File
+	datBackend backend.BackendStorageFile
+}
+
+func (scanner *VolumeFileScanner4SeeDat) VisitSuperBlock(superBlock super_block.SuperBlock) error {
+	scanner.version = superBlock.Version
+	scanner.block = superBlock
+	return nil
+
+}
+func (scanner *VolumeFileScanner4SeeDat) ReadNeedleBody() bool {
+	return true
+}
+
+func (scanner *VolumeFileScanner4SeeDat) VisitNeedle(n *needle.Needle, offset int64, needleHeader, needleBody []byte) error {
+
+	if scanner.datBackend == nil {
+		newFileName := filepath.Join(*volumePath, "dat_fixed")
+		newDatFile, err := os.Create(newFileName)
+		if err != nil {
+			glog.Fatalf("Write New Volume Data %v", err)
+		}
+		scanner.datBackend = backend.NewDiskFile(newDatFile)
+		scanner.datBackend.WriteAt(scanner.block.Bytes(), 0)
+	}
+
+	checksum := Checksum(n)
+
+	if scanner.hashes[checksum] {
+		glog.V(0).Infof("duplicate checksum:%s fid:%d,%s%x @ offset:%d", checksum, *volumeId, n.Id, n.Cookie, offset)
+		return nil
+	}
+	scanner.hashes[checksum] = true
+
+	_, s, _, e := n.Append(scanner.datBackend, scanner.version)
+	fmt.Printf("size %d error %v\n", s, e)
+
+	return nil
+}
+
+func main() {
+	flag.Parse()
+
+	vid := needle.VolumeId(*volumeId)
+
+	outpath, _ := filepath.Abs(filepath.Dir(os.Args[0]))
+
+	scanner := &VolumeFileScanner4SeeDat{
+		dir:    filepath.Join(outpath, "out"),
+		hashes: map[string]bool{},
+	}
+
+	if _, err := os.Stat(scanner.dir); err != nil {
+		if err := os.MkdirAll(scanner.dir, os.ModePerm); err != nil {
+			glog.Fatalf("could not create output dir : %s", err)
+		}
+	}
+
+	err := storage.ScanVolumeFile(*volumePath, *volumeCollection, vid, storage.NeedleMapInMemory, scanner)
+	if err != nil {
+		glog.Fatalf("Reading Volume File [ERROR] %s\n", err)
+	}
+
+}

unmaintained/repeated_vacuum/repeated_vacuum.go

@@ -7,6 +7,9 @@ import (
 	"log"
 	"math/rand"
 
+	"github.com/chrislusf/seaweedfs/weed/security"
+	"github.com/spf13/viper"
+
 	"github.com/chrislusf/seaweedfs/weed/operation"
 	"github.com/chrislusf/seaweedfs/weed/util"
 )
@@ -19,8 +22,11 @@ var (
 func main() {
 	flag.Parse()
 
+	util.LoadConfiguration("security", false)
+	grpcDialOption := security.LoadClientTLS(viper.Sub("grpc"), "client")
+
 	for i := 0; i < *repeat; i++ {
-		assignResult, err := operation.Assign(*master, &operation.VolumeAssignRequest{Count: 1})
+		assignResult, err := operation.Assign(*master, grpcDialOption, &operation.VolumeAssignRequest{Count: 1})
 		if err != nil {
 			log.Fatalf("assign: %v", err)
 		}
@@ -29,14 +35,14 @@ func main() {
 		rand.Read(data)
 		reader := bytes.NewReader(data)
 
-		targetUrl := fmt.Sprintf("http://%s/%s", assignResult.Url,assignResult.Fid)
+		targetUrl := fmt.Sprintf("http://%s/%s", assignResult.Url, assignResult.Fid)
 
-		_, err = operation.Upload(targetUrl, fmt.Sprintf("test%d", i), reader, false, "", nil, "")
+		_, err = operation.Upload(targetUrl, fmt.Sprintf("test%d", i), reader, false, "", nil, assignResult.Auth)
 		if err != nil {
 			log.Fatalf("upload: %v", err)
 		}
 
-		util.Delete(targetUrl, "")
+		util.Delete(targetUrl, string(assignResult.Auth))
 
 		util.Get(fmt.Sprintf("http://%s/vol/vacuum", *master))
 

unmaintained/see_dat/see_dat.go

@@ -0,0 +1,48 @@
+package main
+
+import (
+	"flag"
+	"time"
+
+	"github.com/chrislusf/seaweedfs/weed/glog"
+	"github.com/chrislusf/seaweedfs/weed/storage"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle"
+	"github.com/chrislusf/seaweedfs/weed/storage/super_block"
+)
+
+var (
+	volumePath       = flag.String("dir", "/tmp", "data directory to store files")
+	volumeCollection = flag.String("collection", "", "the volume collection name")
+	volumeId         = flag.Int("volumeId", -1, "a volume id. The volume should already exist in the dir. The volume index file should not exist.")
+)
+
+type VolumeFileScanner4SeeDat struct {
+	version needle.Version
+}
+
+func (scanner *VolumeFileScanner4SeeDat) VisitSuperBlock(superBlock super_block.SuperBlock) error {
+	scanner.version = superBlock.Version
+	return nil
+
+}
+func (scanner *VolumeFileScanner4SeeDat) ReadNeedleBody() bool {
+	return true
+}
+
+func (scanner *VolumeFileScanner4SeeDat) VisitNeedle(n *needle.Needle, offset int64, needleHeader, needleBody []byte) error {
+	t := time.Unix(int64(n.AppendAtNs)/int64(time.Second), int64(n.AppendAtNs)%int64(time.Second))
+	glog.V(0).Infof("%d,%s%x offset %d size %d cookie %x appendedAt %v", *volumeId, n.Id, n.Cookie, offset, n.Size, n.Cookie, t)
+	return nil
+}
+
+func main() {
+	flag.Parse()
+
+	vid := needle.VolumeId(*volumeId)
+
+	scanner := &VolumeFileScanner4SeeDat{}
+	err := storage.ScanVolumeFile(*volumePath, *volumeCollection, vid, storage.NeedleMapInMemory, scanner)
+	if err != nil {
+		glog.Fatalf("Reading Volume File [ERROR] %s\n", err)
+	}
+}

unmaintained/see_idx/see_idx.go

@@ -8,7 +8,7 @@ import (
 	"strconv"
 
 	"github.com/chrislusf/seaweedfs/weed/glog"
-	"github.com/chrislusf/seaweedfs/weed/storage"
+	"github.com/chrislusf/seaweedfs/weed/storage/idx"
 	"github.com/chrislusf/seaweedfs/weed/storage/types"
 )
 
@@ -35,7 +35,7 @@ func main() {
 	}
 	defer indexFile.Close()
 
-	storage.WalkIndexFile(indexFile, func(key types.NeedleId, offset types.Offset, size uint32) error {
+	idx.WalkIndexFile(indexFile, func(key types.NeedleId, offset types.Offset, size uint32) error {
 		fmt.Printf("key:%v offset:%v size:%v\n", key, offset, size)
 		return nil
 	})

unmaintained/see_meta/see_meta.go

@@ -0,0 +1,68 @@
+package main
+
+import (
+	"flag"
+	"fmt"
+	"io"
+	"log"
+	"os"
+
+	"github.com/chrislusf/seaweedfs/weed/filer2"
+	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
+	"github.com/chrislusf/seaweedfs/weed/util"
+	"github.com/golang/protobuf/proto"
+)
+
+var (
+	metaFile = flag.String("meta", "", "meta file generated via fs.meta.save")
+)
+
+func main() {
+	flag.Parse()
+
+	dst, err := os.OpenFile(*metaFile, os.O_RDONLY, 0644)
+	if err != nil {
+		log.Fatalf("failed to open %s: %v", *metaFile, err)
+	}
+	defer dst.Close()
+
+	err = walkMetaFile(dst)
+	if err != nil {
+		log.Fatalf("failed to visit %s: %v", *metaFile, err)
+	}
+
+}
+
+func walkMetaFile(dst *os.File) error {
+
+	sizeBuf := make([]byte, 4)
+
+	for {
+		if n, err := dst.Read(sizeBuf); n != 4 {
+			if err == io.EOF {
+				return nil
+			}
+			return err
+		}
+
+		size := util.BytesToUint32(sizeBuf)
+
+		data := make([]byte, int(size))
+
+		if n, err := dst.Read(data); n != len(data) {
+			return err
+		}
+
+		fullEntry := &filer_pb.FullEntry{}
+		if err := proto.Unmarshal(data, fullEntry); err != nil {
+			return err
+		}
+
+		fmt.Fprintf(os.Stdout, "file %s %v\n", filer2.FullPath(fullEntry.Dir).Child(fullEntry.Entry.Name), fullEntry.Entry.Attributes.String())
+		for i, chunk := range fullEntry.Entry.Chunks {
+			fmt.Fprintf(os.Stdout, "  chunk %d %v\n", i+1, chunk.String())
+		}
+
+	}
+
+}

unmaintained/stress_filer_upload/bench_filer_upload/bench_filer_upload.go

@@ -0,0 +1,136 @@
+package main
+
+import (
+	"bytes"
+	"flag"
+	"fmt"
+	"io"
+	"io/ioutil"
+	"log"
+	"math/rand"
+	"mime/multipart"
+	"net/http"
+	"os"
+	"strings"
+	"sync"
+	"time"
+)
+
+var (
+	size        = flag.Int("size", 1024, "file size")
+	concurrency = flag.Int("c", 4, "concurrent number of uploads")
+	times       = flag.Int("n", 1024, "repeated number of times")
+	fileCount   = flag.Int("fileCount", 1, "number of files to write")
+	destination = flag.String("to", "http://localhost:8888/", "destination directory on filer")
+
+	statsChan = make(chan stat, 8)
+)
+
+type stat struct {
+	size int64
+}
+
+func main() {
+
+	flag.Parse()
+
+	data := make([]byte, *size)
+	println("data len", len(data))
+
+	var wg sync.WaitGroup
+	for x := 0; x < *concurrency; x++ {
+		wg.Add(1)
+
+		go func(x int) {
+			defer wg.Done()
+
+			client := &http.Client{Transport: &http.Transport{
+				MaxConnsPerHost:     1024,
+				MaxIdleConnsPerHost: 1024,
+			}}
+			r := rand.New(rand.NewSource(time.Now().UnixNano() + int64(x)))
+
+			for t := 0; t < *times; t++ {
+				for f := 0; f < *fileCount; f++ {
+					fn := r.Intn(*fileCount)
+					if size, err := uploadFileToFiler(client, data, fmt.Sprintf("file%04d", fn), *destination); err == nil {
+						statsChan <- stat{
+							size: size,
+						}
+					} else {
+						log.Fatalf("client %d upload %d times: %v", x, t, err)
+					}
+				}
+			}
+		}(x)
+	}
+
+	go func() {
+		ticker := time.NewTicker(1000 * time.Millisecond)
+
+		var lastTime time.Time
+		var counter, size int64
+		for {
+			select {
+			case stat := <-statsChan:
+				size += stat.size
+				counter++
+			case x := <-ticker.C:
+				if !lastTime.IsZero() {
+					elapsed := x.Sub(lastTime).Seconds()
+					fmt.Fprintf(os.Stdout, "%.2f files/s, %.2f MB/s\n",
+						float64(counter)/elapsed,
+						float64(size/1024/1024)/elapsed)
+				}
+				lastTime = x
+				size = 0
+				counter = 0
+			}
+		}
+	}()
+
+	wg.Wait()
+
+}
+
+func uploadFileToFiler(client *http.Client, data []byte, filename, destination string) (size int64, err error) {
+
+	if !strings.HasSuffix(destination, "/") {
+		destination = destination + "/"
+	}
+
+	body := &bytes.Buffer{}
+	writer := multipart.NewWriter(body)
+	part, err := writer.CreateFormFile("file", filename)
+	if err != nil {
+		return 0, fmt.Errorf("fail to create form %v: %v", filename, err)
+	}
+
+	part.Write(data)
+
+	err = writer.Close()
+	if err != nil {
+		return 0, fmt.Errorf("fail to write part %v: %v", filename, err)
+	}
+
+	uri := destination + filename
+
+	request, err := http.NewRequest("POST", uri, body)
+	request.Header.Set("Content-Type", writer.FormDataContentType())
+	// request.Close = true  // can not use this, which do not reuse http connection, impacting filer->volume also.
+
+	resp, err := client.Do(request)
+	if err != nil {
+		return 0, fmt.Errorf("http POST %s: %v", uri, err)
+	} else {
+		body := &bytes.Buffer{}
+		_, err := body.ReadFrom(resp.Body)
+		if err != nil {
+			return 0, fmt.Errorf("read http POST %s response: %v", uri, err)
+		}
+		io.Copy(ioutil.Discard, resp.Body)
+		resp.Body.Close()
+	}
+
+	return int64(len(data)), nil
+}

unmaintained/stress_filer_upload/stress_filer_upload_actual/stress_filer_upload.go

@@ -0,0 +1,150 @@
+package main
+
+import (
+	"bytes"
+	"flag"
+	"fmt"
+	"io"
+	"io/ioutil"
+	"log"
+	"math/rand"
+	"mime/multipart"
+	"net/http"
+	"os"
+	"path/filepath"
+	"strings"
+	"sync"
+	"time"
+)
+
+var (
+	dir         = flag.String("dir", ".", "upload files under this directory")
+	concurrency = flag.Int("c", 1, "concurrent number of uploads")
+	times       = flag.Int("n", 1, "repeated number of times")
+	destination = flag.String("to", "http://localhost:8888/", "destination directory on filer")
+
+	statsChan = make(chan stat, 8)
+)
+
+type stat struct {
+	size int64
+}
+
+func main() {
+
+	flag.Parse()
+
+	var fileNames []string
+
+	files, err := ioutil.ReadDir(*dir)
+	if err != nil {
+		log.Fatalf("fail to read dir %v: %v", *dir, err)
+	}
+
+	for _, file := range files {
+		if file.IsDir() {
+			continue
+		}
+		fileNames = append(fileNames, filepath.Join(*dir, file.Name()))
+	}
+
+	var wg sync.WaitGroup
+	for x := 0; x < *concurrency; x++ {
+		wg.Add(1)
+
+		client := &http.Client{}
+
+		go func() {
+			defer wg.Done()
+			rand.Shuffle(len(fileNames), func(i, j int) {
+				fileNames[i], fileNames[j] = fileNames[j], fileNames[i]
+			})
+			for t := 0; t < *times; t++ {
+				for _, filename := range fileNames {
+					if size, err := uploadFileToFiler(client, filename, *destination); err == nil {
+						statsChan <- stat{
+							size: size,
+						}
+					}
+				}
+			}
+		}()
+	}
+
+	go func() {
+		ticker := time.NewTicker(500 * time.Millisecond)
+
+		var lastTime time.Time
+		var counter, size int64
+		for {
+			select {
+			case stat := <-statsChan:
+				size += stat.size
+				counter++
+			case x := <-ticker.C:
+				if !lastTime.IsZero() {
+					elapsed := x.Sub(lastTime).Seconds()
+					fmt.Fprintf(os.Stdout, "%.2f files/s, %.2f MB/s\n",
+						float64(counter)/elapsed,
+						float64(size/1024/1024)/elapsed)
+				}
+				lastTime = x
+				size = 0
+				counter = 0
+			}
+		}
+	}()
+
+	wg.Wait()
+
+}
+
+func uploadFileToFiler(client *http.Client, filename, destination string) (size int64, err error) {
+	file, err := os.Open(filename)
+	if err != nil {
+		panic(err)
+	}
+	defer file.Close()
+
+	fi, err := file.Stat()
+
+	if !strings.HasSuffix(destination, "/") {
+		destination = destination + "/"
+	}
+
+	body := &bytes.Buffer{}
+	writer := multipart.NewWriter(body)
+	part, err := writer.CreateFormFile("file", file.Name())
+	if err != nil {
+		return 0, fmt.Errorf("fail to create form %v: %v", file.Name(), err)
+	}
+	_, err = io.Copy(part, file)
+	if err != nil {
+		return 0, fmt.Errorf("fail to write part %v: %v", file.Name(), err)
+	}
+
+	err = writer.Close()
+	if err != nil {
+		return 0, fmt.Errorf("fail to write part %v: %v", file.Name(), err)
+	}
+
+	uri := destination + file.Name()
+
+	request, err := http.NewRequest("POST", uri, body)
+	request.Header.Set("Content-Type", writer.FormDataContentType())
+
+	resp, err := client.Do(request)
+	if err != nil {
+		return 0, fmt.Errorf("http POST %s: %v", uri, err)
+	} else {
+		body := &bytes.Buffer{}
+		_, err := body.ReadFrom(resp.Body)
+		if err != nil {
+			return 0, fmt.Errorf("read http POST %s response: %v", uri, err)
+		}
+		io.Copy(ioutil.Discard, resp.Body)
+		resp.Body.Close()
+	}
+
+	return fi.Size(), nil
+}

unmaintained/volume_tailer/volume_tailer.go

@@ -0,0 +1,70 @@
+package main
+
+import (
+	"flag"
+	"log"
+	"time"
+
+	"github.com/chrislusf/seaweedfs/weed/operation"
+	"github.com/chrislusf/seaweedfs/weed/security"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle"
+	util2 "github.com/chrislusf/seaweedfs/weed/util"
+	"github.com/spf13/viper"
+	"golang.org/x/tools/godoc/util"
+)
+
+var (
+	master         = flag.String("master", "localhost:9333", "master server host and port")
+	volumeId       = flag.Int("volumeId", -1, "a volume id")
+	rewindDuration = flag.Duration("rewind", -1, "rewind back in time. -1 means from the first entry. 0 means from now.")
+	timeoutSeconds = flag.Int("timeoutSeconds", 0, "disconnect if no activity after these seconds")
+	showTextFile   = flag.Bool("showTextFile", false, "display textual file content")
+)
+
+func main() {
+	flag.Parse()
+
+	util2.LoadConfiguration("security", false)
+	grpcDialOption := security.LoadClientTLS(viper.Sub("grpc"), "client")
+
+	vid := needle.VolumeId(*volumeId)
+
+	var sinceTimeNs int64
+	if *rewindDuration == 0 {
+		sinceTimeNs = time.Now().UnixNano()
+	} else if *rewindDuration == -1 {
+		sinceTimeNs = 0
+	} else if *rewindDuration > 0 {
+		sinceTimeNs = time.Now().Add(-*rewindDuration).UnixNano()
+	}
+
+	err := operation.TailVolume(*master, grpcDialOption, vid, uint64(sinceTimeNs), *timeoutSeconds, func(n *needle.Needle) (err error) {
+		if n.Size == 0 {
+			println("-", n.String())
+			return nil
+		} else {
+			println("+", n.String())
+		}
+
+		if *showTextFile {
+
+			data := n.Data
+			if n.IsGzipped() {
+				if data, err = util2.UnGzipData(data); err != nil {
+					return err
+				}
+			}
+			if util.IsText(data) {
+				println(string(data))
+			}
+
+			println("-", n.String(), "compressed", n.IsGzipped(), "original size", len(data))
+		}
+		return nil
+	})
+
+	if err != nil {
+		log.Printf("Error VolumeTailSender volume %d: %v", vid, err)
+	}
+
+}

util/gostd

@@ -0,0 +1,98 @@
+#!/usr/bin/env bash
+
+############################  GLOBAL VARIABLES
+regex=' '
+branch="master"
+max_length=150
+
+REGEX_SUFFIX_GO=".+\.go$"
+
+############################  FUNCTIONS
+msg() {
+    printf '%b' "$1" >&2
+}
+
+die() {
+    msg "\33[31m[✘]\33[0m ${1}${2}"
+    exit 1
+}
+
+succ() {
+    msg "\33[34m[√]\33[0m ${1}${2}"
+}
+
+gostd() {
+    local branch=$1
+    local reg4exclude=$2
+    local max_length=$3
+
+    for file in `git diff $branch --name-only`
+    do
+        if ! [[ $file =~ $REGEX_SUFFIX_GO ]] || [[ $file =~ $reg4exclude ]]; then
+            continue
+        fi
+
+        error=`go fmt $file 2>&1`
+        if ! [ $? -eq 0 ]; then
+            die "go fmt $file:" "$error"
+        fi
+
+        succ "$file\n"
+
+        grep -n -E --color=always ".{$max_length}" $file | awk '{ printf ("%4s %s\n", "", $0) }'
+    done
+}
+
+get_options() {
+    while getopts "b:e:hl:" opts
+    do
+        case $opts in
+            b)
+                branch=$OPTARG
+                ;;
+            e)
+                regex=$OPTARG
+                ;;
+            h)
+                usage
+                exit 0
+                ;;
+            l)
+                max_length=$OPTARG
+                ;;
+            \?)
+                usage
+                exit 1
+                ;;
+        esac
+    done
+}
+
+usage () {
+    cat << _EOC_
+Usage:
+    gostd [options]
+
+Options:
+    -b <branch/commit>        Specify the git diff branch or commit.
+                              (default: master)
+    -e <regex>                Regex for excluding file or directory.
+    -h                        Print this usage.
+    -l <length>               Show files that exceed the limit line length.
+                              (default: 150)
+
+Examples:
+    gostd
+    gostd -b master -l 100
+    gostd -b 59d532a -e weed/pb -l 100
+_EOC_
+}
+
+main() {
+    get_options "$@"
+
+    gostd "$branch" "$regex" "$max_length"
+}
+
+############################  MAIN()
+main "$@"

weed/command/backup.go

@@ -3,6 +3,13 @@ package command
 import (
 	"fmt"
 
+	"github.com/spf13/viper"
+
+	"github.com/chrislusf/seaweedfs/weed/security"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle"
+	"github.com/chrislusf/seaweedfs/weed/storage/super_block"
+	"github.com/chrislusf/seaweedfs/weed/util"
+
 	"github.com/chrislusf/seaweedfs/weed/operation"
 	"github.com/chrislusf/seaweedfs/weed/storage"
 )
@@ -12,10 +19,12 @@ var (
 )
 
 type BackupOptions struct {
-	master     *string
-	collection *string
-	dir        *string
-	volumeId   *int
+	master      *string
+	collection  *string
+	dir         *string
+	volumeId    *int
+	ttl         *string
+	replication *string
 }
 
 func init() {
@@ -24,32 +33,45 @@ func init() {
 	s.collection = cmdBackup.Flag.String("collection", "", "collection name")
 	s.dir = cmdBackup.Flag.String("dir", ".", "directory to store volume data files")
 	s.volumeId = cmdBackup.Flag.Int("volumeId", -1, "a volume id. The volume .dat and .idx files should already exist in the dir.")
+	s.ttl = cmdBackup.Flag.String("ttl", "", `backup volume's time to live, format: 
+				3m: 3 minutes
+				4h: 4 hours
+				5d: 5 days
+				6w: 6 weeks
+				7M: 7 months
+				8y: 8 years
+				default is the same with origin`)
+	s.replication = cmdBackup.Flag.String("replication", "", "backup volume's replication, default is the same with origin")
 }
 
 var cmdBackup = &Command{
 	UsageLine: "backup -dir=. -volumeId=234 -server=localhost:9333",
 	Short:     "incrementally backup a volume to local folder",
 	Long: `Incrementally backup volume data.
-	
+
 	It is expected that you use this inside a script, to loop through
 	all possible volume ids that needs to be backup to local folder.
-	
+
 	The volume id does not need to exist locally or even remotely.
 	This will help to backup future new volumes.
-	
+
 	Usually backing up is just copying the .dat (and .idx) files.
-	But it's tricky to incremententally copy the differences.
-	
+	But it's tricky to incrementally copy the differences.
+
 	The complexity comes when there are multiple addition, deletion and compaction.
-	This tool will handle them correctly and efficiently, avoiding unnecessary data transporation.
+	This tool will handle them correctly and efficiently, avoiding unnecessary data transportation.
   `,
 }
 
 func runBackup(cmd *Command, args []string) bool {
+
+	util.LoadConfiguration("security", false)
+	grpcDialOption := security.LoadClientTLS(viper.Sub("grpc"), "client")
+
 	if *s.volumeId == -1 {
 		return false
 	}
-	vid := storage.VolumeId(*s.volumeId)
+	vid := needle.VolumeId(*s.volumeId)
 
 	// find volume location, replication, ttl info
 	lookup, err := operation.Lookup(*s.master, vid.String())
@@ -59,29 +81,73 @@ func runBackup(cmd *Command, args []string) bool {
 	}
 	volumeServer := lookup.Locations[0].Url
 
-	stats, err := operation.GetVolumeSyncStatus(volumeServer, uint32(vid))
+	stats, err := operation.GetVolumeSyncStatus(volumeServer, grpcDialOption, uint32(vid))
 	if err != nil {
 		fmt.Printf("Error get volume %d status: %v\n", vid, err)
 		return true
 	}
-	ttl, err := storage.ReadTTL(stats.Ttl)
-	if err != nil {
-		fmt.Printf("Error get volume %d ttl %s: %v\n", vid, stats.Ttl, err)
-		return true
+	var ttl *needle.TTL
+	if *s.ttl != "" {
+		ttl, err = needle.ReadTTL(*s.ttl)
+		if err != nil {
+			fmt.Printf("Error generate volume %d ttl %s: %v\n", vid, *s.ttl, err)
+			return true
+		}
+	} else {
+		ttl, err = needle.ReadTTL(stats.Ttl)
+		if err != nil {
+			fmt.Printf("Error get volume %d ttl %s: %v\n", vid, stats.Ttl, err)
+			return true
+		}
 	}
-	replication, err := storage.NewReplicaPlacementFromString(stats.Replication)
-	if err != nil {
-		fmt.Printf("Error get volume %d replication %s : %v\n", vid, stats.Replication, err)
-		return true
+	var replication *super_block.ReplicaPlacement
+	if *s.replication != "" {
+		replication, err = super_block.NewReplicaPlacementFromString(*s.replication)
+		if err != nil {
+			fmt.Printf("Error generate volume %d replication %s : %v\n", vid, *s.replication, err)
+			return true
+		}
+	} else {
+		replication, err = super_block.NewReplicaPlacementFromString(stats.Replication)
+		if err != nil {
+			fmt.Printf("Error get volume %d replication %s : %v\n", vid, stats.Replication, err)
+			return true
+		}
 	}
-
-	v, err := storage.NewVolume(*s.dir, *s.collection, vid, storage.NeedleMapInMemory, replication, ttl, 0)
+	v, err := storage.NewVolume(*s.dir, *s.collection, vid, storage.NeedleMapInMemory, replication, ttl, 0, 0)
 	if err != nil {
 		fmt.Printf("Error creating or reading from volume %d: %v\n", vid, err)
 		return true
 	}
 
-	if err := v.Synchronize(volumeServer); err != nil {
+	if v.SuperBlock.CompactionRevision < uint16(stats.CompactRevision) {
+		if err = v.Compact2(30 * 1024 * 1024 * 1024); err != nil {
+			fmt.Printf("Compact Volume before synchronizing %v\n", err)
+			return true
+		}
+		if err = v.CommitCompact(); err != nil {
+			fmt.Printf("Commit Compact before synchronizing %v\n", err)
+			return true
+		}
+		v.SuperBlock.CompactionRevision = uint16(stats.CompactRevision)
+		v.DataBackend.WriteAt(v.SuperBlock.Bytes(), 0)
+	}
+
+	datSize, _, _ := v.FileStat()
+
+	if datSize > stats.TailOffset {
+		// remove the old data
+		v.Destroy()
+		// recreate an empty volume
+		v, err = storage.NewVolume(*s.dir, *s.collection, vid, storage.NeedleMapInMemory, replication, ttl, 0, 0)
+		if err != nil {
+			fmt.Printf("Error creating or reading from volume %d: %v\n", vid, err)
+			return true
+		}
+	}
+	defer v.Close()
+
+	if err := v.IncrementalBackup(volumeServer, grpcDialOption); err != nil {
 		fmt.Printf("Error synchronizing volume %d: %v\n", vid, err)
 		return true
 	}

weed/command/benchmark.go

@@ -15,6 +15,9 @@ import (
 	"sync"
 	"time"
 
+	"github.com/spf13/viper"
+	"google.golang.org/grpc"
+
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/operation"
 	"github.com/chrislusf/seaweedfs/weed/security"
@@ -33,19 +36,21 @@ type BenchmarkOptions struct {
 	read             *bool
 	sequentialRead   *bool
 	collection       *string
+	replication      *string
 	cpuprofile       *string
 	maxCpu           *int
-	secretKey        *string
+	grpcDialOption   grpc.DialOption
+	masterClient     *wdclient.MasterClient
 }
 
 var (
-	b            BenchmarkOptions
-	sharedBytes  []byte
-	masterClient *wdclient.MasterClient
+	b           BenchmarkOptions
+	sharedBytes []byte
+	isSecure    bool
 )
 
 func init() {
-	cmdBenchmark.Run = runbenchmark // break init cycle
+	cmdBenchmark.Run = runBenchmark // break init cycle
 	cmdBenchmark.IsDebug = cmdBenchmark.Flag.Bool("debug", false, "verbose debug information")
 	b.masters = cmdBenchmark.Flag.String("master", "localhost:9333", "SeaweedFS master location")
 	b.concurrency = cmdBenchmark.Flag.Int("c", 16, "number of concurrent write or read processes")
@@ -57,14 +62,14 @@ func init() {
 	b.read = cmdBenchmark.Flag.Bool("read", true, "enable read")
 	b.sequentialRead = cmdBenchmark.Flag.Bool("readSequentially", false, "randomly read by ids from \"-list\" specified file")
 	b.collection = cmdBenchmark.Flag.String("collection", "benchmark", "write data to this collection")
+	b.replication = cmdBenchmark.Flag.String("replication", "000", "replication type")
 	b.cpuprofile = cmdBenchmark.Flag.String("cpuprofile", "", "cpu profile output file")
 	b.maxCpu = cmdBenchmark.Flag.Int("maxCpu", 0, "maximum number of CPUs. 0 means all available CPUs")
-	b.secretKey = cmdBenchmark.Flag.String("secure.secret", "", "secret to encrypt Json Web Token(JWT)")
 	sharedBytes = make([]byte, 1024)
 }
 
 var cmdBenchmark = &Command{
-	UsageLine: "benchmark -server=localhost:9333 -c=10 -n=100000",
+	UsageLine: "benchmark -master=localhost:9333 -c=10 -n=100000",
 	Short:     "benchmark on writing millions of files and read out",
 	Long: `benchmark on an empty SeaweedFS file system.
 
@@ -101,7 +106,11 @@ var (
 	readStats  *stats
 )
 
-func runbenchmark(cmd *Command, args []string) bool {
+func runBenchmark(cmd *Command, args []string) bool {
+
+	util.LoadConfiguration("security", false)
+	b.grpcDialOption = security.LoadClientTLS(viper.Sub("grpc"), "client")
+
 	fmt.Printf("This is SeaweedFS version %s %s %s\n", util.VERSION, runtime.GOOS, runtime.GOARCH)
 	if *b.maxCpu < 1 {
 		*b.maxCpu = runtime.NumCPU()
@@ -116,22 +125,22 @@ func runbenchmark(cmd *Command, args []string) bool {
 		defer pprof.StopCPUProfile()
 	}
 
-	masterClient = wdclient.NewMasterClient(context.Background(), "benchmark", strings.Split(*b.masters, ","))
-	go masterClient.KeepConnectedToMaster()
-	masterClient.WaitUntilConnected()
+	b.masterClient = wdclient.NewMasterClient(context.Background(), b.grpcDialOption, "client", strings.Split(*b.masters, ","))
+	go b.masterClient.KeepConnectedToMaster()
+	b.masterClient.WaitUntilConnected()
 
 	if *b.write {
-		bench_write()
+		benchWrite()
 	}
 
 	if *b.read {
-		bench_read()
+		benchRead()
 	}
 
 	return true
 }
 
-func bench_write() {
+func benchWrite() {
 	fileIdLineChan := make(chan string)
 	finishChan := make(chan bool)
 	writeStats = newStats(*b.concurrency)
@@ -158,7 +167,7 @@ func bench_write() {
 	writeStats.printStats()
 }
 
-func bench_read() {
+func benchRead() {
 	fileIdLineChan := make(chan string)
 	finishChan := make(chan bool)
 	readStats = newStats(*b.concurrency)
@@ -188,7 +197,6 @@ func writeFiles(idChan chan int, fileIdLineChan chan string, s *stat) {
 	defer wait.Done()
 	delayedDeleteChan := make(chan *delayedFile, 100)
 	var waitForDeletions sync.WaitGroup
-	secret := security.Secret(*b.secretKey)
 
 	for i := 0; i < 7; i++ {
 		waitForDeletions.Add(1)
@@ -198,8 +206,11 @@ func writeFiles(idChan chan int, fileIdLineChan chan string, s *stat) {
 				if df.enterTime.After(time.Now()) {
 					time.Sleep(df.enterTime.Sub(time.Now()))
 				}
-				if e := util.Delete("http://"+df.fp.Server+"/"+df.fp.Fid,
-					security.GenJwt(secret, df.fp.Fid)); e == nil {
+				var jwtAuthorization security.EncodedJwt
+				if isSecure {
+					jwtAuthorization = operation.LookupJwt(b.masterClient.GetMaster(), df.fp.Fid)
+				}
+				if e := util.Delete(fmt.Sprintf("http://%s/%s", df.fp.Server, df.fp.Fid), string(jwtAuthorization)); e == nil {
 					s.completed++
 				} else {
 					s.failed++
@@ -213,14 +224,22 @@ func writeFiles(idChan chan int, fileIdLineChan chan string, s *stat) {
 	for id := range idChan {
 		start := time.Now()
 		fileSize := int64(*b.fileSize + random.Intn(64))
-		fp := &operation.FilePart{Reader: &FakeReader{id: uint64(id), size: fileSize}, FileSize: fileSize}
-		ar := &operation.VolumeAssignRequest{
-			Count:      1,
-			Collection: *b.collection,
+		fp := &operation.FilePart{
+			Reader:   &FakeReader{id: uint64(id), size: fileSize},
+			FileSize: fileSize,
+			MimeType: "image/bench", // prevent gzip benchmark content
 		}
-		if assignResult, err := operation.Assign(masterClient.GetMaster(), ar); err == nil {
+		ar := &operation.VolumeAssignRequest{
+			Count:       1,
+			Collection:  *b.collection,
+			Replication: *b.replication,
+		}
+		if assignResult, err := operation.Assign(b.masterClient.GetMaster(), b.grpcDialOption, ar); err == nil {
 			fp.Server, fp.Fid, fp.Collection = assignResult.Url, assignResult.Fid, *b.collection
-			if _, err := fp.Upload(0, masterClient.GetMaster(), secret); err == nil {
+			if !isSecure && assignResult.Auth != "" {
+				isSecure = true
+			}
+			if _, err := fp.Upload(0, b.masterClient.GetMaster(), assignResult.Auth, b.grpcDialOption); err == nil {
 				if random.Intn(100) < *b.deletePercentage {
 					s.total++
 					delayedDeleteChan <- &delayedFile{time.Now().Add(time.Second), fp}
@@ -260,7 +279,7 @@ func readFiles(fileIdLineChan chan string, s *stat) {
 			fmt.Printf("reading file %s\n", fid)
 		}
 		start := time.Now()
-		url, err := masterClient.LookupFileId(fid)
+		url, err := b.masterClient.LookupFileId(fid)
 		if err != nil {
 			s.failed++
 			println("!!!! ", fid, " location not found!!!!!")

weed/command/command.go

@@ -13,7 +13,6 @@ var Commands = []*Command{
 	cmdCompact,
 	cmdCopy,
 	cmdFix,
-	cmdFilerExport,
 	cmdFilerReplicate,
 	cmdServer,
 	cmdMaster,
@@ -27,6 +26,7 @@ var Commands = []*Command{
 	cmdVolume,
 	cmdExport,
 	cmdMount,
+	cmdWebDav,
 }
 
 type Command struct {

weed/command/compact.go

@@ -3,6 +3,7 @@ package command
 import (
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/storage"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle"
 )
 
 func init() {
@@ -16,6 +17,9 @@ var cmdCompact = &Command{
   The compacted .dat file is stored as .cpd file.
   The compacted .idx file is stored as .cpx file.
 
+  For method=0, it compacts based on the .dat file, works if .idx file is corrupted.
+  For method=1, it compacts based on the .idx file, works if deletion happened but not written to .dat files.
+
   `,
 }
 
@@ -35,18 +39,18 @@ func runCompact(cmd *Command, args []string) bool {
 
 	preallocate := *compactVolumePreallocate * (1 << 20)
 
-	vid := storage.VolumeId(*compactVolumeId)
+	vid := needle.VolumeId(*compactVolumeId)
 	v, err := storage.NewVolume(*compactVolumePath, *compactVolumeCollection, vid,
-		storage.NeedleMapInMemory, nil, nil, preallocate)
+		storage.NeedleMapInMemory, nil, nil, preallocate, 0)
 	if err != nil {
 		glog.Fatalf("Load Volume [ERROR] %s\n", err)
 	}
 	if *compactMethod == 0 {
-		if err = v.Compact(preallocate); err != nil {
+		if err = v.Compact(preallocate, 0); err != nil {
 			glog.Fatalf("Compact Volume [ERROR] %s\n", err)
 		}
 	} else {
-		if err = v.Compact2(); err != nil {
+		if err = v.Compact2(preallocate); err != nil {
 			glog.Fatalf("Compact Volume [ERROR] %s\n", err)
 		}
 	}

weed/command/export.go

@@ -4,6 +4,7 @@ import (
 	"archive/tar"
 	"bytes"
 	"fmt"
+	"io"
 	"os"
 	"path"
 	"path/filepath"
@@ -14,8 +15,10 @@ import (
 
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/storage"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle_map"
+	"github.com/chrislusf/seaweedfs/weed/storage/super_block"
 	"github.com/chrislusf/seaweedfs/weed/storage/types"
-	"io"
 )
 
 const (
@@ -66,10 +69,10 @@ var (
 	localLocation, _             = time.LoadLocation("Local")
 )
 
-func printNeedle(vid storage.VolumeId, n *storage.Needle, version storage.Version, deleted bool) {
-	key := storage.NewFileIdFromNeedle(vid, n).String()
+func printNeedle(vid needle.VolumeId, n *needle.Needle, version needle.Version, deleted bool) {
+	key := needle.NewFileIdFromNeedle(vid, n).String()
 	size := n.DataSize
-	if version == storage.Version1 {
+	if version == needle.Version1 {
 		size = n.Size
 	}
 	fmt.Printf("%s\t%s\t%d\t%t\t%s\t%s\t%s\t%t\n",
@@ -84,6 +87,62 @@ func printNeedle(vid storage.VolumeId, n *storage.Needle, version storage.Versio
 	)
 }
 
+type VolumeFileScanner4Export struct {
+	version   needle.Version
+	counter   int
+	needleMap *needle_map.MemDb
+	vid       needle.VolumeId
+}
+
+func (scanner *VolumeFileScanner4Export) VisitSuperBlock(superBlock super_block.SuperBlock) error {
+	scanner.version = superBlock.Version
+	return nil
+
+}
+func (scanner *VolumeFileScanner4Export) ReadNeedleBody() bool {
+	return true
+}
+
+func (scanner *VolumeFileScanner4Export) VisitNeedle(n *needle.Needle, offset int64, needleHeader, needleBody []byte) error {
+	needleMap := scanner.needleMap
+	vid := scanner.vid
+
+	nv, ok := needleMap.Get(n.Id)
+	glog.V(3).Infof("key %d offset %d size %d disk_size %d gzip %v ok %v nv %+v",
+		n.Id, offset, n.Size, n.DiskSize(scanner.version), n.IsGzipped(), ok, nv)
+	if ok && nv.Size > 0 && nv.Size != types.TombstoneFileSize && nv.Offset.ToAcutalOffset() == offset {
+		if newerThanUnix >= 0 && n.HasLastModifiedDate() && n.LastModified < uint64(newerThanUnix) {
+			glog.V(3).Infof("Skipping this file, as it's old enough: LastModified %d vs %d",
+				n.LastModified, newerThanUnix)
+			return nil
+		}
+		scanner.counter++
+		if *limit > 0 && scanner.counter > *limit {
+			return io.EOF
+		}
+		if tarOutputFile != nil {
+			return writeFile(vid, n)
+		} else {
+			printNeedle(vid, n, scanner.version, false)
+			return nil
+		}
+	}
+	if !ok {
+		if *showDeleted && tarOutputFile == nil {
+			if n.DataSize > 0 {
+				printNeedle(vid, n, scanner.version, true)
+			} else {
+				n.Name = []byte("*tombstone")
+				printNeedle(vid, n, scanner.version, true)
+			}
+		}
+		glog.V(2).Infof("This seems deleted %d size %d", n.Id, n.Size)
+	} else {
+		glog.V(2).Infof("Skipping later-updated Id %d size %d", n.Id, n.Size)
+	}
+	return nil
+}
+
 func runExport(cmd *Command, args []string) bool {
 
 	var err error
@@ -133,67 +192,23 @@ func runExport(cmd *Command, args []string) bool {
 	if *export.collection != "" {
 		fileName = *export.collection + "_" + fileName
 	}
-	vid := storage.VolumeId(*export.volumeId)
-	indexFile, err := os.OpenFile(path.Join(*export.dir, fileName+".idx"), os.O_RDONLY, 0644)
-	if err != nil {
-		glog.Fatalf("Create Volume Index [ERROR] %s\n", err)
-	}
-	defer indexFile.Close()
+	vid := needle.VolumeId(*export.volumeId)
 
-	needleMap, err := storage.LoadBtreeNeedleMap(indexFile)
-	if err != nil {
-		glog.Fatalf("cannot load needle map from %s: %s", indexFile.Name(), err)
+	needleMap := needle_map.NewMemDb()
+	if err := needleMap.LoadFromIdx(path.Join(*export.dir, fileName+".idx")); err != nil {
+		glog.Fatalf("cannot load needle map from %s.idx: %s", fileName, err)
 	}
 
-	var version storage.Version
+	volumeFileScanner := &VolumeFileScanner4Export{
+		needleMap: needleMap,
+		vid:       vid,
+	}
 
 	if tarOutputFile == nil {
 		fmt.Printf("key\tname\tsize\tgzip\tmime\tmodified\tttl\tdeleted\n")
 	}
 
-	var counter = 0
-
-	err = storage.ScanVolumeFile(*export.dir, *export.collection, vid,
-		storage.NeedleMapInMemory,
-		func(superBlock storage.SuperBlock) error {
-			version = superBlock.Version()
-			return nil
-		}, true, func(n *storage.Needle, offset int64) error {
-			nv, ok := needleMap.Get(n.Id)
-			glog.V(3).Infof("key %d offset %d size %d disk_size %d gzip %v ok %v nv %+v",
-				n.Id, offset, n.Size, n.DiskSize(version), n.IsGzipped(), ok, nv)
-			if ok && nv.Size > 0 && int64(nv.Offset)*types.NeedlePaddingSize == offset {
-				if newerThanUnix >= 0 && n.HasLastModifiedDate() && n.LastModified < uint64(newerThanUnix) {
-					glog.V(3).Infof("Skipping this file, as it's old enough: LastModified %d vs %d",
-						n.LastModified, newerThanUnix)
-					return nil
-				}
-				counter++
-				if *limit > 0 && counter > *limit {
-					return io.EOF
-				}
-				if tarOutputFile != nil {
-					return writeFile(vid, n)
-				} else {
-					printNeedle(vid, n, version, false)
-					return nil
-				}
-			}
-			if !ok {
-				if *showDeleted && tarOutputFile == nil {
-					if n.DataSize > 0 {
-						printNeedle(vid, n, version, true)
-					} else {
-						n.Name = []byte("*tombstone")
-						printNeedle(vid, n, version, true)
-					}
-				}
-				glog.V(2).Infof("This seems deleted %d size %d", n.Id, n.Size)
-			} else {
-				glog.V(2).Infof("Skipping later-updated Id %d size %d", n.Id, n.Size)
-			}
-			return nil
-		})
+	err = storage.ScanVolumeFile(*export.dir, *export.collection, vid, storage.NeedleMapInMemory, volumeFileScanner)
 	if err != nil && err != io.EOF {
 		glog.Fatalf("Export Volume File [ERROR] %s\n", err)
 	}
@@ -208,8 +223,8 @@ type nameParams struct {
 	Ext  string
 }
 
-func writeFile(vid storage.VolumeId, n *storage.Needle) (err error) {
-	key := storage.NewFileIdFromNeedle(vid, n).String()
+func writeFile(vid needle.VolumeId, n *needle.Needle) (err error) {
+	key := needle.NewFileIdFromNeedle(vid, n).String()
 	fileNameTemplateBuffer.Reset()
 	if err = fileNameTemplate.Execute(fileNameTemplateBuffer,
 		nameParams{

weed/command/filer.go

@@ -6,6 +6,9 @@ import (
 	"strings"
 	"time"
 
+	"github.com/chrislusf/seaweedfs/weed/security"
+	"github.com/spf13/viper"
+
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
 	"github.com/chrislusf/seaweedfs/weed/server"
@@ -21,17 +24,19 @@ type FilerOptions struct {
 	masters                 *string
 	ip                      *string
 	port                    *int
-	grpcPort                *int
 	publicPort              *int
 	collection              *string
 	defaultReplicaPlacement *string
 	redirectOnRead          *bool
 	disableDirListing       *bool
 	maxMB                   *int
-	secretKey               *string
 	dirListingLimit         *int
 	dataCenter              *string
 	enableNotification      *bool
+	disableHttp             *bool
+
+	// default leveldb directory, used in "weed server" mode
+	defaultLevelDbDirectory *string
 }
 
 func init() {
@@ -40,15 +45,14 @@ func init() {
 	f.collection = cmdFiler.Flag.String("collection", "", "all data will be stored in this collection")
 	f.ip = cmdFiler.Flag.String("ip", "", "filer server http listen ip address")
 	f.port = cmdFiler.Flag.Int("port", 8888, "filer server http listen port")
-	f.grpcPort = cmdFiler.Flag.Int("port.grpc", 0, "filer grpc server listen port, default to http port + 10000")
-	f.publicPort = cmdFiler.Flag.Int("port.public", 0, "port opened to public")
+	f.publicPort = cmdFiler.Flag.Int("port.readonly", 0, "readonly port opened to public")
 	f.defaultReplicaPlacement = cmdFiler.Flag.String("defaultReplicaPlacement", "000", "default replication type if not specified")
 	f.redirectOnRead = cmdFiler.Flag.Bool("redirectOnRead", false, "whether proxy or redirect to volume server during file GET request")
 	f.disableDirListing = cmdFiler.Flag.Bool("disableDirListing", false, "turn off directory listing")
 	f.maxMB = cmdFiler.Flag.Int("maxMB", 32, "split files larger than the limit")
-	f.secretKey = cmdFiler.Flag.String("secure.secret", "", "secret to encrypt Json Web Token(JWT)")
-	f.dirListingLimit = cmdFiler.Flag.Int("dirListLimit", 1000, "limit sub dir listing size")
+	f.dirListingLimit = cmdFiler.Flag.Int("dirListLimit", 100000, "limit sub dir listing size")
 	f.dataCenter = cmdFiler.Flag.String("dataCenter", "", "prefer to write to volumes in this data center")
+	f.disableHttp = cmdFiler.Flag.Bool("disableHttp", false, "disable http request, only gRpc operations are allowed")
 }
 
 var cmdFiler = &Command{
@@ -67,13 +71,15 @@ var cmdFiler = &Command{
 
 	The configuration file "filer.toml" is read from ".", "$HOME/.seaweedfs/", or "/etc/seaweedfs/", in that order.
 
-	The example filer.toml configuration file can be generated by "weed scaffold filer"
+	The example filer.toml configuration file can be generated by "weed scaffold -config=filer"
 
 `,
 }
 
 func runFiler(cmd *Command, args []string) bool {
 
+	util.LoadConfiguration("security", false)
+
 	f.startFiler()
 
 	return true
@@ -88,16 +94,23 @@ func (fo *FilerOptions) startFiler() {
 		publicVolumeMux = http.NewServeMux()
 	}
 
+	defaultLevelDbDirectory := "./filerldb2"
+	if fo.defaultLevelDbDirectory != nil {
+		defaultLevelDbDirectory = *fo.defaultLevelDbDirectory + "/filerldb2"
+	}
+
 	fs, nfs_err := weed_server.NewFilerServer(defaultMux, publicVolumeMux, &weed_server.FilerOption{
-		Masters:            strings.Split(*f.masters, ","),
+		Masters:            strings.Split(*fo.masters, ","),
 		Collection:         *fo.collection,
 		DefaultReplication: *fo.defaultReplicaPlacement,
 		RedirectOnRead:     *fo.redirectOnRead,
 		DisableDirListing:  *fo.disableDirListing,
 		MaxMB:              *fo.maxMB,
-		SecretKey:          *fo.secretKey,
 		DirListingLimit:    *fo.dirListingLimit,
 		DataCenter:         *fo.dataCenter,
+		DefaultLevelDbDir:  defaultLevelDbDirectory,
+		DisableHttp:        *fo.disableHttp,
+		Port:               *fo.port,
 	})
 	if nfs_err != nil {
 		glog.Fatalf("Filer startup error: %v", nfs_err)
@@ -117,9 +130,9 @@ func (fo *FilerOptions) startFiler() {
 		}()
 	}
 
-	glog.V(0).Infoln("Start Seaweed Filer", util.VERSION, "at port", strconv.Itoa(*fo.port))
+	glog.V(0).Infof("Start Seaweed Filer %s at %s:%d", util.VERSION, *fo.ip, *fo.port)
 	filerListener, e := util.NewListener(
-		":"+strconv.Itoa(*fo.port),
+		*fo.ip+":"+strconv.Itoa(*fo.port),
 		time.Duration(10)*time.Second,
 	)
 	if e != nil {
@@ -127,15 +140,12 @@ func (fo *FilerOptions) startFiler() {
 	}
 
 	// starting grpc server
-	grpcPort := *fo.grpcPort
-	if grpcPort == 0 {
-		grpcPort = *fo.port + 10000
-	}
+	grpcPort := *fo.port + 10000
 	grpcL, err := util.NewListener(":"+strconv.Itoa(grpcPort), 0)
 	if err != nil {
 		glog.Fatalf("failed to listen on grpc port %d: %v", grpcPort, err)
 	}
-	grpcS := util.NewGrpcServer()
+	grpcS := util.NewGrpcServer(security.LoadServerTLS(viper.Sub("grpc"), "filer"))
 	filer_pb.RegisterSeaweedFilerServer(grpcS, fs)
 	reflection.Register(grpcS)
 	go grpcS.Serve(grpcL)

weed/command/filer_copy.go

@@ -1,53 +1,58 @@
 package command
 
 import (
+	"context"
 	"fmt"
+	"io"
 	"io/ioutil"
+	"net/http"
 	"net/url"
 	"os"
 	"path/filepath"
+	"strconv"
 	"strings"
+	"sync"
+	"time"
 
-	"context"
 	"github.com/chrislusf/seaweedfs/weed/operation"
 	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
 	"github.com/chrislusf/seaweedfs/weed/security"
 	"github.com/chrislusf/seaweedfs/weed/util"
-	"io"
-	"net/http"
-	"path"
-	"strconv"
-	"time"
+	"github.com/chrislusf/seaweedfs/weed/wdclient"
+	"github.com/spf13/viper"
+	"google.golang.org/grpc"
 )
 
 var (
-	copy CopyOptions
+	copy      CopyOptions
+	waitGroup sync.WaitGroup
 )
 
 type CopyOptions struct {
-	filerGrpcPort *int
-	master        *string
-	include       *string
-	replication   *string
-	collection    *string
-	ttl           *string
-	maxMB         *int
-	secretKey     *string
-
-	secret security.Secret
+	include           *string
+	replication       *string
+	collection        *string
+	ttl               *string
+	maxMB             *int
+	masterClient      *wdclient.MasterClient
+	concurrenctFiles  *int
+	concurrenctChunks *int
+	compressionLevel  *int
+	grpcDialOption    grpc.DialOption
+	masters           []string
 }
 
 func init() {
 	cmdCopy.Run = runCopy // break init cycle
 	cmdCopy.IsDebug = cmdCopy.Flag.Bool("debug", false, "verbose debug information")
-	copy.master = cmdCopy.Flag.String("master", "localhost:9333", "SeaweedFS master location")
 	copy.include = cmdCopy.Flag.String("include", "", "pattens of files to copy, e.g., *.pdf, *.html, ab?d.txt, works together with -dir")
 	copy.replication = cmdCopy.Flag.String("replication", "", "replication type")
 	copy.collection = cmdCopy.Flag.String("collection", "", "optional collection name")
 	copy.ttl = cmdCopy.Flag.String("ttl", "", "time to live, e.g.: 1m, 1h, 1d, 1M, 1y")
-	copy.maxMB = cmdCopy.Flag.Int("maxMB", 0, "split files larger than the limit")
-	copy.filerGrpcPort = cmdCopy.Flag.Int("filer.port.grpc", 0, "filer grpc server listen port, default to filer port + 10000")
-	copy.secretKey = cmdCopy.Flag.String("secure.secret", "", "secret to encrypt Json Web Token(JWT)")
+	copy.maxMB = cmdCopy.Flag.Int("maxMB", 32, "split files larger than the limit")
+	copy.concurrenctFiles = cmdCopy.Flag.Int("c", 8, "concurrent file copy goroutines")
+	copy.concurrenctChunks = cmdCopy.Flag.Int("concurrentChunks", 8, "concurrent chunk copy goroutines for each file")
+	copy.compressionLevel = cmdCopy.Flag.Int("compressionLevel", 9, "local file compression level 1 ~ 9")
 }
 
 var cmdCopy = &Command{
@@ -61,17 +66,15 @@ var cmdCopy = &Command{
   All files under the folder and subfolders will be copyed.
   Optional parameter "-include" allows you to specify the file name patterns.
 
-  If any file has a ".gz" extension, the content are considered gzipped already, and will be stored as is.
-  This can save volume server's gzipped processing and allow customizable gzip compression level.
-  The file name will strip out ".gz" and stored. For example, "jquery.js.gz" will be stored as "jquery.js".
-
   If "maxMB" is set to a positive number, files larger than it would be split into chunks.
 
   `,
 }
 
 func runCopy(cmd *Command, args []string) bool {
-	copy.secret = security.Secret(*copy.secretKey)
+
+	util.LoadConfiguration("security", false)
+
 	if len(args) <= 1 {
 		return false
 	}
@@ -85,7 +88,8 @@ func runCopy(cmd *Command, args []string) bool {
 	}
 	urlPath := filerUrl.Path
 	if !strings.HasSuffix(urlPath, "/") {
-		urlPath = urlPath + "/"
+		fmt.Printf("The last argument should be a folder and end with \"/\": %v\n", err)
+		return false
 	}
 
 	if filerUrl.Port() == "" {
@@ -100,98 +104,198 @@ func runCopy(cmd *Command, args []string) bool {
 	}
 
 	filerGrpcPort := filerPort + 10000
-	if *copy.filerGrpcPort != 0 {
-		filerGrpcPort = uint64(*copy.filerGrpcPort)
-	}
-
 	filerGrpcAddress := fmt.Sprintf("%s:%d", filerUrl.Hostname(), filerGrpcPort)
+	copy.grpcDialOption = security.LoadClientTLS(viper.Sub("grpc"), "client")
 
-	for _, fileOrDir := range fileOrDirs {
-		if !doEachCopy(fileOrDir, filerUrl.Host, filerGrpcAddress, urlPath) {
-			return false
-		}
+	ctx := context.Background()
+
+	masters, collection, replication, maxMB, err := readFilerConfiguration(ctx, copy.grpcDialOption, filerGrpcAddress)
+	if err != nil {
+		fmt.Printf("read from filer %s: %v\n", filerGrpcAddress, err)
+		return false
 	}
+	if *copy.collection == "" {
+		*copy.collection = collection
+	}
+	if *copy.replication == "" {
+		*copy.replication = replication
+	}
+	if *copy.maxMB == 0 {
+		*copy.maxMB = int(maxMB)
+	}
+	copy.masters = masters
+
+	copy.masterClient = wdclient.NewMasterClient(ctx, copy.grpcDialOption, "client", copy.masters)
+	go copy.masterClient.KeepConnectedToMaster()
+	copy.masterClient.WaitUntilConnected()
+
+	if *cmdCopy.IsDebug {
+		util.SetupProfiling("filer.copy.cpu.pprof", "filer.copy.mem.pprof")
+	}
+
+	fileCopyTaskChan := make(chan FileCopyTask, *copy.concurrenctFiles)
+
+	go func() {
+		defer close(fileCopyTaskChan)
+		for _, fileOrDir := range fileOrDirs {
+			if err := genFileCopyTask(fileOrDir, urlPath, fileCopyTaskChan); err != nil {
+				fmt.Fprintf(os.Stderr, "gen file list error: %v\n", err)
+				break
+			}
+		}
+	}()
+	for i := 0; i < *copy.concurrenctFiles; i++ {
+		waitGroup.Add(1)
+		go func() {
+			defer waitGroup.Done()
+			worker := FileCopyWorker{
+				options:          &copy,
+				filerHost:        filerUrl.Host,
+				filerGrpcAddress: filerGrpcAddress,
+			}
+			if err := worker.copyFiles(ctx, fileCopyTaskChan); err != nil {
+				fmt.Fprintf(os.Stderr, "copy file error: %v\n", err)
+				return
+			}
+		}()
+	}
+	waitGroup.Wait()
+
 	return true
 }
 
-func doEachCopy(fileOrDir string, filerAddress, filerGrpcAddress string, path string) bool {
-	f, err := os.Open(fileOrDir)
-	if err != nil {
-		fmt.Printf("Failed to open file %s: %v\n", fileOrDir, err)
-		return false
-	}
-	defer f.Close()
+func readFilerConfiguration(ctx context.Context, grpcDialOption grpc.DialOption, filerGrpcAddress string) (masters []string, collection, replication string, maxMB uint32, err error) {
+	err = withFilerClient(ctx, filerGrpcAddress, grpcDialOption, func(client filer_pb.SeaweedFilerClient) error {
+		resp, err := client.GetFilerConfiguration(ctx, &filer_pb.GetFilerConfigurationRequest{})
+		if err != nil {
+			return fmt.Errorf("get filer %s configuration: %v", filerGrpcAddress, err)
+		}
+		masters, collection, replication, maxMB = resp.Masters, resp.Collection, resp.Replication, resp.MaxMb
+		return nil
+	})
+	return
+}
 
-	fi, err := f.Stat()
+func genFileCopyTask(fileOrDir string, destPath string, fileCopyTaskChan chan FileCopyTask) error {
+
+	fi, err := os.Stat(fileOrDir)
 	if err != nil {
-		fmt.Printf("Failed to get stat for file %s: %v\n", fileOrDir, err)
-		return false
+		fmt.Fprintf(os.Stderr, "Failed to get stat for file %s: %v\n", fileOrDir, err)
+		return nil
 	}
 
 	mode := fi.Mode()
 	if mode.IsDir() {
 		files, _ := ioutil.ReadDir(fileOrDir)
 		for _, subFileOrDir := range files {
-			if !doEachCopy(fileOrDir+"/"+subFileOrDir.Name(), filerAddress, filerGrpcAddress, path+fi.Name()+"/") {
-				return false
+			if err = genFileCopyTask(fileOrDir+"/"+subFileOrDir.Name(), destPath+fi.Name()+"/", fileCopyTaskChan); err != nil {
+				return err
 			}
 		}
-		return true
+		return nil
 	}
 
+	uid, gid := util.GetFileUidGid(fi)
+
+	fileCopyTaskChan <- FileCopyTask{
+		sourceLocation:     fileOrDir,
+		destinationUrlPath: destPath,
+		fileSize:           fi.Size(),
+		fileMode:           fi.Mode(),
+		uid:                uid,
+		gid:                gid,
+	}
+
+	return nil
+}
+
+type FileCopyWorker struct {
+	options          *CopyOptions
+	filerHost        string
+	filerGrpcAddress string
+}
+
+func (worker *FileCopyWorker) copyFiles(ctx context.Context, fileCopyTaskChan chan FileCopyTask) error {
+	for task := range fileCopyTaskChan {
+		if err := worker.doEachCopy(ctx, task); err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+type FileCopyTask struct {
+	sourceLocation     string
+	destinationUrlPath string
+	fileSize           int64
+	fileMode           os.FileMode
+	uid                uint32
+	gid                uint32
+}
+
+func (worker *FileCopyWorker) doEachCopy(ctx context.Context, task FileCopyTask) error {
+
+	f, err := os.Open(task.sourceLocation)
+	if err != nil {
+		fmt.Printf("Failed to open file %s: %v\n", task.sourceLocation, err)
+		if _, ok := err.(*os.PathError); ok {
+			fmt.Printf("skipping %s\n", task.sourceLocation)
+			return nil
+		}
+		return err
+	}
+	defer f.Close()
+
 	// this is a regular file
-	if *copy.include != "" {
-		if ok, _ := filepath.Match(*copy.include, filepath.Base(fileOrDir)); !ok {
-			return true
+	if *worker.options.include != "" {
+		if ok, _ := filepath.Match(*worker.options.include, filepath.Base(task.sourceLocation)); !ok {
+			return nil
 		}
 	}
 
 	// find the chunk count
-	chunkSize := int64(*copy.maxMB * 1024 * 1024)
+	chunkSize := int64(*worker.options.maxMB * 1024 * 1024)
 	chunkCount := 1
-	if chunkSize > 0 && fi.Size() > chunkSize {
-		chunkCount = int(fi.Size()/chunkSize) + 1
+	if chunkSize > 0 && task.fileSize > chunkSize {
+		chunkCount = int(task.fileSize/chunkSize) + 1
 	}
 
 	if chunkCount == 1 {
-		return uploadFileAsOne(filerAddress, filerGrpcAddress, path, f, fi)
+		return worker.uploadFileAsOne(ctx, task, f)
 	}
 
-	return uploadFileInChunks(filerAddress, filerGrpcAddress, path, f, fi, chunkCount, chunkSize)
+	return worker.uploadFileInChunks(ctx, task, f, chunkCount, chunkSize)
 }
 
-func uploadFileAsOne(filerAddress, filerGrpcAddress string, urlFolder string, f *os.File, fi os.FileInfo) bool {
+func (worker *FileCopyWorker) uploadFileAsOne(ctx context.Context, task FileCopyTask, f *os.File) error {
 
 	// upload the file content
 	fileName := filepath.Base(f.Name())
 	mimeType := detectMimeType(f)
-	isGzipped := isGzipped(fileName)
 
 	var chunks []*filer_pb.FileChunk
 
-	if fi.Size() > 0 {
+	if task.fileSize > 0 {
 
 		// assign a volume
-		assignResult, err := operation.Assign(*copy.master, &operation.VolumeAssignRequest{
+		assignResult, err := operation.Assign(worker.options.masterClient.GetMaster(), worker.options.grpcDialOption, &operation.VolumeAssignRequest{
 			Count:       1,
-			Replication: *copy.replication,
-			Collection:  *copy.collection,
-			Ttl:         *copy.ttl,
+			Replication: *worker.options.replication,
+			Collection:  *worker.options.collection,
+			Ttl:         *worker.options.ttl,
 		})
 		if err != nil {
-			fmt.Printf("Failed to assign from %s: %v\n", *copy.master, err)
+			fmt.Printf("Failed to assign from %v: %v\n", worker.options.masters, err)
 		}
 
 		targetUrl := "http://" + assignResult.Url + "/" + assignResult.Fid
 
-		uploadResult, err := operation.Upload(targetUrl, fileName, f, isGzipped, mimeType, nil, "")
+		uploadResult, err := operation.UploadWithLocalCompressionLevel(targetUrl, fileName, f, false, mimeType, nil, assignResult.Auth, *worker.options.compressionLevel)
 		if err != nil {
-			fmt.Printf("upload data %v to %s: %v\n", fileName, targetUrl, err)
-			return false
+			return fmt.Errorf("upload data %v to %s: %v\n", fileName, targetUrl, err)
 		}
 		if uploadResult.Error != "" {
-			fmt.Printf("upload %v to %s result: %v\n", fileName, targetUrl, uploadResult.Error)
-			return false
+			return fmt.Errorf("upload %v to %s result: %v\n", fileName, targetUrl, uploadResult.Error)
 		}
 		fmt.Printf("uploaded %s to %s\n", fileName, targetUrl)
 
@@ -203,128 +307,150 @@ func uploadFileAsOne(filerAddress, filerGrpcAddress string, urlFolder string, f
 			ETag:   uploadResult.ETag,
 		})
 
-		fmt.Printf("copied %s => http://%s%s%s\n", fileName, filerAddress, urlFolder, fileName)
+		fmt.Printf("copied %s => http://%s%s%s\n", fileName, worker.filerHost, task.destinationUrlPath, fileName)
 	}
 
-	if err := withFilerClient(filerGrpcAddress, func(client filer_pb.SeaweedFilerClient) error {
+	if err := withFilerClient(ctx, worker.filerGrpcAddress, worker.options.grpcDialOption, func(client filer_pb.SeaweedFilerClient) error {
 		request := &filer_pb.CreateEntryRequest{
-			Directory: urlFolder,
+			Directory: task.destinationUrlPath,
 			Entry: &filer_pb.Entry{
 				Name: fileName,
 				Attributes: &filer_pb.FuseAttributes{
 					Crtime:      time.Now().Unix(),
 					Mtime:       time.Now().Unix(),
-					Gid:         uint32(os.Getgid()),
-					Uid:         uint32(os.Getuid()),
-					FileSize:    uint64(fi.Size()),
-					FileMode:    uint32(fi.Mode()),
+					Gid:         task.gid,
+					Uid:         task.uid,
+					FileSize:    uint64(task.fileSize),
+					FileMode:    uint32(task.fileMode),
 					Mime:        mimeType,
-					Replication: *copy.replication,
-					Collection:  *copy.collection,
-					TtlSec:      int32(util.ParseInt(*copy.ttl, 0)),
+					Replication: *worker.options.replication,
+					Collection:  *worker.options.collection,
+					TtlSec:      int32(util.ParseInt(*worker.options.ttl, 0)),
 				},
 				Chunks: chunks,
 			},
 		}
 
-		if _, err := client.CreateEntry(context.Background(), request); err != nil {
+		if _, err := client.CreateEntry(ctx, request); err != nil {
 			return fmt.Errorf("update fh: %v", err)
 		}
 		return nil
 	}); err != nil {
-		fmt.Printf("upload data %v to http://%s%s%s: %v\n", fileName, filerAddress, urlFolder, fileName, err)
-		return false
+		return fmt.Errorf("upload data %v to http://%s%s%s: %v\n", fileName, worker.filerHost, task.destinationUrlPath, fileName, err)
 	}
 
-	return true
+	return nil
 }
 
-func uploadFileInChunks(filerAddress, filerGrpcAddress string, urlFolder string, f *os.File, fi os.FileInfo, chunkCount int, chunkSize int64) bool {
+func (worker *FileCopyWorker) uploadFileInChunks(ctx context.Context, task FileCopyTask, f *os.File, chunkCount int, chunkSize int64) error {
 
 	fileName := filepath.Base(f.Name())
 	mimeType := detectMimeType(f)
 
+	chunksChan := make(chan *filer_pb.FileChunk, chunkCount)
+
+	concurrentChunks := make(chan struct{}, *worker.options.concurrenctChunks)
+	var wg sync.WaitGroup
+	var uploadError error
+
+	fmt.Printf("uploading %s in %d chunks ...\n", fileName, chunkCount)
+	for i := int64(0); i < int64(chunkCount) && uploadError == nil; i++ {
+		wg.Add(1)
+		concurrentChunks <- struct{}{}
+		go func(i int64) {
+			defer func() {
+				wg.Done()
+				<-concurrentChunks
+			}()
+			// assign a volume
+			assignResult, err := operation.Assign(worker.options.masterClient.GetMaster(), worker.options.grpcDialOption, &operation.VolumeAssignRequest{
+				Count:       1,
+				Replication: *worker.options.replication,
+				Collection:  *worker.options.collection,
+				Ttl:         *worker.options.ttl,
+			})
+			if err != nil {
+				fmt.Printf("Failed to assign from %v: %v\n", worker.options.masters, err)
+			}
+
+			targetUrl := "http://" + assignResult.Url + "/" + assignResult.Fid
+
+			uploadResult, err := operation.Upload(targetUrl,
+				fileName+"-"+strconv.FormatInt(i+1, 10),
+				io.NewSectionReader(f, i*chunkSize, chunkSize),
+				false, "", nil, assignResult.Auth)
+			if err != nil {
+				uploadError = fmt.Errorf("upload data %v to %s: %v\n", fileName, targetUrl, err)
+				return
+			}
+			if uploadResult.Error != "" {
+				uploadError = fmt.Errorf("upload %v to %s result: %v\n", fileName, targetUrl, uploadResult.Error)
+				return
+			}
+			chunksChan <- &filer_pb.FileChunk{
+				FileId: assignResult.Fid,
+				Offset: i * chunkSize,
+				Size:   uint64(uploadResult.Size),
+				Mtime:  time.Now().UnixNano(),
+				ETag:   uploadResult.ETag,
+			}
+			fmt.Printf("uploaded %s-%d to %s [%d,%d)\n", fileName, i+1, targetUrl, i*chunkSize, i*chunkSize+int64(uploadResult.Size))
+		}(i)
+	}
+	wg.Wait()
+	close(chunksChan)
+
 	var chunks []*filer_pb.FileChunk
-
-	for i := int64(0); i < int64(chunkCount); i++ {
-
-		// assign a volume
-		assignResult, err := operation.Assign(*copy.master, &operation.VolumeAssignRequest{
-			Count:       1,
-			Replication: *copy.replication,
-			Collection:  *copy.collection,
-			Ttl:         *copy.ttl,
-		})
-		if err != nil {
-			fmt.Printf("Failed to assign from %s: %v\n", *copy.master, err)
-		}
-
-		targetUrl := "http://" + assignResult.Url + "/" + assignResult.Fid
-
-		uploadResult, err := operation.Upload(targetUrl,
-			fileName+"-"+strconv.FormatInt(i+1, 10),
-			io.LimitReader(f, chunkSize),
-			false, "application/octet-stream", nil, "")
-		if err != nil {
-			fmt.Printf("upload data %v to %s: %v\n", fileName, targetUrl, err)
-			return false
-		}
-		if uploadResult.Error != "" {
-			fmt.Printf("upload %v to %s result: %v\n", fileName, targetUrl, uploadResult.Error)
-			return false
-		}
-		chunks = append(chunks, &filer_pb.FileChunk{
-			FileId: assignResult.Fid,
-			Offset: i * chunkSize,
-			Size:   uint64(uploadResult.Size),
-			Mtime:  time.Now().UnixNano(),
-			ETag:   uploadResult.ETag,
-		})
-		fmt.Printf("uploaded %s-%d to %s [%d,%d)\n", fileName, i+1, targetUrl, i*chunkSize, i*chunkSize+int64(uploadResult.Size))
+	for chunk := range chunksChan {
+		chunks = append(chunks, chunk)
 	}
 
-	if err := withFilerClient(filerGrpcAddress, func(client filer_pb.SeaweedFilerClient) error {
+	if uploadError != nil {
+		var fileIds []string
+		for _, chunk := range chunks {
+			fileIds = append(fileIds, chunk.FileId)
+		}
+		operation.DeleteFiles(worker.options.masterClient.GetMaster(), worker.options.grpcDialOption, fileIds)
+		return uploadError
+	}
+
+	if err := withFilerClient(ctx, worker.filerGrpcAddress, worker.options.grpcDialOption, func(client filer_pb.SeaweedFilerClient) error {
 		request := &filer_pb.CreateEntryRequest{
-			Directory: urlFolder,
+			Directory: task.destinationUrlPath,
 			Entry: &filer_pb.Entry{
 				Name: fileName,
 				Attributes: &filer_pb.FuseAttributes{
 					Crtime:      time.Now().Unix(),
 					Mtime:       time.Now().Unix(),
-					Gid:         uint32(os.Getgid()),
-					Uid:         uint32(os.Getuid()),
-					FileSize:    uint64(fi.Size()),
-					FileMode:    uint32(fi.Mode()),
+					Gid:         task.gid,
+					Uid:         task.uid,
+					FileSize:    uint64(task.fileSize),
+					FileMode:    uint32(task.fileMode),
 					Mime:        mimeType,
-					Replication: *copy.replication,
-					Collection:  *copy.collection,
-					TtlSec:      int32(util.ParseInt(*copy.ttl, 0)),
+					Replication: *worker.options.replication,
+					Collection:  *worker.options.collection,
+					TtlSec:      int32(util.ParseInt(*worker.options.ttl, 0)),
 				},
 				Chunks: chunks,
 			},
 		}
 
-		if _, err := client.CreateEntry(context.Background(), request); err != nil {
+		if _, err := client.CreateEntry(ctx, request); err != nil {
 			return fmt.Errorf("update fh: %v", err)
 		}
 		return nil
 	}); err != nil {
-		fmt.Printf("upload data %v to http://%s%s%s: %v\n", fileName, filerAddress, urlFolder, fileName, err)
-		return false
+		return fmt.Errorf("upload data %v to http://%s%s%s: %v\n", fileName, worker.filerHost, task.destinationUrlPath, fileName, err)
 	}
 
-	fmt.Printf("copied %s => http://%s%s%s\n", fileName, filerAddress, urlFolder, fileName)
+	fmt.Printf("copied %s => http://%s%s%s\n", fileName, worker.filerHost, task.destinationUrlPath, fileName)
 
-	return true
-}
-
-func isGzipped(filename string) bool {
-	return strings.ToLower(path.Ext(filename)) == ".gz"
+	return nil
 }
 
 func detectMimeType(f *os.File) string {
 	head := make([]byte, 512)
-	f.Seek(0, 0)
+	f.Seek(0, io.SeekStart)
 	n, err := f.Read(head)
 	if err == io.EOF {
 		return ""
@@ -333,20 +459,16 @@ func detectMimeType(f *os.File) string {
 		fmt.Printf("read head of %v: %v\n", f.Name(), err)
 		return "application/octet-stream"
 	}
-	f.Seek(0, 0)
+	f.Seek(0, io.SeekStart)
 	mimeType := http.DetectContentType(head[:n])
 	return mimeType
 }
 
-func withFilerClient(filerAddress string, fn func(filer_pb.SeaweedFilerClient) error) error {
+func withFilerClient(ctx context.Context, filerAddress string, grpcDialOption grpc.DialOption, fn func(filer_pb.SeaweedFilerClient) error) error {
 
-	grpcConnection, err := util.GrpcDial(filerAddress)
-	if err != nil {
-		return fmt.Errorf("fail to dial %s: %v", filerAddress, err)
-	}
-	defer grpcConnection.Close()
+	return util.WithCachedGrpcClient(ctx, func(clientConn *grpc.ClientConn) error {
+		client := filer_pb.NewSeaweedFilerClient(clientConn)
+		return fn(client)
+	}, filerAddress, grpcDialOption)
 
-	client := filer_pb.NewSeaweedFilerClient(grpcConnection)
-
-	return fn(client)
 }

weed/command/filer_export.go

@@ -1,174 +0,0 @@
-package command
-
-import (
-	"github.com/chrislusf/seaweedfs/weed/filer2"
-	"github.com/chrislusf/seaweedfs/weed/glog"
-	"github.com/chrislusf/seaweedfs/weed/notification"
-	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
-	"github.com/chrislusf/seaweedfs/weed/server"
-	"github.com/spf13/viper"
-)
-
-func init() {
-	cmdFilerExport.Run = runFilerExport // break init cycle
-}
-
-var cmdFilerExport = &Command{
-	UsageLine: "filer.export -sourceStore=mysql -targetStroe=cassandra",
-	Short:     "export meta data in filer store",
-	Long: `Iterate the file tree and export all metadata out
-
-	Both source and target store:
-        * should be a store name already specified in filer.toml
-        * do not need to be enabled state
-
-	If target store is empty, only the directory tree will be listed.
-
-	If target store is "notification", the list of entries will be sent to notification.
-	This is usually used to bootstrap filer replication to a remote system.
-
-  `,
-}
-
-var (
-	// filerExportOutputFile  = cmdFilerExport.Flag.String("output", "", "the output file. If empty, only list out the directory tree")
-	filerExportSourceStore = cmdFilerExport.Flag.String("sourceStore", "", "the source store name in filer.toml, default to currently enabled store")
-	filerExportTargetStore = cmdFilerExport.Flag.String("targetStore", "", "the target store name in filer.toml, or \"notification\" to export all files to message queue")
-	dir                    = cmdFilerExport.Flag.String("dir", "/", "only process files under this directory")
-	dirListLimit           = cmdFilerExport.Flag.Int("dirListLimit", 100000, "limit directory list size")
-	dryRun                 = cmdFilerExport.Flag.Bool("dryRun", false, "not actually moving data")
-)
-
-type statistics struct {
-	directoryCount int
-	fileCount      int
-}
-
-func runFilerExport(cmd *Command, args []string) bool {
-
-	weed_server.LoadConfiguration("filer", true)
-	config := viper.GetViper()
-
-	var sourceStore, targetStore filer2.FilerStore
-
-	for _, store := range filer2.Stores {
-		if store.GetName() == *filerExportSourceStore || *filerExportSourceStore == "" && config.GetBool(store.GetName()+".enabled") {
-			viperSub := config.Sub(store.GetName())
-			if err := store.Initialize(viperSub); err != nil {
-				glog.Fatalf("Failed to initialize source store for %s: %+v",
-					store.GetName(), err)
-			} else {
-				sourceStore = store
-			}
-			break
-		}
-	}
-
-	for _, store := range filer2.Stores {
-		if store.GetName() == *filerExportTargetStore {
-			viperSub := config.Sub(store.GetName())
-			if err := store.Initialize(viperSub); err != nil {
-				glog.Fatalf("Failed to initialize target store for %s: %+v",
-					store.GetName(), err)
-			} else {
-				targetStore = store
-			}
-			break
-		}
-	}
-
-	if sourceStore == nil {
-		glog.Errorf("Failed to find source store %s", *filerExportSourceStore)
-		println("existing data sources are:")
-		for _, store := range filer2.Stores {
-			println("    " + store.GetName())
-		}
-		return false
-	}
-
-	if targetStore == nil && *filerExportTargetStore != "" && *filerExportTargetStore != "notification" {
-		glog.Errorf("Failed to find target store %s", *filerExportTargetStore)
-		println("existing data sources are:")
-		for _, store := range filer2.Stores {
-			println("    " + store.GetName())
-		}
-		return false
-	}
-
-	stat := statistics{}
-
-	var fn func(level int, entry *filer2.Entry) error
-
-	if *filerExportTargetStore == "notification" {
-		weed_server.LoadConfiguration("notification", false)
-		v := viper.GetViper()
-		notification.LoadConfiguration(v.Sub("notification"))
-
-		fn = func(level int, entry *filer2.Entry) error {
-			printout(level, entry)
-			if *dryRun {
-				return nil
-			}
-			return notification.Queue.SendMessage(
-				string(entry.FullPath),
-				&filer_pb.EventNotification{
-					NewEntry: entry.ToProtoEntry(),
-				},
-			)
-		}
-	} else if targetStore == nil {
-		fn = printout
-	} else {
-		fn = func(level int, entry *filer2.Entry) error {
-			printout(level, entry)
-			if *dryRun {
-				return nil
-			}
-			return targetStore.InsertEntry(entry)
-		}
-	}
-
-	doTraverse(&stat, sourceStore, filer2.FullPath(*dir), 0, fn)
-
-	glog.Infof("processed %d directories, %d files", stat.directoryCount, stat.fileCount)
-
-	return true
-}
-
-func doTraverse(stat *statistics, filerStore filer2.FilerStore, parentPath filer2.FullPath, level int, fn func(level int, entry *filer2.Entry) error) {
-
-	limit := *dirListLimit
-	lastEntryName := ""
-	for {
-		entries, err := filerStore.ListDirectoryEntries(parentPath, lastEntryName, false, limit)
-		if err != nil {
-			break
-		}
-		for _, entry := range entries {
-			if fnErr := fn(level, entry); fnErr != nil {
-				glog.Errorf("failed to process entry: %s", entry.FullPath)
-			}
-			if entry.IsDirectory() {
-				stat.directoryCount++
-				doTraverse(stat, filerStore, entry.FullPath, level+1, fn)
-			} else {
-				stat.fileCount++
-			}
-		}
-		if len(entries) < limit {
-			break
-		}
-	}
-}
-
-func printout(level int, entry *filer2.Entry) error {
-	for i := 0; i < level; i++ {
-		if i == level-1 {
-			print("+-")
-		} else {
-			print("| ")
-		}
-	}
-	println(entry.FullPath.Name())
-	return nil
-}

weed/command/filer_replication.go

@@ -1,6 +1,7 @@
 package command
 
 import (
+	"context"
 	"strings"
 
 	"github.com/chrislusf/seaweedfs/weed/glog"
@@ -12,7 +13,7 @@ import (
 	_ "github.com/chrislusf/seaweedfs/weed/replication/sink/gcssink"
 	_ "github.com/chrislusf/seaweedfs/weed/replication/sink/s3sink"
 	"github.com/chrislusf/seaweedfs/weed/replication/sub"
-	"github.com/chrislusf/seaweedfs/weed/server"
+	"github.com/chrislusf/seaweedfs/weed/util"
 	"github.com/spf13/viper"
 )
 
@@ -28,19 +29,22 @@ var cmdFilerReplicate = &Command{
 	filer.replicate listens on filer notifications. If any file is updated, it will fetch the updated content,
 	and write to the other destination.
 
-	Run "weed scaffold -config replication" to generate a replication.toml file and customize the parameters.
+	Run "weed scaffold -config=replication" to generate a replication.toml file and customize the parameters.
 
   `,
 }
 
 func runFilerReplicate(cmd *Command, args []string) bool {
 
-	weed_server.LoadConfiguration("replication", true)
-	weed_server.LoadConfiguration("notification", true)
+	util.LoadConfiguration("security", false)
+	util.LoadConfiguration("replication", true)
+	util.LoadConfiguration("notification", true)
 	config := viper.GetViper()
 
 	var notificationInput sub.NotificationInput
 
+	validateOneEnabledInput(config)
+
 	for _, input := range sub.NotificationInputs {
 		if config.GetBool("notification." + input.GetName() + ".enabled") {
 			viperSub := config.Sub("notification." + input.GetName())
@@ -113,7 +117,7 @@ func runFilerReplicate(cmd *Command, args []string) bool {
 		} else {
 			glog.V(1).Infof("modify: %s", key)
 		}
-		if err = replicator.Replicate(key, m); err != nil {
+		if err = replicator.Replicate(context.Background(), key, m); err != nil {
 			glog.Errorf("replicate %s: %+v", key, err)
 		} else {
 			glog.V(1).Infof("replicated %s", key)
@@ -122,3 +126,16 @@ func runFilerReplicate(cmd *Command, args []string) bool {
 
 	return true
 }
+
+func validateOneEnabledInput(config *viper.Viper) {
+	enabledInput := ""
+	for _, input := range sub.NotificationInputs {
+		if config.GetBool("notification." + input.GetName() + ".enabled") {
+			if enabledInput == "" {
+				enabledInput = input.GetName()
+			} else {
+				glog.Fatalf("Notification input is enabled for both %s and %s", enabledInput, input.GetName())
+			}
+		}
+	}
+}

weed/command/fix.go

@@ -7,6 +7,9 @@ import (
 
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/storage"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle"
+	"github.com/chrislusf/seaweedfs/weed/storage/needle_map"
+	"github.com/chrislusf/seaweedfs/weed/storage/super_block"
 	"github.com/chrislusf/seaweedfs/weed/storage/types"
 )
 
@@ -28,6 +31,32 @@ var (
 	fixVolumeId         = cmdFix.Flag.Int("volumeId", -1, "a volume id. The volume should already exist in the dir. The volume index file should not exist.")
 )
 
+type VolumeFileScanner4Fix struct {
+	version needle.Version
+	nm      *needle_map.MemDb
+}
+
+func (scanner *VolumeFileScanner4Fix) VisitSuperBlock(superBlock super_block.SuperBlock) error {
+	scanner.version = superBlock.Version
+	return nil
+
+}
+func (scanner *VolumeFileScanner4Fix) ReadNeedleBody() bool {
+	return false
+}
+
+func (scanner *VolumeFileScanner4Fix) VisitNeedle(n *needle.Needle, offset int64, needleHeader, needleBody []byte) error {
+	glog.V(2).Infof("key %d offset %d size %d disk_size %d gzip %v", n.Id, offset, n.Size, n.DiskSize(scanner.version), n.IsGzipped())
+	if n.Size > 0 && n.Size != types.TombstoneFileSize {
+		pe := scanner.nm.Set(n.Id, types.ToOffset(offset), n.Size)
+		glog.V(2).Infof("saved %d with error %v", n.Size, pe)
+	} else {
+		glog.V(2).Infof("skipping deleted file ...")
+		return scanner.nm.Delete(n.Id)
+	}
+	return nil
+}
+
 func runFix(cmd *Command, args []string) bool {
 
 	if *fixVolumeId == -1 {
@@ -45,27 +74,14 @@ func runFix(cmd *Command, args []string) bool {
 	}
 	defer indexFile.Close()
 
-	nm := storage.NewBtreeNeedleMap(indexFile)
-	defer nm.Close()
+	nm := needle_map.NewMemDb()
 
-	var version storage.Version
-	vid := storage.VolumeId(*fixVolumeId)
-	err = storage.ScanVolumeFile(*fixVolumePath, *fixVolumeCollection, vid,
-		storage.NeedleMapInMemory,
-		func(superBlock storage.SuperBlock) error {
-			version = superBlock.Version()
-			return nil
-		}, false, func(n *storage.Needle, offset int64) error {
-			glog.V(2).Infof("key %d offset %d size %d disk_size %d gzip %v", n.Id, offset, n.Size, n.DiskSize(version), n.IsGzipped())
-			if n.Size > 0 {
-				pe := nm.Put(n.Id, types.Offset(offset/types.NeedlePaddingSize), n.Size)
-				glog.V(2).Infof("saved %d with error %v", n.Size, pe)
-			} else {
-				glog.V(2).Infof("skipping deleted file ...")
-				return nm.Delete(n.Id, types.Offset(offset/types.NeedlePaddingSize))
-			}
-			return nil
-		})
+	vid := needle.VolumeId(*fixVolumeId)
+	scanner := &VolumeFileScanner4Fix{
+		nm: nm,
+	}
+
+	err = storage.ScanVolumeFile(*fixVolumePath, *fixVolumeCollection, vid, storage.NeedleMapInMemory, scanner)
 	if err != nil {
 		glog.Fatalf("Export Volume File [ERROR] %s\n", err)
 		os.Remove(indexFileName)

weed/command/master.go

@@ -6,112 +6,143 @@ import (
 	"runtime"
 	"strconv"
 	"strings"
-	"time"
 
+	"github.com/chrislusf/raft/protobuf"
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/pb/master_pb"
+	"github.com/chrislusf/seaweedfs/weed/security"
 	"github.com/chrislusf/seaweedfs/weed/server"
+	"github.com/chrislusf/seaweedfs/weed/storage/backend"
 	"github.com/chrislusf/seaweedfs/weed/util"
 	"github.com/gorilla/mux"
-	"github.com/soheilhy/cmux"
+	"github.com/spf13/viper"
 	"google.golang.org/grpc/reflection"
 )
 
+var (
+	m MasterOptions
+)
+
+type MasterOptions struct {
+	port               *int
+	ip                 *string
+	ipBind             *string
+	metaFolder         *string
+	peers              *string
+	volumeSizeLimitMB  *uint
+	volumePreallocate  *bool
+	pulseSeconds       *int
+	defaultReplication *string
+	garbageThreshold   *float64
+	whiteList          *string
+	disableHttp        *bool
+	metricsAddress     *string
+	metricsIntervalSec *int
+}
+
 func init() {
 	cmdMaster.Run = runMaster // break init cycle
+	m.port = cmdMaster.Flag.Int("port", 9333, "http listen port")
+	m.ip = cmdMaster.Flag.String("ip", "localhost", "master <ip>|<server> address")
+	m.ipBind = cmdMaster.Flag.String("ip.bind", "0.0.0.0", "ip address to bind to")
+	m.metaFolder = cmdMaster.Flag.String("mdir", os.TempDir(), "data directory to store meta data")
+	m.peers = cmdMaster.Flag.String("peers", "", "all master nodes in comma separated ip:port list, example: 127.0.0.1:9093,127.0.0.1:9094")
+	m.volumeSizeLimitMB = cmdMaster.Flag.Uint("volumeSizeLimitMB", 30*1000, "Master stops directing writes to oversized volumes.")
+	m.volumePreallocate = cmdMaster.Flag.Bool("volumePreallocate", false, "Preallocate disk space for volumes.")
+	m.pulseSeconds = cmdMaster.Flag.Int("pulseSeconds", 5, "number of seconds between heartbeats")
+	m.defaultReplication = cmdMaster.Flag.String("defaultReplication", "000", "Default replication type if not specified.")
+	m.garbageThreshold = cmdMaster.Flag.Float64("garbageThreshold", 0.3, "threshold to vacuum and reclaim spaces")
+	m.whiteList = cmdMaster.Flag.String("whiteList", "", "comma separated Ip addresses having write permission. No limit if empty.")
+	m.disableHttp = cmdMaster.Flag.Bool("disableHttp", false, "disable http requests, only gRPC operations are allowed.")
+	m.metricsAddress = cmdMaster.Flag.String("metrics.address", "", "Prometheus gateway address")
+	m.metricsIntervalSec = cmdMaster.Flag.Int("metrics.intervalSeconds", 15, "Prometheus push interval in seconds")
 }
 
 var cmdMaster = &Command{
 	UsageLine: "master -port=9333",
 	Short:     "start a master server",
-	Long: `start a master server to provide volume=>location mapping service
-  and sequence number of file ids
+	Long: `start a master server to provide volume=>location mapping service and sequence number of file ids
+
+	The configuration file "security.toml" is read from ".", "$HOME/.seaweedfs/", or "/etc/seaweedfs/", in that order.
+
+	The example security.toml configuration file can be generated by "weed scaffold -config=security"
 
   `,
 }
 
 var (
-	mport                   = cmdMaster.Flag.Int("port", 9333, "http listen port")
-	masterIp                = cmdMaster.Flag.String("ip", "localhost", "master <ip>|<server> address")
-	masterBindIp            = cmdMaster.Flag.String("ip.bind", "0.0.0.0", "ip address to bind to")
-	metaFolder              = cmdMaster.Flag.String("mdir", os.TempDir(), "data directory to store meta data")
-	masterPeers             = cmdMaster.Flag.String("peers", "", "all master nodes in comma separated ip:port list, example: 127.0.0.1:9093,127.0.0.1:9094")
-	volumeSizeLimitMB       = cmdMaster.Flag.Uint("volumeSizeLimitMB", 30*1000, "Master stops directing writes to oversized volumes.")
-	volumePreallocate       = cmdMaster.Flag.Bool("volumePreallocate", false, "Preallocate disk space for volumes.")
-	mpulse                  = cmdMaster.Flag.Int("pulseSeconds", 5, "number of seconds between heartbeats")
-	defaultReplicaPlacement = cmdMaster.Flag.String("defaultReplication", "000", "Default replication type if not specified.")
-	// mTimeout                = cmdMaster.Flag.Int("idleTimeout", 30, "connection idle seconds")
-	mMaxCpu               = cmdMaster.Flag.Int("maxCpu", 0, "maximum number of CPUs. 0 means all available CPUs")
-	garbageThreshold      = cmdMaster.Flag.Float64("garbageThreshold", 0.3, "threshold to vacuum and reclaim spaces")
-	masterWhiteListOption = cmdMaster.Flag.String("whiteList", "", "comma separated Ip addresses having write permission. No limit if empty.")
-	masterSecureKey       = cmdMaster.Flag.String("secure.secret", "", "secret to encrypt Json Web Token(JWT)")
-	masterCpuProfile      = cmdMaster.Flag.String("cpuprofile", "", "cpu profile output file")
-	masterMemProfile      = cmdMaster.Flag.String("memprofile", "", "memory profile output file")
-
-	masterWhiteList []string
+	masterCpuProfile = cmdMaster.Flag.String("cpuprofile", "", "cpu profile output file")
+	masterMemProfile = cmdMaster.Flag.String("memprofile", "", "memory profile output file")
 )
 
 func runMaster(cmd *Command, args []string) bool {
-	if *mMaxCpu < 1 {
-		*mMaxCpu = runtime.NumCPU()
-	}
-	runtime.GOMAXPROCS(*mMaxCpu)
+
+	util.LoadConfiguration("security", false)
+	util.LoadConfiguration("master", false)
+
+	runtime.GOMAXPROCS(runtime.NumCPU())
 	util.SetupProfiling(*masterCpuProfile, *masterMemProfile)
 
-	if err := util.TestFolderWritable(*metaFolder); err != nil {
-		glog.Fatalf("Check Meta Folder (-mdir) Writable %s : %s", *metaFolder, err)
+	if err := util.TestFolderWritable(*m.metaFolder); err != nil {
+		glog.Fatalf("Check Meta Folder (-mdir) Writable %s : %s", *m.metaFolder, err)
 	}
-	if *masterWhiteListOption != "" {
-		masterWhiteList = strings.Split(*masterWhiteListOption, ",")
+
+	var masterWhiteList []string
+	if *m.whiteList != "" {
+		masterWhiteList = strings.Split(*m.whiteList, ",")
 	}
-	if *volumeSizeLimitMB > 30*1000 {
+	if *m.volumeSizeLimitMB > util.VolumeSizeLimitGB*1000 {
 		glog.Fatalf("volumeSizeLimitMB should be smaller than 30000")
 	}
 
+	startMaster(m, masterWhiteList)
+
+	return true
+}
+
+func startMaster(masterOption MasterOptions, masterWhiteList []string) {
+
+	backend.LoadConfiguration(viper.GetViper())
+
+	myMasterAddress, peers := checkPeers(*masterOption.ip, *masterOption.port, *masterOption.peers)
+
 	r := mux.NewRouter()
-	ms := weed_server.NewMasterServer(r, *mport, *metaFolder,
-		*volumeSizeLimitMB, *volumePreallocate,
-		*mpulse, *defaultReplicaPlacement, *garbageThreshold,
-		masterWhiteList, *masterSecureKey,
-	)
-
-	listeningAddress := *masterBindIp + ":" + strconv.Itoa(*mport)
-
-	glog.V(0).Infoln("Start Seaweed Master", util.VERSION, "at", listeningAddress)
-
-	listener, e := util.NewListener(listeningAddress, 0)
+	ms := weed_server.NewMasterServer(r, masterOption.toMasterOption(masterWhiteList), peers)
+	listeningAddress := *masterOption.ipBind + ":" + strconv.Itoa(*masterOption.port)
+	glog.V(0).Infof("Start Seaweed Master %s at %s", util.VERSION, listeningAddress)
+	masterListener, e := util.NewListener(listeningAddress, 0)
 	if e != nil {
 		glog.Fatalf("Master startup error: %v", e)
 	}
-
-	go func() {
-		time.Sleep(100 * time.Millisecond)
-		myMasterAddress, peers := checkPeers(*masterIp, *mport, *masterPeers)
-		raftServer := weed_server.NewRaftServer(r, peers, myMasterAddress, *metaFolder, ms.Topo, *mpulse)
-		ms.SetRaftServer(raftServer)
-	}()
-
-	// start grpc and http server
-	m := cmux.New(listener)
-
-	grpcL := m.Match(cmux.HTTP2HeaderField("content-type", "application/grpc"))
-	httpL := m.Match(cmux.Any())
-
-	// Create your protocol servers.
-	grpcS := util.NewGrpcServer()
-	master_pb.RegisterSeaweedServer(grpcS, ms)
-	reflection.Register(grpcS)
-
-	httpS := &http.Server{Handler: r}
-
-	go grpcS.Serve(grpcL)
-	go httpS.Serve(httpL)
-
-	if err := m.Serve(); err != nil {
-		glog.Fatalf("master server failed to serve: %v", err)
+	// start raftServer
+	raftServer := weed_server.NewRaftServer(security.LoadClientTLS(viper.Sub("grpc"), "master"),
+		peers, myMasterAddress, *masterOption.metaFolder, ms.Topo, *masterOption.pulseSeconds)
+	if raftServer == nil {
+		glog.Fatalf("please verify %s is writable, see https://github.com/chrislusf/seaweedfs/issues/717", *masterOption.metaFolder)
 	}
+	ms.SetRaftServer(raftServer)
+	r.HandleFunc("/cluster/status", raftServer.StatusHandler).Methods("GET")
+	// starting grpc server
+	grpcPort := *masterOption.port + 10000
+	grpcL, err := util.NewListener(*masterOption.ipBind+":"+strconv.Itoa(grpcPort), 0)
+	if err != nil {
+		glog.Fatalf("master failed to listen on grpc port %d: %v", grpcPort, err)
+	}
+	// Create your protocol servers.
+	grpcS := util.NewGrpcServer(security.LoadServerTLS(viper.Sub("grpc"), "master"))
+	master_pb.RegisterSeaweedServer(grpcS, ms)
+	protobuf.RegisterRaftServer(grpcS, raftServer)
+	reflection.Register(grpcS)
+	glog.V(0).Infof("Start Seaweed Master %s grpc server at %s:%d", util.VERSION, *masterOption.ipBind, grpcPort)
+	go grpcS.Serve(grpcL)
 
-	return true
+	go ms.MasterClient.KeepConnectedToMaster()
+
+	// start http server
+	httpS := &http.Server{Handler: r}
+	go httpS.Serve(masterListener)
+
+	select {}
 }
 
 func checkPeers(masterIp string, masterPort int, peers string) (masterAddress string, cleanedPeers []string) {
@@ -128,12 +159,27 @@ func checkPeers(masterIp string, masterPort int, peers string) (masterAddress st
 		}
 	}
 
-	peerCount := len(cleanedPeers)
 	if !hasSelf {
-		peerCount += 1
+		cleanedPeers = append(cleanedPeers, masterAddress)
 	}
-	if peerCount%2 == 0 {
+	if len(cleanedPeers)%2 == 0 {
 		glog.Fatalf("Only odd number of masters are supported!")
 	}
 	return
 }
+
+func (m *MasterOptions) toMasterOption(whiteList []string) *weed_server.MasterOption {
+	return &weed_server.MasterOption{
+		Port:                    *m.port,
+		MetaFolder:              *m.metaFolder,
+		VolumeSizeLimitMB:       *m.volumeSizeLimitMB,
+		VolumePreallocate:       *m.volumePreallocate,
+		PulseSeconds:            *m.pulseSeconds,
+		DefaultReplicaPlacement: *m.defaultReplication,
+		GarbageThreshold:        *m.garbageThreshold,
+		WhiteList:               whiteList,
+		DisableHttp:             *m.disableHttp,
+		MetricsAddress:          *m.metricsAddress,
+		MetricsIntervalSec:      *m.metricsIntervalSec,
+	}
+}

weed/command/mount.go

@@ -8,33 +8,39 @@ import (
 
 type MountOptions struct {
 	filer              *string
-	filerGrpcPort      *int
 	filerMountRootPath *string
 	dir                *string
-	dirListingLimit    *int
+	dirListCacheLimit  *int64
 	collection         *string
 	replication        *string
 	ttlSec             *int
 	chunkSizeLimitMB   *int
 	dataCenter         *string
+	allowOthers        *bool
+	umaskString        *string
 }
 
 var (
-	mountOptions MountOptions
+	mountOptions    MountOptions
+	mountCpuProfile *string
+	mountMemProfile *string
 )
 
 func init() {
 	cmdMount.Run = runMount // break init cycle
 	mountOptions.filer = cmdMount.Flag.String("filer", "localhost:8888", "weed filer location")
-	mountOptions.filerGrpcPort = cmdMount.Flag.Int("filer.grpc.port", 0, "filer grpc server listen port, default to http port + 10000")
 	mountOptions.filerMountRootPath = cmdMount.Flag.String("filer.path", "/", "mount this remote path from filer server")
 	mountOptions.dir = cmdMount.Flag.String("dir", ".", "mount weed filer to this directory")
-	mountOptions.dirListingLimit = cmdMount.Flag.Int("dirListLimit", 100000, "limit directory listing size")
+	mountOptions.dirListCacheLimit = cmdMount.Flag.Int64("dirListCacheLimit", 1000000, "limit cache size to speed up directory long format listing")
 	mountOptions.collection = cmdMount.Flag.String("collection", "", "collection to create the files")
 	mountOptions.replication = cmdMount.Flag.String("replication", "", "replication(e.g. 000, 001) to create to files. If empty, let filer decide.")
 	mountOptions.ttlSec = cmdMount.Flag.Int("ttl", 0, "file ttl in seconds")
-	mountOptions.chunkSizeLimitMB = cmdMount.Flag.Int("chunkSizeLimitMB", 16, "local write buffer size, also chunk large files")
+	mountOptions.chunkSizeLimitMB = cmdMount.Flag.Int("chunkSizeLimitMB", 4, "local write buffer size, also chunk large files")
 	mountOptions.dataCenter = cmdMount.Flag.String("dataCenter", "", "prefer to write to the data center")
+	mountOptions.allowOthers = cmdMount.Flag.Bool("allowOthers", true, "allows other users to access the file system")
+	mountOptions.umaskString = cmdMount.Flag.String("umask", "022", "octal umask, e.g., 022, 0111")
+	mountCpuProfile = cmdMount.Flag.String("cpuprofile", "", "cpu profile output file")
+	mountMemProfile = cmdMount.Flag.String("memprofile", "", "memory profile output file")
 }
 
 var cmdMount = &Command{
@@ -47,7 +53,7 @@ var cmdMount = &Command{
   2) have a "weed filer" running
   These 2 requirements can be achieved with one command "weed server -filer=true"
 
-  This uses bazil.org/fuse, which enables writing FUSE file systems on
+  This uses github.com/seaweedfs/fuse, which enables writing FUSE file systems on
   Linux, and OS X.
 
   On OS X, it requires OSXFUSE (http://osxfuse.github.com/).
@@ -55,21 +61,18 @@ var cmdMount = &Command{
   `,
 }
 
-func parseFilerGrpcAddress(filer string, optionalGrpcPort int) (filerGrpcAddress string, err error) {
+func parseFilerGrpcAddress(filer string) (filerGrpcAddress string, err error) {
 	hostnameAndPort := strings.Split(filer, ":")
 	if len(hostnameAndPort) != 2 {
-		return "", fmt.Errorf("The filer should have hostname:port format: %v", hostnameAndPort)
+		return "", fmt.Errorf("filer should have hostname:port format: %v", hostnameAndPort)
 	}
 
 	filerPort, parseErr := strconv.ParseUint(hostnameAndPort[1], 10, 64)
 	if parseErr != nil {
-		return "", fmt.Errorf("The filer filer port parse error: %v", parseErr)
+		return "", fmt.Errorf("filer port parse error: %v", parseErr)
 	}
 
 	filerGrpcPort := int(filerPort) + 10000
-	if optionalGrpcPort != 0 {
-		filerGrpcPort = optionalGrpcPort
-	}
 
 	return fmt.Sprintf("%s:%d", hostnameAndPort[0], filerGrpcPort), nil
 }

weed/command/mount_darwin.go

@@ -0,0 +1,13 @@
+package command
+
+import (
+	"github.com/seaweedfs/fuse"
+)
+
+func osSpecificMountOptions() []fuse.MountOption {
+	return []fuse.MountOption{}
+}
+
+func checkMountPointAvailable(dir string) bool {
+	return true
+}

weed/command/mount_freebsd.go

@@ -0,0 +1,13 @@
+package command
+
+import (
+	"github.com/seaweedfs/fuse"
+)
+
+func osSpecificMountOptions() []fuse.MountOption {
+	return []fuse.MountOption{}
+}
+
+func checkMountPointAvailable(dir string) bool {
+	return true
+}

weed/command/mount_linux.go

@@ -0,0 +1,157 @@
+package command
+
+import (
+	"bufio"
+	"fmt"
+	"io"
+	"os"
+	"strings"
+
+	"github.com/seaweedfs/fuse"
+)
+
+const (
+	/* 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
+	   (1)(2)(3)   (4)   (5)      (6)      (7)   (8) (9)   (10)         (11)
+
+	   (1) mount ID:  unique identifier of the mount (may be reused after umount)
+	   (2) parent ID:  ID of parent (or of self for the top of the mount tree)
+	   (3) major:minor:  value of st_dev for files on filesystem
+	   (4) root:  root of the mount within the filesystem
+	   (5) mount point:  mount point relative to the process's root
+	   (6) mount options:  per mount options
+	   (7) optional fields:  zero or more fields of the form "tag[:value]"
+	   (8) separator:  marks the end of the optional fields
+	   (9) filesystem type:  name of filesystem of the form "type[.subtype]"
+	   (10) mount source:  filesystem specific information or "none"
+	   (11) super options:  per super block options*/
+	mountinfoFormat = "%d %d %d:%d %s %s %s %s"
+)
+
+// Info reveals information about a particular mounted filesystem. This
+// struct is populated from the content in the /proc/<pid>/mountinfo file.
+type Info struct {
+	// ID is a unique identifier of the mount (may be reused after umount).
+	ID int
+
+	// Parent indicates the ID of the mount parent (or of self for the top of the
+	// mount tree).
+	Parent int
+
+	// Major indicates one half of the device ID which identifies the device class.
+	Major int
+
+	// Minor indicates one half of the device ID which identifies a specific
+	// instance of device.
+	Minor int
+
+	// Root of the mount within the filesystem.
+	Root string
+
+	// Mountpoint indicates the mount point relative to the process's root.
+	Mountpoint string
+
+	// Opts represents mount-specific options.
+	Opts string
+
+	// Optional represents optional fields.
+	Optional string
+
+	// Fstype indicates the type of filesystem, such as EXT3.
+	Fstype string
+
+	// Source indicates filesystem specific information or "none".
+	Source string
+
+	// VfsOpts represents per super block options.
+	VfsOpts string
+}
+
+// Mounted determines if a specified mountpoint has been mounted.
+// On Linux it looks at /proc/self/mountinfo and on Solaris at mnttab.
+func mounted(mountPoint string) (bool, error) {
+	entries, err := parseMountTable()
+	if err != nil {
+		return false, err
+	}
+
+	// Search the table for the mountPoint
+	for _, e := range entries {
+		if e.Mountpoint == mountPoint {
+			return true, nil
+		}
+	}
+	return false, nil
+}
+
+// Parse /proc/self/mountinfo because comparing Dev and ino does not work from
+// bind mounts
+func parseMountTable() ([]*Info, error) {
+	f, err := os.Open("/proc/self/mountinfo")
+	if err != nil {
+		return nil, err
+	}
+	defer f.Close()
+
+	return parseInfoFile(f)
+}
+
+func parseInfoFile(r io.Reader) ([]*Info, error) {
+	var (
+		s   = bufio.NewScanner(r)
+		out []*Info
+	)
+
+	for s.Scan() {
+		if err := s.Err(); err != nil {
+			return nil, err
+		}
+
+		var (
+			p              = &Info{}
+			text           = s.Text()
+			optionalFields string
+		)
+
+		if _, err := fmt.Sscanf(text, mountinfoFormat,
+			&p.ID, &p.Parent, &p.Major, &p.Minor,
+			&p.Root, &p.Mountpoint, &p.Opts, &optionalFields); err != nil {
+			return nil, fmt.Errorf("Scanning '%s' failed: %s", text, err)
+		}
+		// Safe as mountinfo encodes mountpoints with spaces as \040.
+		index := strings.Index(text, " - ")
+		postSeparatorFields := strings.Fields(text[index+3:])
+		if len(postSeparatorFields) < 3 {
+			return nil, fmt.Errorf("Error found less than 3 fields post '-' in %q", text)
+		}
+
+		if optionalFields != "-" {
+			p.Optional = optionalFields
+		}
+
+		p.Fstype = postSeparatorFields[0]
+		p.Source = postSeparatorFields[1]
+		p.VfsOpts = strings.Join(postSeparatorFields[2:], " ")
+		out = append(out, p)
+	}
+	return out, nil
+}
+
+func osSpecificMountOptions() []fuse.MountOption {
+	return []fuse.MountOption{
+		fuse.AllowNonEmptyMount(),
+	}
+}
+
+func checkMountPointAvailable(dir string) bool {
+	mountPoint := dir
+	if mountPoint != "/" && strings.HasSuffix(mountPoint, "/") {
+		mountPoint = mountPoint[0 : len(mountPoint)-1]
+	}
+
+	if mounted, err := mounted(mountPoint); err != nil || mounted {
+		return false
+	}
+
+	return true
+}

weed/command/mount_notsupported.go

@@ -1,5 +1,6 @@
 // +build !linux
 // +build !darwin
+// +build !freebsd
 
 package command
 

weed/command/mount_std.go

@@ -1,88 +1,179 @@
-// +build linux darwin
+// +build linux darwin freebsd
 
 package command
 
 import (
 	"fmt"
+	"os"
+	"os/user"
+	"path"
 	"runtime"
+	"strconv"
 	"strings"
 	"time"
 
-	"bazil.org/fuse"
-	"bazil.org/fuse/fs"
+	"github.com/jacobsa/daemonize"
+	"github.com/spf13/viper"
+
 	"github.com/chrislusf/seaweedfs/weed/filesys"
 	"github.com/chrislusf/seaweedfs/weed/glog"
+	"github.com/chrislusf/seaweedfs/weed/security"
 	"github.com/chrislusf/seaweedfs/weed/util"
+	"github.com/seaweedfs/fuse"
+	"github.com/seaweedfs/fuse/fs"
 )
 
 func runMount(cmd *Command, args []string) bool {
+
+	util.SetupProfiling(*mountCpuProfile, *mountMemProfile)
+
+	umask, umaskErr := strconv.ParseUint(*mountOptions.umaskString, 8, 64)
+	if umaskErr != nil {
+		fmt.Printf("can not parse umask %s", *mountOptions.umaskString)
+		return false
+	}
+
+	return RunMount(
+		*mountOptions.filer,
+		*mountOptions.filerMountRootPath,
+		*mountOptions.dir,
+		*mountOptions.collection,
+		*mountOptions.replication,
+		*mountOptions.dataCenter,
+		*mountOptions.chunkSizeLimitMB,
+		*mountOptions.allowOthers,
+		*mountOptions.ttlSec,
+		*mountOptions.dirListCacheLimit,
+		os.FileMode(umask),
+	)
+}
+
+func RunMount(filer, filerMountRootPath, dir, collection, replication, dataCenter string, chunkSizeLimitMB int,
+	allowOthers bool, ttlSec int, dirListCacheLimit int64, umask os.FileMode) bool {
+
+	util.LoadConfiguration("security", false)
+
 	fmt.Printf("This is SeaweedFS version %s %s %s\n", util.VERSION, runtime.GOOS, runtime.GOARCH)
-	if *mountOptions.dir == "" {
+	if dir == "" {
 		fmt.Printf("Please specify the mount directory via \"-dir\"")
 		return false
 	}
-	if *mountOptions.chunkSizeLimitMB <= 0 {
+	if chunkSizeLimitMB <= 0 {
 		fmt.Printf("Please specify a reasonable buffer size.")
 		return false
 	}
 
-	fuse.Unmount(*mountOptions.dir)
+	fuse.Unmount(dir)
 
-	c, err := fuse.Mount(
-		*mountOptions.dir,
-		fuse.VolumeName("SeaweedFS"),
-		fuse.FSName("SeaweedFS"),
+	uid, gid := uint32(0), uint32(0)
+
+	// detect mount folder mode
+	mountMode := os.ModeDir | 0755
+	fileInfo, err := os.Stat(dir)
+	if err == nil {
+		mountMode = os.ModeDir | fileInfo.Mode()
+		uid, gid = util.GetFileUidGid(fileInfo)
+		fmt.Printf("mount point owner uid=%d gid=%d mode=%s\n", uid, gid, fileInfo.Mode())
+	}
+
+	if uid == 0 {
+		if u, err := user.Current(); err == nil {
+			if parsedId, pe := strconv.ParseUint(u.Uid, 10, 32); pe == nil {
+				uid = uint32(parsedId)
+			}
+			if parsedId, pe := strconv.ParseUint(u.Gid, 10, 32); pe == nil {
+				gid = uint32(parsedId)
+			}
+			fmt.Printf("current uid=%d gid=%d\n", uid, gid)
+		}
+	}
+
+	// Ensure target mount point availability
+	if isValid := checkMountPointAvailable(dir); !isValid {
+		glog.Fatalf("Expected mount to still be active, target mount point: %s, please check!", dir)
+		return false
+	}
+
+	mountName := path.Base(dir)
+
+	options := []fuse.MountOption{
+		fuse.VolumeName(mountName),
+		fuse.FSName(filer + ":" + filerMountRootPath),
+		fuse.Subtype("seaweedfs"),
 		fuse.NoAppleDouble(),
 		fuse.NoAppleXattr(),
+		fuse.NoBrowse(),
+		fuse.AutoXattr(),
 		fuse.ExclCreate(),
 		fuse.DaemonTimeout("3600"),
-		fuse.AllowOther(),
 		fuse.AllowSUID(),
 		fuse.DefaultPermissions(),
-		// fuse.MaxReadahead(1024*128), // TODO: not tested yet, possibly improving read performance
+		fuse.MaxReadahead(1024 * 128),
 		fuse.AsyncRead(),
 		fuse.WritebackCache(),
-	)
+		fuse.AllowNonEmptyMount(),
+	}
+
+	options = append(options, osSpecificMountOptions()...)
+
+	if allowOthers {
+		options = append(options, fuse.AllowOther())
+	}
+
+	c, err := fuse.Mount(dir, options...)
 	if err != nil {
-		glog.Fatal(err)
-		return false
+		glog.V(0).Infof("mount: %v", err)
+		daemonize.SignalOutcome(err)
+		return true
 	}
 
 	util.OnInterrupt(func() {
-		fuse.Unmount(*mountOptions.dir)
+		fuse.Unmount(dir)
 		c.Close()
 	})
 
-	filerGrpcAddress, err := parseFilerGrpcAddress(*mountOptions.filer, *mountOptions.filerGrpcPort)
+	filerGrpcAddress, err := parseFilerGrpcAddress(filer)
 	if err != nil {
-		glog.Fatal(err)
-		return false
+		glog.V(0).Infof("parseFilerGrpcAddress: %v", err)
+		daemonize.SignalOutcome(err)
+		return true
 	}
 
-	mountRoot := *mountOptions.filerMountRootPath
+	mountRoot := filerMountRootPath
 	if mountRoot != "/" && strings.HasSuffix(mountRoot, "/") {
 		mountRoot = mountRoot[0 : len(mountRoot)-1]
 	}
 
+	daemonize.SignalOutcome(nil)
+
 	err = fs.Serve(c, filesys.NewSeaweedFileSystem(&filesys.Option{
 		FilerGrpcAddress:   filerGrpcAddress,
+		GrpcDialOption:     security.LoadClientTLS(viper.Sub("grpc"), "client"),
 		FilerMountRootPath: mountRoot,
-		Collection:         *mountOptions.collection,
-		Replication:        *mountOptions.replication,
-		TtlSec:             int32(*mountOptions.ttlSec),
-		ChunkSizeLimit:     int64(*mountOptions.chunkSizeLimitMB) * 1024 * 1024,
-		DataCenter:         *mountOptions.dataCenter,
-		DirListingLimit:    *mountOptions.dirListingLimit,
+		Collection:         collection,
+		Replication:        replication,
+		TtlSec:             int32(ttlSec),
+		ChunkSizeLimit:     int64(chunkSizeLimitMB) * 1024 * 1024,
+		DataCenter:         dataCenter,
+		DirListCacheLimit:  dirListCacheLimit,
 		EntryCacheTtl:      3 * time.Second,
+		MountUid:           uid,
+		MountGid:           gid,
+		MountMode:          mountMode,
+		MountCtime:         fileInfo.ModTime(),
+		MountMtime:         time.Now(),
+		Umask:              umask,
 	}))
 	if err != nil {
-		fuse.Unmount(*mountOptions.dir)
+		fuse.Unmount(dir)
 	}
 
 	// check if the mount process has an error to report
 	<-c.Ready
 	if err := c.MountError; err != nil {
-		glog.Fatal(err)
+		glog.V(0).Infof("mount process: %v", err)
+		daemonize.SignalOutcome(err)
+		return true
 	}
 
 	return true

weed/command/s3.go

@@ -4,7 +4,11 @@ import (
 	"net/http"
 	"time"
 
+	"github.com/chrislusf/seaweedfs/weed/security"
+	"github.com/spf13/viper"
+
 	"fmt"
+
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/s3api"
 	"github.com/chrislusf/seaweedfs/weed/util"
@@ -12,12 +16,11 @@ import (
 )
 
 var (
-	s3options S3Options
+	s3StandaloneOptions S3Options
 )
 
 type S3Options struct {
 	filer            *string
-	filerGrpcPort    *int
 	filerBucketsPath *string
 	port             *int
 	domainName       *string
@@ -27,13 +30,12 @@ type S3Options struct {
 
 func init() {
 	cmdS3.Run = runS3 // break init cycle
-	s3options.filer = cmdS3.Flag.String("filer", "localhost:8888", "filer server address")
-	s3options.filerGrpcPort = cmdS3.Flag.Int("filer.grpcPort", 0, "filer server grpc port, default to filer http port plus 10000")
-	s3options.filerBucketsPath = cmdS3.Flag.String("filer.dir.buckets", "/buckets", "folder on filer to store all buckets")
-	s3options.port = cmdS3.Flag.Int("port", 8333, "s3options server http listen port")
-	s3options.domainName = cmdS3.Flag.String("domainName", "", "suffix of the host name, {bucket}.{domainName}")
-	s3options.tlsPrivateKey = cmdS3.Flag.String("key.file", "", "path to the TLS private key file")
-	s3options.tlsCertificate = cmdS3.Flag.String("cert.file", "", "path to the TLS certificate file")
+	s3StandaloneOptions.filer = cmdS3.Flag.String("filer", "localhost:8888", "filer server address")
+	s3StandaloneOptions.filerBucketsPath = cmdS3.Flag.String("filer.dir.buckets", "/buckets", "folder on filer to store all buckets")
+	s3StandaloneOptions.port = cmdS3.Flag.Int("port", 8333, "s3 server http listen port")
+	s3StandaloneOptions.domainName = cmdS3.Flag.String("domainName", "", "suffix of the host name, {bucket}.{domainName}")
+	s3StandaloneOptions.tlsPrivateKey = cmdS3.Flag.String("key.file", "", "path to the TLS private key file")
+	s3StandaloneOptions.tlsCertificate = cmdS3.Flag.String("cert.file", "", "path to the TLS certificate file")
 }
 
 var cmdS3 = &Command{
@@ -46,7 +48,15 @@ var cmdS3 = &Command{
 
 func runS3(cmd *Command, args []string) bool {
 
-	filerGrpcAddress, err := parseFilerGrpcAddress(*s3options.filer, *s3options.filerGrpcPort)
+	util.LoadConfiguration("security", false)
+
+	return s3StandaloneOptions.startS3Server()
+
+}
+
+func (s3opt *S3Options) startS3Server() bool {
+
+	filerGrpcAddress, err := parseFilerGrpcAddress(*s3opt.filer)
 	if err != nil {
 		glog.Fatal(err)
 		return false
@@ -55,10 +65,11 @@ func runS3(cmd *Command, args []string) bool {
 	router := mux.NewRouter().SkipClean(true)
 
 	_, s3ApiServer_err := s3api.NewS3ApiServer(router, &s3api.S3ApiServerOption{
-		Filer:            *s3options.filer,
+		Filer:            *s3opt.filer,
 		FilerGrpcAddress: filerGrpcAddress,
-		DomainName:       *s3options.domainName,
-		BucketsPath:      *s3options.filerBucketsPath,
+		DomainName:       *s3opt.domainName,
+		BucketsPath:      *s3opt.filerBucketsPath,
+		GrpcDialOption:   security.LoadClientTLS(viper.Sub("grpc"), "client"),
 	})
 	if s3ApiServer_err != nil {
 		glog.Fatalf("S3 API Server startup error: %v", s3ApiServer_err)
@@ -66,22 +77,22 @@ func runS3(cmd *Command, args []string) bool {
 
 	httpS := &http.Server{Handler: router}
 
-	listenAddress := fmt.Sprintf(":%d", *s3options.port)
+	listenAddress := fmt.Sprintf(":%d", *s3opt.port)
 	s3ApiListener, err := util.NewListener(listenAddress, time.Duration(10)*time.Second)
 	if err != nil {
 		glog.Fatalf("S3 API Server listener on %s error: %v", listenAddress, err)
 	}
 
-	if *s3options.tlsPrivateKey != "" {
-		if err = httpS.ServeTLS(s3ApiListener, *s3options.tlsCertificate, *s3options.tlsPrivateKey); err != nil {
+	if *s3opt.tlsPrivateKey != "" {
+		glog.V(0).Infof("Start Seaweed S3 API Server %s at https port %d", util.VERSION, *s3opt.port)
+		if err = httpS.ServeTLS(s3ApiListener, *s3opt.tlsCertificate, *s3opt.tlsPrivateKey); err != nil {
 			glog.Fatalf("S3 API Server Fail to serve: %v", err)
 		}
-		glog.V(0).Infof("Start Seaweed S3 API Server %s at https port %d", util.VERSION, *s3options.port)
 	} else {
+		glog.V(0).Infof("Start Seaweed S3 API Server %s at http port %d", util.VERSION, *s3opt.port)
 		if err = httpS.Serve(s3ApiListener); err != nil {
 			glog.Fatalf("S3 API Server Fail to serve: %v", err)
 		}
-		glog.V(0).Infof("Start Seaweed S3 API Server %s at http port %d", util.VERSION, *s3options.port)
 	}
 
 	return true

weed/command/scaffold.go

@@ -10,7 +10,7 @@ func init() {
 }
 
 var cmdScaffold = &Command{
-	UsageLine: "scaffold [filer]",
+	UsageLine: "scaffold -config=[filer|notification|replication|security|master]",
 	Short:     "generate basic configuration files",
 	Long: `Generate filer.toml with all possible configurations for you to customize.
 
@@ -19,7 +19,7 @@ var cmdScaffold = &Command{
 
 var (
 	outputPath = cmdScaffold.Flag.String("output", "", "if not empty, save the configuration file to this directory")
-	config     = cmdScaffold.Flag.String("config", "filer", "[filer|notification|replication] the configuration file to generate")
+	config     = cmdScaffold.Flag.String("config", "filer", "[filer|notification|replication|security|master] the configuration file to generate")
 )
 
 func runScaffold(cmd *Command, args []string) bool {
@@ -32,6 +32,10 @@ func runScaffold(cmd *Command, args []string) bool {
 		content = NOTIFICATION_TOML_EXAMPLE
 	case "replication":
 		content = REPLICATION_TOML_EXAMPLE
+	case "security":
+		content = SECURITY_TOML_EXAMPLE
+	case "master":
+		content = MASTER_TOML_EXAMPLE
 	}
 	if content == "" {
 		println("need a valid -config option")
@@ -55,27 +59,34 @@ const (
 #    $HOME/.seaweedfs/filer.toml
 #    /etc/seaweedfs/filer.toml
 
-[memory]
-# local in memory, mostly for testing purpose
-enabled = false
+####################################################
+# Customizable filer server options
+####################################################
+[filer.options]
+# with http DELETE, by default the filer would check whether a folder is empty.
+# recursive_delete will delete all sub folders and files, similar to "rm -Rf"
+recursive_delete = false
 
-[leveldb]
+
+####################################################
+# The following are filer store options
+####################################################
+
+[leveldb2]
 # local on disk, mostly for simple single-machine setup, fairly scalable
+# faster than previous leveldb, recommended.
 enabled = true
 dir = "."					# directory to store level db files
 
-####################################################
-# multiple filers on shared storage, fairly scalable
-####################################################
-
-[mysql]
+[mysql]  # or tidb
 # CREATE TABLE IF NOT EXISTS filemeta (
-#   dirhash     BIGINT        COMMENT 'first 64 bits of MD5 hash value of directory field',
-#   name        VARCHAR(1000) COMMENT 'directory or file name',
-#   directory   VARCHAR(4096) COMMENT 'full path to parent directory',
-#   meta        BLOB,
+#   dirhash     BIGINT         COMMENT 'first 64 bits of MD5 hash value of directory field',
+#   name        VARCHAR(1000)  COMMENT 'directory or file name',
+#   directory   TEXT           COMMENT 'full path to parent directory',
+#   meta        LONGBLOB,
 #   PRIMARY KEY (dirhash, name)
 # ) DEFAULT CHARSET=utf8;
+
 enabled = false
 hostname = "localhost"
 port = 3306
@@ -84,12 +95,13 @@ password = ""
 database = ""              # create or use an existing database
 connection_max_idle = 2
 connection_max_open = 100
+interpolateParams = false
 
-[postgres]
+[postgres] # or cockroachdb
 # CREATE TABLE IF NOT EXISTS filemeta (
 #   dirhash     BIGINT,
-#   name        VARCHAR(1000),
-#   directory   VARCHAR(4096),
+#   name        VARCHAR(65535),
+#   directory   VARCHAR(65535),
 #   meta        bytea,
 #   PRIMARY KEY (dirhash, name)
 # );
@@ -120,7 +132,7 @@ hosts=[
 enabled = false
 address  = "localhost:6379"
 password = ""
-db = 0
+database = 0
 
 [redis_cluster]
 enabled = false
@@ -132,6 +144,21 @@ addresses = [
     "localhost:30005",
     "localhost:30006",
 ]
+password = ""
+// allows reads from slave servers or the master, but all writes still go to the master
+readOnly = true
+// automatically use the closest Redis server for reads
+routeByLatency = true
+
+[etcd]
+enabled = false
+servers = "localhost:2379"
+timeout = "3s"
+
+[tikv]
+enabled = false
+pdAddress = "192.168.199.113:2379"
+
 
 `
 
@@ -178,6 +205,17 @@ google_application_credentials = "/path/to/x.json" # path to json credential fil
 project_id = ""                       # an existing project id
 topic = "seaweedfs_filer_topic"       # a topic, auto created if does not exists
 
+[notification.gocdk_pub_sub]
+# The Go Cloud Development Kit (https://gocloud.dev).
+# PubSub API (https://godoc.org/gocloud.dev/pubsub).
+# Supports AWS SNS/SQS, Azure Service Bus, Google PubSub, NATS and RabbitMQ.
+enabled = false
+# This URL will Dial the RabbitMQ server at the URL in the environment
+# variable RABBIT_SERVER_URL and open the exchange "myexchange".
+# The exchange must have already been created by some other means, like
+# the RabbitMQ management plugin.
+topic_url = "rabbit://myexchange"
+sub_url = "rabbit://myqueue"
 `
 
 	REPLICATION_TOML_EXAMPLE = `
@@ -191,19 +229,25 @@ topic = "seaweedfs_filer_topic"       # a topic, auto created if does not exists
 [source.filer]
 enabled = true
 grpcAddress = "localhost:18888"
-directory = "/buckets"    # all files under this directory tree are replicated
+# all files under this directory tree are replicated.
+# this is not a directory on your hard drive, but on your filer.
+# i.e., all files with this "prefix" are sent to notification message queue.
+directory = "/buckets"
 
 [sink.filer]
 enabled = false
 grpcAddress = "localhost:18888"
-directory = "/backup"    # all replicated files are under this directory tree
+# all replicated files are under this directory tree
+# this is not a directory on your hard drive, but on your filer.
+# i.e., all received files will be "prefixed" to this directory.
+directory = "/backup"
 replication = ""
 collection = ""
 ttlSec = 0
 
 [sink.s3]
 # read credentials doc at https://docs.aws.amazon.com/sdk-for-go/v1/developer-guide/sessions.html
-# default loads credentials from the shared credentials file (~/.aws/credentials). 
+# default loads credentials from the shared credentials file (~/.aws/credentials).
 enabled = false
 aws_access_key_id     = ""     # if empty, loads from the shared credentials file (~/.aws/credentials).
 aws_secret_access_key = ""     # if empty, loads from the shared credentials file (~/.aws/credentials).
@@ -233,5 +277,97 @@ b2_master_application_key  = ""
 bucket = "mybucket"            # an existing bucket
 directory = "/"                # destination directory
 
+`
+
+	SECURITY_TOML_EXAMPLE = `
+# Put this file to one of the location, with descending priority
+#    ./security.toml
+#    $HOME/.seaweedfs/security.toml
+#    /etc/seaweedfs/security.toml
+# this file is read by master, volume server, and filer
+
+# the jwt signing key is read by master and volume server.
+# a jwt defaults to expire after 10 seconds.
+[jwt.signing]
+key = ""
+expires_after_seconds = 10           # seconds
+
+# jwt for read is only supported with master+volume setup. Filer does not support this mode.
+[jwt.signing.read]
+key = ""
+expires_after_seconds = 10           # seconds
+
+# all grpc tls authentications are mutual
+# the values for the following ca, cert, and key are paths to the PERM files.
+# the host name is not checked, so the PERM files can be shared.
+[grpc]
+ca = ""
+
+[grpc.volume]
+cert = ""
+key  = ""
+
+[grpc.master]
+cert = ""
+key  = ""
+
+[grpc.filer]
+cert = ""
+key  = ""
+
+# use this for any place needs a grpc client
+# i.e., "weed backup|benchmark|filer.copy|filer.replicate|mount|s3|upload"
+[grpc.client]
+cert = ""
+key  = ""
+
+
+# volume server https options
+# Note: work in progress!
+#     this does not work with other clients, e.g., "weed filer|mount" etc, yet.
+[https.client]
+enabled = true
+[https.volume]
+cert = ""
+key  = ""
+
+
+`
+
+	MASTER_TOML_EXAMPLE = `
+# Put this file to one of the location, with descending priority
+#    ./master.toml
+#    $HOME/.seaweedfs/master.toml
+#    /etc/seaweedfs/master.toml
+# this file is read by master
+
+[master.maintenance]
+# periodically run these scripts are the same as running them from 'weed shell'
+scripts = """
+  ec.encode -fullPercent=95 -quietFor=1h
+  ec.rebuild -force
+  ec.balance -force
+  volume.balance -force
+"""
+sleep_minutes = 17          # sleep minutes between each script execution
+
+[master.filer]
+default_filer_url = "http://localhost:8888/"
+
+[master.sequencer]
+type = "memory"     # Choose [memory|etcd] type for storing the file id sequence
+# when sequencer.type = etcd, set listen client urls of etcd cluster that store file id sequence
+# example : http://127.0.0.1:2379,http://127.0.0.1:2389
+sequencer_etcd_urls = "http://127.0.0.1:2379"
+
+
+[storage.backend]
+	[storage.backend.s3.default]
+	enabled = false
+	aws_access_key_id     = ""     # if empty, loads from the shared credentials file (~/.aws/credentials).
+	aws_secret_access_key = ""     # if empty, loads from the shared credentials file (~/.aws/credentials).
+	region = "us-east-2"
+	bucket = "your_bucket_name"    # an existing bucket
+
 `
 )

weed/command/scaffold_test.go

@@ -0,0 +1,44 @@
+package command
+
+import (
+	"bytes"
+	"fmt"
+	"testing"
+
+	"github.com/spf13/viper"
+)
+
+func TestReadingTomlConfiguration(t *testing.T) {
+
+	viper.SetConfigType("toml")
+
+	// any approach to require this configuration into your program.
+	var tomlExample = []byte(`
+[database]
+server = "192.168.1.1"
+ports = [ 8001, 8001, 8002 ]
+connection_max = 5000
+enabled = true
+
+[servers]
+
+  # You can indent as you please. Tabs or spaces. TOML don't care.
+  [servers.alpha]
+  ip = "10.0.0.1"
+  dc = "eqdc10"
+
+  [servers.beta]
+  ip = "10.0.0.2"
+  dc = "eqdc10"
+
+`)
+
+	viper.ReadConfig(bytes.NewBuffer(tomlExample))
+
+	fmt.Printf("database is %v\n", viper.Get("database"))
+	fmt.Printf("servers is %v\n", viper.GetStringMap("servers"))
+
+	alpha := viper.Sub("servers.alpha")
+
+	fmt.Printf("alpha ip is %v\n", alpha.GetString("ip"))
+}

weed/command/server.go

@@ -1,22 +1,15 @@
 package command
 
 import (
-	"net/http"
+	"fmt"
 	"os"
 	"runtime"
 	"runtime/pprof"
-	"strconv"
 	"strings"
-	"sync"
 	"time"
 
 	"github.com/chrislusf/seaweedfs/weed/glog"
-	"github.com/chrislusf/seaweedfs/weed/pb/master_pb"
-	"github.com/chrislusf/seaweedfs/weed/server"
 	"github.com/chrislusf/seaweedfs/weed/util"
-	"github.com/gorilla/mux"
-	"github.com/soheilhy/cmux"
-	"google.golang.org/grpc/reflection"
 )
 
 type ServerOptions struct {
@@ -26,7 +19,9 @@ type ServerOptions struct {
 
 var (
 	serverOptions ServerOptions
+	masterOptions MasterOptions
 	filerOptions  FilerOptions
+	s3Options     S3Options
 )
 
 func init() {
@@ -35,50 +30,52 @@ func init() {
 
 var cmdServer = &Command{
 	UsageLine: "server -port=8080 -dir=/tmp -volume.max=5 -ip=server_name",
-	Short:     "start a server, including volume server, and automatically elect a master server",
+	Short:     "start a master server, a volume server, and optionally a filer and a S3 gateway",
 	Long: `start both a volume server to provide storage spaces
   and a master server to provide volume=>location mapping service and sequence number of file ids
 
   This is provided as a convenient way to start both volume server and master server.
-  The servers are exactly the same as starting them separately.
+  The servers acts exactly the same as starting them separately.
+  So other volume servers can connect to this master server also.
 
-  So other volume servers can use this embedded master server also.
-
-  Optionally, one filer server can be started. Logically, filer servers should not be in a cluster.
-  They run with meta data on disk, not shared. So each filer server is different.
+  Optionally, a filer server can be started.
+  Also optionally, a S3 gateway can be started.
 
   `,
 }
 
 var (
-	serverIp                      = cmdServer.Flag.String("ip", "localhost", "ip or server name")
-	serverBindIp                  = cmdServer.Flag.String("ip.bind", "0.0.0.0", "ip address to bind to")
-	serverMaxCpu                  = cmdServer.Flag.Int("maxCpu", 0, "maximum number of CPUs. 0 means all available CPUs")
-	serverTimeout                 = cmdServer.Flag.Int("idleTimeout", 30, "connection idle seconds")
-	serverDataCenter              = cmdServer.Flag.String("dataCenter", "", "current volume server's data center name")
-	serverRack                    = cmdServer.Flag.String("rack", "", "current volume server's rack name")
-	serverWhiteListOption         = cmdServer.Flag.String("whiteList", "", "comma separated Ip addresses having write permission. No limit if empty.")
-	serverPeers                   = cmdServer.Flag.String("master.peers", "", "all master nodes in comma separated ip:masterPort list")
-	serverSecureKey               = cmdServer.Flag.String("secure.secret", "", "secret to encrypt Json Web Token(JWT)")
-	serverGarbageThreshold        = cmdServer.Flag.Float64("garbageThreshold", 0.3, "threshold to vacuum and reclaim spaces")
-	masterPort                    = cmdServer.Flag.Int("master.port", 9333, "master server http listen port")
-	masterMetaFolder              = cmdServer.Flag.String("master.dir", "", "data directory to store meta data, default to same as -dir specified")
-	masterVolumeSizeLimitMB       = cmdServer.Flag.Uint("master.volumeSizeLimitMB", 30*1000, "Master stops directing writes to oversized volumes.")
-	masterVolumePreallocate       = cmdServer.Flag.Bool("master.volumePreallocate", false, "Preallocate disk space for volumes.")
-	masterDefaultReplicaPlacement = cmdServer.Flag.String("master.defaultReplicaPlacement", "000", "Default replication type if not specified.")
-	volumeDataFolders             = cmdServer.Flag.String("dir", os.TempDir(), "directories to store data files. dir[,dir]...")
-	volumeMaxDataVolumeCounts     = cmdServer.Flag.String("volume.max", "7", "maximum numbers of volumes, count[,count]...")
-	pulseSeconds                  = cmdServer.Flag.Int("pulseSeconds", 5, "number of seconds between heartbeats")
-	isStartingFiler               = cmdServer.Flag.Bool("filer", false, "whether to start filer")
+	serverIp                  = cmdServer.Flag.String("ip", "localhost", "ip or server name")
+	serverBindIp              = cmdServer.Flag.String("ip.bind", "0.0.0.0", "ip address to bind to")
+	serverTimeout             = cmdServer.Flag.Int("idleTimeout", 30, "connection idle seconds")
+	serverDataCenter          = cmdServer.Flag.String("dataCenter", "", "current volume server's data center name")
+	serverRack                = cmdServer.Flag.String("rack", "", "current volume server's rack name")
+	serverWhiteListOption     = cmdServer.Flag.String("whiteList", "", "comma separated Ip addresses having write permission. No limit if empty.")
+	serverDisableHttp         = cmdServer.Flag.Bool("disableHttp", false, "disable http requests, only gRPC operations are allowed.")
+	volumeDataFolders         = cmdServer.Flag.String("dir", os.TempDir(), "directories to store data files. dir[,dir]...")
+	volumeMaxDataVolumeCounts = cmdServer.Flag.String("volume.max", "7", "maximum numbers of volumes, count[,count]...")
+	pulseSeconds              = cmdServer.Flag.Int("pulseSeconds", 5, "number of seconds between heartbeats")
+	isStartingFiler           = cmdServer.Flag.Bool("filer", false, "whether to start filer")
+	isStartingS3              = cmdServer.Flag.Bool("s3", false, "whether to start S3 gateway")
 
 	serverWhiteList []string
 )
 
 func init() {
 	serverOptions.cpuprofile = cmdServer.Flag.String("cpuprofile", "", "cpu profile output file")
+
+	masterOptions.port = cmdServer.Flag.Int("master.port", 9333, "master server http listen port")
+	masterOptions.metaFolder = cmdServer.Flag.String("master.dir", "", "data directory to store meta data, default to same as -dir specified")
+	masterOptions.peers = cmdServer.Flag.String("master.peers", "", "all master nodes in comma separated ip:masterPort list")
+	masterOptions.volumeSizeLimitMB = cmdServer.Flag.Uint("master.volumeSizeLimitMB", 30*1000, "Master stops directing writes to oversized volumes.")
+	masterOptions.volumePreallocate = cmdServer.Flag.Bool("master.volumePreallocate", false, "Preallocate disk space for volumes.")
+	masterOptions.defaultReplication = cmdServer.Flag.String("master.defaultReplication", "000", "Default replication type if not specified.")
+	masterOptions.garbageThreshold = cmdServer.Flag.Float64("garbageThreshold", 0.3, "threshold to vacuum and reclaim spaces")
+	masterOptions.metricsAddress = cmdServer.Flag.String("metrics.address", "", "Prometheus gateway address")
+	masterOptions.metricsIntervalSec = cmdServer.Flag.Int("metrics.intervalSeconds", 15, "Prometheus push interval in seconds")
+
 	filerOptions.collection = cmdServer.Flag.String("filer.collection", "", "all data will be stored in this collection")
 	filerOptions.port = cmdServer.Flag.Int("filer.port", 8888, "filer server http listen port")
-	filerOptions.grpcPort = cmdServer.Flag.Int("filer.port.grpc", 0, "filer grpc server listen port, default to http port + 10000")
 	filerOptions.publicPort = cmdServer.Flag.Int("filer.port.public", 0, "filer server public http listen port")
 	filerOptions.defaultReplicaPlacement = cmdServer.Flag.String("filer.defaultReplicaPlacement", "", "Default replication type if not specified during runtime.")
 	filerOptions.redirectOnRead = cmdServer.Flag.Bool("filer.redirectOnRead", false, "whether proxy or redirect to volume server during file GET request")
@@ -88,15 +85,25 @@ func init() {
 
 	serverOptions.v.port = cmdServer.Flag.Int("volume.port", 8080, "volume server http listen port")
 	serverOptions.v.publicPort = cmdServer.Flag.Int("volume.port.public", 0, "volume server public port")
-	serverOptions.v.indexType = cmdServer.Flag.String("volume.index", "memory", "Choose [memory|leveldb|boltdb|btree] mode for memory~performance balance.")
-	serverOptions.v.fixJpgOrientation = cmdServer.Flag.Bool("volume.images.fix.orientation", true, "Adjust jpg orientation when uploading.")
+	serverOptions.v.indexType = cmdServer.Flag.String("volume.index", "memory", "Choose [memory|leveldb|leveldbMedium|leveldbLarge] mode for memory~performance balance.")
+	serverOptions.v.fixJpgOrientation = cmdServer.Flag.Bool("volume.images.fix.orientation", false, "Adjust jpg orientation when uploading.")
 	serverOptions.v.readRedirect = cmdServer.Flag.Bool("volume.read.redirect", true, "Redirect moved or non-local volumes.")
+	serverOptions.v.compactionMBPerSecond = cmdServer.Flag.Int("volume.compactionMBps", 0, "limit compaction speed in mega bytes per second")
 	serverOptions.v.publicUrl = cmdServer.Flag.String("volume.publicUrl", "", "publicly accessible address")
 
+	s3Options.filerBucketsPath = cmdServer.Flag.String("s3.filer.dir.buckets", "/buckets", "folder on filer to store all buckets")
+	s3Options.port = cmdServer.Flag.Int("s3.port", 8333, "s3 server http listen port")
+	s3Options.domainName = cmdServer.Flag.String("s3.domainName", "", "suffix of the host name, {bucket}.{domainName}")
+	s3Options.tlsPrivateKey = cmdServer.Flag.String("s3.key.file", "", "path to the TLS private key file")
+	s3Options.tlsCertificate = cmdServer.Flag.String("s3.cert.file", "", "path to the TLS certificate file")
+
 }
 
 func runServer(cmd *Command, args []string) bool {
-	filerOptions.secretKey = serverSecureKey
+
+	util.LoadConfiguration("security", false)
+	util.LoadConfiguration("master", false)
+
 	if *serverOptions.cpuprofile != "" {
 		f, err := os.Create(*serverOptions.cpuprofile)
 		if err != nil {
@@ -110,40 +117,56 @@ func runServer(cmd *Command, args []string) bool {
 		*isStartingFiler = true
 	}
 
-	master := *serverIp + ":" + strconv.Itoa(*masterPort)
-	filerOptions.ip = serverIp
+	if *isStartingS3 {
+		*isStartingFiler = true
+	}
+
+	_, peerList := checkPeers(*serverIp, *masterOptions.port, *masterOptions.peers)
+	peers := strings.Join(peerList, ",")
+	masterOptions.peers = &peers
+
+	masterOptions.ip = serverIp
+	masterOptions.ipBind = serverBindIp
+	filerOptions.masters = &peers
+	filerOptions.ip = serverBindIp
 	serverOptions.v.ip = serverIp
 	serverOptions.v.bindIp = serverBindIp
-	serverOptions.v.masters = &master
+	serverOptions.v.masters = &peers
 	serverOptions.v.idleConnectionTimeout = serverTimeout
-	serverOptions.v.maxCpu = serverMaxCpu
 	serverOptions.v.dataCenter = serverDataCenter
 	serverOptions.v.rack = serverRack
+
 	serverOptions.v.pulseSeconds = pulseSeconds
+	masterOptions.pulseSeconds = pulseSeconds
+
+	masterOptions.whiteList = serverWhiteListOption
 
 	filerOptions.dataCenter = serverDataCenter
+	filerOptions.disableHttp = serverDisableHttp
+	masterOptions.disableHttp = serverDisableHttp
+
+	filerAddress := fmt.Sprintf("%s:%d", *serverIp, *filerOptions.port)
+	s3Options.filer = &filerAddress
 
 	if *filerOptions.defaultReplicaPlacement == "" {
-		*filerOptions.defaultReplicaPlacement = *masterDefaultReplicaPlacement
+		*filerOptions.defaultReplicaPlacement = *masterOptions.defaultReplication
 	}
 
-	if *serverMaxCpu < 1 {
-		*serverMaxCpu = runtime.NumCPU()
-	}
-	runtime.GOMAXPROCS(*serverMaxCpu)
+	runtime.GOMAXPROCS(runtime.NumCPU())
 
 	folders := strings.Split(*volumeDataFolders, ",")
 
-	if *masterVolumeSizeLimitMB > 30*1000 {
+	if *masterOptions.volumeSizeLimitMB > util.VolumeSizeLimitGB*1000 {
 		glog.Fatalf("masterVolumeSizeLimitMB should be less than 30000")
 	}
 
-	if *masterMetaFolder == "" {
-		*masterMetaFolder = folders[0]
+	if *masterOptions.metaFolder == "" {
+		*masterOptions.metaFolder = folders[0]
 	}
-	if err := util.TestFolderWritable(*masterMetaFolder); err != nil {
-		glog.Fatalf("Check Meta Folder (-mdir=\"%s\") Writable: %s", *masterMetaFolder, err)
+	if err := util.TestFolderWritable(*masterOptions.metaFolder); err != nil {
+		glog.Fatalf("Check Meta Folder (-mdir=\"%s\") Writable: %s", *masterOptions.metaFolder, err)
 	}
+	filerOptions.defaultLevelDbDirectory = masterOptions.metaFolder
 
 	if *serverWhiteListOption != "" {
 		serverWhiteList = strings.Split(*serverWhiteListOption, ",")
@@ -158,63 +181,21 @@ func runServer(cmd *Command, args []string) bool {
 		}()
 	}
 
-	var raftWaitForMaster sync.WaitGroup
-	var volumeWait sync.WaitGroup
-
-	raftWaitForMaster.Add(1)
-	volumeWait.Add(1)
-
-	go func() {
-		r := mux.NewRouter()
-		ms := weed_server.NewMasterServer(r, *masterPort, *masterMetaFolder,
-			*masterVolumeSizeLimitMB, *masterVolumePreallocate,
-			*pulseSeconds, *masterDefaultReplicaPlacement, *serverGarbageThreshold,
-			serverWhiteList, *serverSecureKey,
-		)
-
-		glog.V(0).Infoln("Start Seaweed Master", util.VERSION, "at", *serverIp+":"+strconv.Itoa(*masterPort))
-		masterListener, e := util.NewListener(*serverBindIp+":"+strconv.Itoa(*masterPort), 0)
-		if e != nil {
-			glog.Fatalf("Master startup error: %v", e)
-		}
-
+	if *isStartingS3 {
 		go func() {
-			raftWaitForMaster.Wait()
-			time.Sleep(100 * time.Millisecond)
-			myAddress, peers := checkPeers(*serverIp, *masterPort, *serverPeers)
-			raftServer := weed_server.NewRaftServer(r, peers, myAddress, *masterMetaFolder, ms.Topo, *pulseSeconds)
-			ms.SetRaftServer(raftServer)
-			volumeWait.Done()
+			time.Sleep(2 * time.Second)
+
+			s3Options.startS3Server()
+
 		}()
+	}
 
-		raftWaitForMaster.Done()
+	// start volume server
+	{
+		go serverOptions.v.startVolumeServer(*volumeDataFolders, *volumeMaxDataVolumeCounts, *serverWhiteListOption)
+	}
 
-		// start grpc and http server
-		m := cmux.New(masterListener)
-
-		grpcL := m.Match(cmux.HTTP2HeaderField("content-type", "application/grpc"))
-		httpL := m.Match(cmux.Any())
-
-		// Create your protocol servers.
-		grpcS := util.NewGrpcServer()
-		master_pb.RegisterSeaweedServer(grpcS, ms)
-		reflection.Register(grpcS)
-
-		httpS := &http.Server{Handler: r}
-
-		go grpcS.Serve(grpcL)
-		go httpS.Serve(httpL)
-
-		if err := m.Serve(); err != nil {
-			glog.Fatalf("master server failed to serve: %v", err)
-		}
-
-	}()
-
-	volumeWait.Wait()
-	time.Sleep(100 * time.Millisecond)
-
-	serverOptions.v.startVolumeServer(*volumeDataFolders, *volumeMaxDataVolumeCounts, *serverWhiteListOption)
+	startMaster(masterOptions, serverWhiteList)
 
 	return true
 }

weed/command/shell.go

@@ -1,61 +1,47 @@
 package command
 
 import (
-	"bufio"
 	"fmt"
-	"os"
 
-	"github.com/chrislusf/seaweedfs/weed/glog"
+	"github.com/chrislusf/seaweedfs/weed/security"
+	"github.com/chrislusf/seaweedfs/weed/shell"
+	"github.com/chrislusf/seaweedfs/weed/util"
+	"github.com/spf13/viper"
+)
+
+var (
+	shellOptions         shell.ShellOptions
+	shellInitialFilerUrl *string
 )
 
 func init() {
 	cmdShell.Run = runShell // break init cycle
+	shellOptions.Masters = cmdShell.Flag.String("master", "localhost:9333", "comma-separated master servers")
+	shellInitialFilerUrl = cmdShell.Flag.String("filer.url", "http://localhost:8888/", "initial filer url")
 }
 
 var cmdShell = &Command{
 	UsageLine: "shell",
-	Short:     "run interactive commands, now just echo",
-	Long: `run interactive commands.
+	Short:     "run interactive administrative commands",
+	Long: `run interactive administrative commands.
 
   `,
 }
 
-var ()
-
 func runShell(command *Command, args []string) bool {
-	r := bufio.NewReader(os.Stdin)
-	o := bufio.NewWriter(os.Stdout)
-	e := bufio.NewWriter(os.Stderr)
-	prompt := func() {
-		var err error
-		if _, err = o.WriteString("> "); err != nil {
-			glog.V(0).Infoln("error writing to stdout:", err)
-		}
-		if err = o.Flush(); err != nil {
-			glog.V(0).Infoln("error flushing stdout:", err)
-		}
-	}
-	readLine := func() string {
-		ret, err := r.ReadString('\n')
-		if err != nil {
-			fmt.Fprint(e, err)
-			os.Exit(1)
-		}
-		return ret
-	}
-	execCmd := func(cmd string) int {
-		if cmd != "" {
-			if _, err := o.WriteString(cmd); err != nil {
-				glog.V(0).Infoln("error writing to stdout:", err)
-			}
-		}
-		return 0
+
+	util.LoadConfiguration("security", false)
+	shellOptions.GrpcDialOption = security.LoadClientTLS(viper.Sub("grpc"), "client")
+
+	var filerPwdErr error
+	shellOptions.FilerHost, shellOptions.FilerPort, shellOptions.Directory, filerPwdErr = util.ParseFilerUrl(*shellInitialFilerUrl)
+	if filerPwdErr != nil {
+		fmt.Printf("failed to parse url filer.url=%s : %v\n", *shellInitialFilerUrl, filerPwdErr)
+		return false
 	}
 
-	cmd := ""
-	for {
-		prompt()
-		cmd = readLine()
-		execCmd(cmd)
-	}
+	shell.RunShell(shellOptions)
+
+	return true
+
 }

weed/command/upload.go

@@ -6,8 +6,11 @@ import (
 	"os"
 	"path/filepath"
 
-	"github.com/chrislusf/seaweedfs/weed/operation"
 	"github.com/chrislusf/seaweedfs/weed/security"
+	"github.com/chrislusf/seaweedfs/weed/util"
+	"github.com/spf13/viper"
+
+	"github.com/chrislusf/seaweedfs/weed/operation"
 )
 
 var (
@@ -23,7 +26,6 @@ type UploadOptions struct {
 	dataCenter  *string
 	ttl         *string
 	maxMB       *int
-	secretKey   *string
 }
 
 func init() {
@@ -36,8 +38,7 @@ func init() {
 	upload.collection = cmdUpload.Flag.String("collection", "", "optional collection name")
 	upload.dataCenter = cmdUpload.Flag.String("dataCenter", "", "optional data center name")
 	upload.ttl = cmdUpload.Flag.String("ttl", "", "time to live, e.g.: 1m, 1h, 1d, 1M, 1y")
-	upload.maxMB = cmdUpload.Flag.Int("maxMB", 0, "split files larger than the limit")
-	upload.secretKey = cmdUpload.Flag.String("secure.secret", "", "secret to encrypt Json Web Token(JWT)")
+	upload.maxMB = cmdUpload.Flag.Int("maxMB", 32, "split files larger than the limit")
 }
 
 var cmdUpload = &Command{
@@ -53,18 +54,17 @@ var cmdUpload = &Command{
   All files under the folder and subfolders will be uploaded, each with its own file key.
   Optional parameter "-include" allows you to specify the file name patterns.
 
-  If any file has a ".gz" extension, the content are considered gzipped already, and will be stored as is.
-  This can save volume server's gzipped processing and allow customizable gzip compression level.
-  The file name will strip out ".gz" and stored. For example, "jquery.js.gz" will be stored as "jquery.js".
-
-  If "maxMB" is set to a positive number, files larger than it would be split into chunks and uploaded separatedly.
+  If "maxMB" is set to a positive number, files larger than it would be split into chunks and uploaded separately.
   The list of file ids of those chunks would be stored in an additional chunk, and this additional chunk's file id would be returned.
 
   `,
 }
 
 func runUpload(cmd *Command, args []string) bool {
-	secret := security.Secret(*upload.secretKey)
+
+	util.LoadConfiguration("security", false)
+	grpcDialOption := security.LoadClientTLS(viper.Sub("grpc"), "client")
+
 	if len(args) == 0 {
 		if *upload.dir == "" {
 			return false
@@ -81,9 +81,9 @@ func runUpload(cmd *Command, args []string) bool {
 					if e != nil {
 						return e
 					}
-					results, e := operation.SubmitFiles(*upload.master, parts,
+					results, e := operation.SubmitFiles(*upload.master, grpcDialOption, parts,
 						*upload.replication, *upload.collection, *upload.dataCenter,
-						*upload.ttl, *upload.maxMB, secret)
+						*upload.ttl, *upload.maxMB)
 					bytes, _ := json.Marshal(results)
 					fmt.Println(string(bytes))
 					if e != nil {
@@ -100,9 +100,9 @@ func runUpload(cmd *Command, args []string) bool {
 		if e != nil {
 			fmt.Println(e.Error())
 		}
-		results, _ := operation.SubmitFiles(*upload.master, parts,
+		results, _ := operation.SubmitFiles(*upload.master, grpcDialOption, parts,
 			*upload.replication, *upload.collection, *upload.dataCenter,
-			*upload.ttl, *upload.maxMB, secret)
+			*upload.ttl, *upload.maxMB)
 		bytes, _ := json.Marshal(results)
 		fmt.Println(string(bytes))
 	}

weed/command/volume.go

@@ -1,6 +1,7 @@
 package command
 
 import (
+	"fmt"
 	"net/http"
 	"os"
 	"runtime"
@@ -9,6 +10,11 @@ import (
 	"strings"
 	"time"
 
+	"github.com/chrislusf/seaweedfs/weed/security"
+	"github.com/chrislusf/seaweedfs/weed/util/httpdown"
+	"github.com/spf13/viper"
+	"google.golang.org/grpc"
+
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/pb/volume_server_pb"
 	"github.com/chrislusf/seaweedfs/weed/server"
@@ -32,7 +38,6 @@ type VolumeServerOptions struct {
 	masters               *string
 	pulseSeconds          *int
 	idleConnectionTimeout *int
-	maxCpu                *int
 	dataCenter            *string
 	rack                  *string
 	whiteList             []string
@@ -41,6 +46,7 @@ type VolumeServerOptions struct {
 	readRedirect          *bool
 	cpuProfile            *string
 	memProfile            *string
+	compactionMBPerSecond *int
 }
 
 func init() {
@@ -53,14 +59,14 @@ func init() {
 	v.masters = cmdVolume.Flag.String("mserver", "localhost:9333", "comma-separated master servers")
 	v.pulseSeconds = cmdVolume.Flag.Int("pulseSeconds", 5, "number of seconds between heartbeats, must be smaller than or equal to the master's setting")
 	v.idleConnectionTimeout = cmdVolume.Flag.Int("idleTimeout", 30, "connection idle seconds")
-	v.maxCpu = cmdVolume.Flag.Int("maxCpu", 0, "maximum number of CPUs. 0 means all available CPUs")
 	v.dataCenter = cmdVolume.Flag.String("dataCenter", "", "current volume server's data center name")
 	v.rack = cmdVolume.Flag.String("rack", "", "current volume server's rack name")
-	v.indexType = cmdVolume.Flag.String("index", "memory", "Choose [memory|leveldb|boltdb|btree] mode for memory~performance balance.")
-	v.fixJpgOrientation = cmdVolume.Flag.Bool("images.fix.orientation", true, "Adjust jpg orientation when uploading.")
+	v.indexType = cmdVolume.Flag.String("index", "memory", "Choose [memory|leveldb|leveldbMedium|leveldbLarge] mode for memory~performance balance.")
+	v.fixJpgOrientation = cmdVolume.Flag.Bool("images.fix.orientation", false, "Adjust jpg orientation when uploading.")
 	v.readRedirect = cmdVolume.Flag.Bool("read.redirect", true, "Redirect moved or non-local volumes.")
 	v.cpuProfile = cmdVolume.Flag.String("cpuprofile", "", "cpu profile output file")
 	v.memProfile = cmdVolume.Flag.String("memprofile", "", "memory profile output file")
+	v.compactionMBPerSecond = cmdVolume.Flag.Int("compactionMBps", 0, "limit background compaction or copying speed in mega bytes per second")
 }
 
 var cmdVolume = &Command{
@@ -78,10 +84,10 @@ var (
 )
 
 func runVolume(cmd *Command, args []string) bool {
-	if *v.maxCpu < 1 {
-		*v.maxCpu = runtime.NumCPU()
-	}
-	runtime.GOMAXPROCS(*v.maxCpu)
+
+	util.LoadConfiguration("security", false)
+
+	runtime.GOMAXPROCS(runtime.NumCPU())
 	util.SetupProfiling(*v.cpuProfile, *v.memProfile)
 
 	v.startVolumeServer(*volumeFolders, *maxVolumeCounts, *volumeWhiteListOption)
@@ -91,7 +97,7 @@ func runVolume(cmd *Command, args []string) bool {
 
 func (v VolumeServerOptions) startVolumeServer(volumeFolders, maxVolumeCounts, volumeWhiteListOption string) {
 
-	//Set multiple folders and each folder's max volume count limit'
+	// Set multiple folders and each folder's max volume count limit'
 	v.folders = strings.Split(volumeFolders, ",")
 	maxCountStrings := strings.Split(maxVolumeCounts, ",")
 	for _, maxString := range maxCountStrings {
@@ -110,7 +116,7 @@ func (v VolumeServerOptions) startVolumeServer(volumeFolders, maxVolumeCounts, v
 		}
 	}
 
-	//security related white list configuration
+	// security related white list configuration
 	if volumeWhiteListOption != "" {
 		v.whiteList = strings.Split(volumeWhiteListOption, ",")
 	}
@@ -125,11 +131,10 @@ func (v VolumeServerOptions) startVolumeServer(volumeFolders, maxVolumeCounts, v
 	if *v.publicUrl == "" {
 		*v.publicUrl = *v.ip + ":" + strconv.Itoa(*v.publicPort)
 	}
-	isSeperatedPublicPort := *v.publicPort != *v.port
 
 	volumeMux := http.NewServeMux()
 	publicVolumeMux := volumeMux
-	if isSeperatedPublicPort {
+	if v.isSeparatedPublicPort() {
 		publicVolumeMux = http.NewServeMux()
 	}
 
@@ -137,10 +142,10 @@ func (v VolumeServerOptions) startVolumeServer(volumeFolders, maxVolumeCounts, v
 	switch *v.indexType {
 	case "leveldb":
 		volumeNeedleMapKind = storage.NeedleMapLevelDb
-	case "boltdb":
-		volumeNeedleMapKind = storage.NeedleMapBoltDb
-	case "btree":
-		volumeNeedleMapKind = storage.NeedleMapBtree
+	case "leveldbMedium":
+		volumeNeedleMapKind = storage.NeedleMapLevelDbMedium
+	case "leveldbLarge":
+		volumeNeedleMapKind = storage.NeedleMapLevelDbLarge
 	}
 
 	masters := *v.masters
@@ -152,46 +157,134 @@ func (v VolumeServerOptions) startVolumeServer(volumeFolders, maxVolumeCounts, v
 		strings.Split(masters, ","), *v.pulseSeconds, *v.dataCenter, *v.rack,
 		v.whiteList,
 		*v.fixJpgOrientation, *v.readRedirect,
+		*v.compactionMBPerSecond,
 	)
 
-	listeningAddress := *v.bindIp + ":" + strconv.Itoa(*v.port)
-	glog.V(0).Infoln("Start Seaweed volume server", util.VERSION, "at", listeningAddress)
-	listener, e := util.NewListener(listeningAddress, time.Duration(*v.idleConnectionTimeout)*time.Second)
-	if e != nil {
-		glog.Fatalf("Volume server listener error:%v", e)
-	}
-	if isSeperatedPublicPort {
-		publicListeningAddress := *v.bindIp + ":" + strconv.Itoa(*v.publicPort)
-		glog.V(0).Infoln("Start Seaweed volume server", util.VERSION, "public at", publicListeningAddress)
-		publicListener, e := util.NewListener(publicListeningAddress, time.Duration(*v.idleConnectionTimeout)*time.Second)
-		if e != nil {
-			glog.Fatalf("Volume server listener error:%v", e)
+	// starting grpc server
+	grpcS := v.startGrpcService(volumeServer)
+
+	// starting public http server
+	var publicHttpDown httpdown.Server
+	if v.isSeparatedPublicPort() {
+		publicHttpDown = v.startPublicHttpService(publicVolumeMux)
+		if nil == publicHttpDown {
+			glog.Fatalf("start public http service failed")
 		}
-		go func() {
-			if e := http.Serve(publicListener, publicVolumeMux); e != nil {
-				glog.Fatalf("Volume server fail to serve public: %v", e)
-			}
-		}()
 	}
 
+	// starting the cluster http server
+	clusterHttpServer := v.startClusterHttpService(volumeMux)
+
+	stopChain := make(chan struct{})
 	util.OnInterrupt(func() {
+		fmt.Println("volume server has be killed")
+		var startTime time.Time
+
+		// firstly, stop the public http service to prevent from receiving new user request
+		if nil != publicHttpDown {
+			startTime = time.Now()
+			if err := publicHttpDown.Stop(); err != nil {
+				glog.Warningf("stop the public http server failed, %v", err)
+			}
+			delta := time.Now().Sub(startTime).Nanoseconds() / 1e6
+			glog.V(0).Infof("stop public http server, elapsed %dms", delta)
+		}
+
+		startTime = time.Now()
+		if err := clusterHttpServer.Stop(); err != nil {
+			glog.Warningf("stop the cluster http server failed, %v", err)
+		}
+		delta := time.Now().Sub(startTime).Nanoseconds() / 1e6
+		glog.V(0).Infof("graceful stop cluster http server, elapsed [%d]", delta)
+
+		startTime = time.Now()
+		grpcS.GracefulStop()
+		delta = time.Now().Sub(startTime).Nanoseconds() / 1e6
+		glog.V(0).Infof("graceful stop gRPC, elapsed [%d]", delta)
+
+		startTime = time.Now()
 		volumeServer.Shutdown()
+		delta = time.Now().Sub(startTime).Nanoseconds() / 1e6
+		glog.V(0).Infof("stop volume server, elapsed [%d]", delta)
+
 		pprof.StopCPUProfile()
+
+		close(stopChain) // notify exit
 	})
 
-	// starting grpc server
+	select {
+	case <-stopChain:
+	}
+	glog.Warningf("the volume server exit.")
+}
+
+// check whether configure the public port
+func (v VolumeServerOptions) isSeparatedPublicPort() bool {
+	return *v.publicPort != *v.port
+}
+
+func (v VolumeServerOptions) startGrpcService(vs volume_server_pb.VolumeServerServer) *grpc.Server {
 	grpcPort := *v.port + 10000
 	grpcL, err := util.NewListener(*v.bindIp+":"+strconv.Itoa(grpcPort), 0)
 	if err != nil {
 		glog.Fatalf("failed to listen on grpc port %d: %v", grpcPort, err)
 	}
-	grpcS := util.NewGrpcServer()
-	volume_server_pb.RegisterVolumeServerServer(grpcS, volumeServer)
+	grpcS := util.NewGrpcServer(security.LoadServerTLS(viper.Sub("grpc"), "volume"))
+	volume_server_pb.RegisterVolumeServerServer(grpcS, vs)
 	reflection.Register(grpcS)
-	go grpcS.Serve(grpcL)
+	go func() {
+		if err := grpcS.Serve(grpcL); err != nil {
+			glog.Fatalf("start gRPC service failed, %s", err)
+		}
+	}()
+	return grpcS
+}
 
-	if e := http.Serve(listener, volumeMux); e != nil {
-		glog.Fatalf("Volume server fail to serve: %v", e)
+func (v VolumeServerOptions) startPublicHttpService(handler http.Handler) httpdown.Server {
+	publicListeningAddress := *v.bindIp + ":" + strconv.Itoa(*v.publicPort)
+	glog.V(0).Infoln("Start Seaweed volume server", util.VERSION, "public at", publicListeningAddress)
+	publicListener, e := util.NewListener(publicListeningAddress, time.Duration(*v.idleConnectionTimeout)*time.Second)
+	if e != nil {
+		glog.Fatalf("Volume server listener error:%v", e)
 	}
 
+	pubHttp := httpdown.HTTP{StopTimeout: 5 * time.Minute, KillTimeout: 5 * time.Minute}
+	publicHttpDown := pubHttp.Serve(&http.Server{Handler: handler}, publicListener)
+	go func() {
+		if err := publicHttpDown.Wait(); err != nil {
+			glog.Errorf("public http down wait failed, %v", err)
+		}
+	}()
+
+	return publicHttpDown
+}
+
+func (v VolumeServerOptions) startClusterHttpService(handler http.Handler) httpdown.Server {
+	var (
+		certFile, keyFile string
+	)
+	if viper.GetString("https.volume.key") != "" {
+		certFile = viper.GetString("https.volume.cert")
+		keyFile = viper.GetString("https.volume.key")
+	}
+
+	listeningAddress := *v.bindIp + ":" + strconv.Itoa(*v.port)
+	glog.V(0).Infof("Start Seaweed volume server %s at %s", util.VERSION, listeningAddress)
+	listener, e := util.NewListener(listeningAddress, time.Duration(*v.idleConnectionTimeout)*time.Second)
+	if e != nil {
+		glog.Fatalf("Volume server listener error:%v", e)
+	}
+
+	httpDown := httpdown.HTTP{
+		KillTimeout: 5 * time.Minute,
+		StopTimeout: 5 * time.Minute,
+		CertFile:    certFile,
+		KeyFile:     keyFile}
+	clusterHttpServer := httpDown.Serve(&http.Server{Handler: handler}, listener)
+	go func() {
+		if e := clusterHttpServer.Wait(); e != nil {
+			glog.Fatalf("Volume server fail to serve: %v", e)
+		}
+	}()
+	return clusterHttpServer
 }

weed/command/webdav.go

@@ -0,0 +1,109 @@
+package command
+
+import (
+	"fmt"
+	"net/http"
+	"os/user"
+	"strconv"
+	"time"
+
+	"github.com/chrislusf/seaweedfs/weed/glog"
+	"github.com/chrislusf/seaweedfs/weed/security"
+	"github.com/chrislusf/seaweedfs/weed/server"
+	"github.com/chrislusf/seaweedfs/weed/util"
+	"github.com/spf13/viper"
+)
+
+var (
+	webDavStandaloneOptions WebDavOption
+)
+
+type WebDavOption struct {
+	filer          *string
+	port           *int
+	collection     *string
+	tlsPrivateKey  *string
+	tlsCertificate *string
+}
+
+func init() {
+	cmdWebDav.Run = runWebDav // break init cycle
+	webDavStandaloneOptions.filer = cmdWebDav.Flag.String("filer", "localhost:8888", "filer server address")
+	webDavStandaloneOptions.port = cmdWebDav.Flag.Int("port", 7333, "webdav server http listen port")
+	webDavStandaloneOptions.collection = cmdWebDav.Flag.String("collection", "", "collection to create the files")
+	webDavStandaloneOptions.tlsPrivateKey = cmdWebDav.Flag.String("key.file", "", "path to the TLS private key file")
+	webDavStandaloneOptions.tlsCertificate = cmdWebDav.Flag.String("cert.file", "", "path to the TLS certificate file")
+}
+
+var cmdWebDav = &Command{
+	UsageLine: "webdav -port=7333 -filer=<ip:port>",
+	Short:     "<unstable> start a webdav server that is backed by a filer",
+	Long: `start a webdav server that is backed by a filer.
+
+`,
+}
+
+func runWebDav(cmd *Command, args []string) bool {
+
+	util.LoadConfiguration("security", false)
+
+	glog.V(0).Infof("Starting Seaweed WebDav Server %s at https port %d", util.VERSION, *webDavStandaloneOptions.port)
+
+	return webDavStandaloneOptions.startWebDav()
+
+}
+
+func (wo *WebDavOption) startWebDav() bool {
+
+	filerGrpcAddress, err := parseFilerGrpcAddress(*wo.filer)
+	if err != nil {
+		glog.Fatal(err)
+		return false
+	}
+
+	// detect current user
+	uid, gid := uint32(0), uint32(0)
+	if u, err := user.Current(); err == nil {
+		if parsedId, pe := strconv.ParseUint(u.Uid, 10, 32); pe == nil {
+			uid = uint32(parsedId)
+		}
+		if parsedId, pe := strconv.ParseUint(u.Gid, 10, 32); pe == nil {
+			gid = uint32(parsedId)
+		}
+	}
+
+	ws, webdavServer_err := weed_server.NewWebDavServer(&weed_server.WebDavOption{
+		Filer:            *wo.filer,
+		FilerGrpcAddress: filerGrpcAddress,
+		GrpcDialOption:   security.LoadClientTLS(viper.Sub("grpc"), "client"),
+		Collection:       *wo.collection,
+		Uid:              uid,
+		Gid:              gid,
+	})
+	if webdavServer_err != nil {
+		glog.Fatalf("WebDav Server startup error: %v", webdavServer_err)
+	}
+
+	httpS := &http.Server{Handler: ws.Handler}
+
+	listenAddress := fmt.Sprintf(":%d", *wo.port)
+	webDavListener, err := util.NewListener(listenAddress, time.Duration(10)*time.Second)
+	if err != nil {
+		glog.Fatalf("WebDav Server listener on %s error: %v", listenAddress, err)
+	}
+
+	if *wo.tlsPrivateKey != "" {
+		glog.V(0).Infof("Start Seaweed WebDav Server %s at https port %d", util.VERSION, *wo.port)
+		if err = httpS.ServeTLS(webDavListener, *wo.tlsCertificate, *wo.tlsPrivateKey); err != nil {
+			glog.Fatalf("WebDav Server Fail to serve: %v", err)
+		}
+	} else {
+		glog.V(0).Infof("Start Seaweed WebDav Server %s at http port %d", util.VERSION, *wo.port)
+		if err = httpS.Serve(webDavListener); err != nil {
+			glog.Fatalf("WebDav Server Fail to serve: %v", err)
+		}
+	}
+
+	return true
+
+}

weed/filer2/abstract_sql/abstract_sql_store.go

@@ -1,6 +1,7 @@
 package abstract_sql
 
 import (
+	"context"
 	"database/sql"
 	"fmt"
 
@@ -9,16 +10,54 @@ import (
 )
 
 type AbstractSqlStore struct {
-	DB               *sql.DB
-	SqlInsert        string
-	SqlUpdate        string
-	SqlFind          string
-	SqlDelete        string
-	SqlListExclusive string
-	SqlListInclusive string
+	DB                      *sql.DB
+	SqlInsert               string
+	SqlUpdate               string
+	SqlFind                 string
+	SqlDelete               string
+	SqlDeleteFolderChildren string
+	SqlListExclusive        string
+	SqlListInclusive        string
 }
 
-func (store *AbstractSqlStore) InsertEntry(entry *filer2.Entry) (err error) {
+type TxOrDB interface {
+	ExecContext(ctx context.Context, query string, args ...interface{}) (sql.Result, error)
+	QueryRowContext(ctx context.Context, query string, args ...interface{}) *sql.Row
+	QueryContext(ctx context.Context, query string, args ...interface{}) (*sql.Rows, error)
+}
+
+func (store *AbstractSqlStore) BeginTransaction(ctx context.Context) (context.Context, error) {
+	tx, err := store.DB.BeginTx(ctx, &sql.TxOptions{
+		Isolation: sql.LevelReadCommitted,
+		ReadOnly:  false,
+	})
+	if err != nil {
+		return ctx, err
+	}
+
+	return context.WithValue(ctx, "tx", tx), nil
+}
+func (store *AbstractSqlStore) CommitTransaction(ctx context.Context) error {
+	if tx, ok := ctx.Value("tx").(*sql.Tx); ok {
+		return tx.Commit()
+	}
+	return nil
+}
+func (store *AbstractSqlStore) RollbackTransaction(ctx context.Context) error {
+	if tx, ok := ctx.Value("tx").(*sql.Tx); ok {
+		return tx.Rollback()
+	}
+	return nil
+}
+
+func (store *AbstractSqlStore) getTxOrDB(ctx context.Context) TxOrDB {
+	if tx, ok := ctx.Value("tx").(*sql.Tx); ok {
+		return tx
+	}
+	return store.DB
+}
+
+func (store *AbstractSqlStore) InsertEntry(ctx context.Context, entry *filer2.Entry) (err error) {
 
 	dir, name := entry.FullPath.DirAndName()
 	meta, err := entry.EncodeAttributesAndChunks()
@@ -26,7 +65,7 @@ func (store *AbstractSqlStore) InsertEntry(entry *filer2.Entry) (err error) {
 		return fmt.Errorf("encode %s: %s", entry.FullPath, err)
 	}
 
-	res, err := store.DB.Exec(store.SqlInsert, hashToLong(dir), name, dir, meta)
+	res, err := store.getTxOrDB(ctx).ExecContext(ctx, store.SqlInsert, hashToLong(dir), name, dir, meta)
 	if err != nil {
 		return fmt.Errorf("insert %s: %s", entry.FullPath, err)
 	}
@@ -38,7 +77,7 @@ func (store *AbstractSqlStore) InsertEntry(entry *filer2.Entry) (err error) {
 	return nil
 }
 
-func (store *AbstractSqlStore) UpdateEntry(entry *filer2.Entry) (err error) {
+func (store *AbstractSqlStore) UpdateEntry(ctx context.Context, entry *filer2.Entry) (err error) {
 
 	dir, name := entry.FullPath.DirAndName()
 	meta, err := entry.EncodeAttributesAndChunks()
@@ -46,7 +85,7 @@ func (store *AbstractSqlStore) UpdateEntry(entry *filer2.Entry) (err error) {
 		return fmt.Errorf("encode %s: %s", entry.FullPath, err)
 	}
 
-	res, err := store.DB.Exec(store.SqlUpdate, meta, hashToLong(dir), name, dir)
+	res, err := store.getTxOrDB(ctx).ExecContext(ctx, store.SqlUpdate, meta, hashToLong(dir), name, dir)
 	if err != nil {
 		return fmt.Errorf("update %s: %s", entry.FullPath, err)
 	}
@@ -58,10 +97,10 @@ func (store *AbstractSqlStore) UpdateEntry(entry *filer2.Entry) (err error) {
 	return nil
 }
 
-func (store *AbstractSqlStore) FindEntry(fullpath filer2.FullPath) (*filer2.Entry, error) {
+func (store *AbstractSqlStore) FindEntry(ctx context.Context, fullpath filer2.FullPath) (*filer2.Entry, error) {
 
 	dir, name := fullpath.DirAndName()
-	row := store.DB.QueryRow(store.SqlFind, hashToLong(dir), name, dir)
+	row := store.getTxOrDB(ctx).QueryRowContext(ctx, store.SqlFind, hashToLong(dir), name, dir)
 	var data []byte
 	if err := row.Scan(&data); err != nil {
 		return nil, filer2.ErrNotFound
@@ -77,11 +116,11 @@ func (store *AbstractSqlStore) FindEntry(fullpath filer2.FullPath) (*filer2.Entr
 	return entry, nil
 }
 
-func (store *AbstractSqlStore) DeleteEntry(fullpath filer2.FullPath) error {
+func (store *AbstractSqlStore) DeleteEntry(ctx context.Context, fullpath filer2.FullPath) error {
 
 	dir, name := fullpath.DirAndName()
 
-	res, err := store.DB.Exec(store.SqlDelete, hashToLong(dir), name, dir)
+	res, err := store.getTxOrDB(ctx).ExecContext(ctx, store.SqlDelete, hashToLong(dir), name, dir)
 	if err != nil {
 		return fmt.Errorf("delete %s: %s", fullpath, err)
 	}
@@ -94,14 +133,29 @@ func (store *AbstractSqlStore) DeleteEntry(fullpath filer2.FullPath) error {
 	return nil
 }
 
-func (store *AbstractSqlStore) ListDirectoryEntries(fullpath filer2.FullPath, startFileName string, inclusive bool, limit int) (entries []*filer2.Entry, err error) {
+func (store *AbstractSqlStore) DeleteFolderChildren(ctx context.Context, fullpath filer2.FullPath) error {
+
+	res, err := store.getTxOrDB(ctx).ExecContext(ctx, store.SqlDeleteFolderChildren, hashToLong(string(fullpath)), fullpath)
+	if err != nil {
+		return fmt.Errorf("deleteFolderChildren %s: %s", fullpath, err)
+	}
+
+	_, err = res.RowsAffected()
+	if err != nil {
+		return fmt.Errorf("deleteFolderChildren %s but no rows affected: %s", fullpath, err)
+	}
+
+	return nil
+}
+
+func (store *AbstractSqlStore) ListDirectoryEntries(ctx context.Context, fullpath filer2.FullPath, startFileName string, inclusive bool, limit int) (entries []*filer2.Entry, err error) {
 
 	sqlText := store.SqlListExclusive
 	if inclusive {
 		sqlText = store.SqlListInclusive
 	}
 
-	rows, err := store.DB.Query(sqlText, hashToLong(string(fullpath)), startFileName, string(fullpath), limit)
+	rows, err := store.getTxOrDB(ctx).QueryContext(ctx, sqlText, hashToLong(string(fullpath)), startFileName, string(fullpath), limit)
 	if err != nil {
 		return nil, fmt.Errorf("list %s : %v", fullpath, err)
 	}

weed/filer2/cassandra/cassandra_store.go

@@ -1,6 +1,7 @@
 package cassandra
 
 import (
+	"context"
 	"fmt"
 	"github.com/chrislusf/seaweedfs/weed/filer2"
 	"github.com/chrislusf/seaweedfs/weed/glog"
@@ -39,7 +40,17 @@ func (store *CassandraStore) initialize(keyspace string, hosts []string) (err er
 	return
 }
 
-func (store *CassandraStore) InsertEntry(entry *filer2.Entry) (err error) {
+func (store *CassandraStore) BeginTransaction(ctx context.Context) (context.Context, error) {
+	return ctx, nil
+}
+func (store *CassandraStore) CommitTransaction(ctx context.Context) error {
+	return nil
+}
+func (store *CassandraStore) RollbackTransaction(ctx context.Context) error {
+	return nil
+}
+
+func (store *CassandraStore) InsertEntry(ctx context.Context, entry *filer2.Entry) (err error) {
 
 	dir, name := entry.FullPath.DirAndName()
 	meta, err := entry.EncodeAttributesAndChunks()
@@ -48,20 +59,20 @@ func (store *CassandraStore) InsertEntry(entry *filer2.Entry) (err error) {
 	}
 
 	if err := store.session.Query(
-		"INSERT INTO filemeta (directory,name,meta) VALUES(?,?,?)",
-		dir, name, meta).Exec(); err != nil {
+		"INSERT INTO filemeta (directory,name,meta) VALUES(?,?,?) USING TTL ? ",
+		dir, name, meta, entry.TtlSec).Exec(); err != nil {
 		return fmt.Errorf("insert %s: %s", entry.FullPath, err)
 	}
 
 	return nil
 }
 
-func (store *CassandraStore) UpdateEntry(entry *filer2.Entry) (err error) {
+func (store *CassandraStore) UpdateEntry(ctx context.Context, entry *filer2.Entry) (err error) {
 
-	return store.InsertEntry(entry)
+	return store.InsertEntry(ctx, entry)
 }
 
-func (store *CassandraStore) FindEntry(fullpath filer2.FullPath) (entry *filer2.Entry, err error) {
+func (store *CassandraStore) FindEntry(ctx context.Context, fullpath filer2.FullPath) (entry *filer2.Entry, err error) {
 
 	dir, name := fullpath.DirAndName()
 	var data []byte
@@ -74,7 +85,7 @@ func (store *CassandraStore) FindEntry(fullpath filer2.FullPath) (entry *filer2.
 	}
 
 	if len(data) == 0 {
-		return nil, fmt.Errorf("not found: %s", fullpath)
+		return nil, filer2.ErrNotFound
 	}
 
 	entry = &filer2.Entry{
@@ -88,7 +99,7 @@ func (store *CassandraStore) FindEntry(fullpath filer2.FullPath) (entry *filer2.
 	return entry, nil
 }
 
-func (store *CassandraStore) DeleteEntry(fullpath filer2.FullPath) error {
+func (store *CassandraStore) DeleteEntry(ctx context.Context, fullpath filer2.FullPath) error {
 
 	dir, name := fullpath.DirAndName()
 
@@ -101,7 +112,18 @@ func (store *CassandraStore) DeleteEntry(fullpath filer2.FullPath) error {
 	return nil
 }
 
-func (store *CassandraStore) ListDirectoryEntries(fullpath filer2.FullPath, startFileName string, inclusive bool,
+func (store *CassandraStore) DeleteFolderChildren(ctx context.Context, fullpath filer2.FullPath) error {
+
+	if err := store.session.Query(
+		"DELETE FROM filemeta WHERE directory=?",
+		fullpath).Exec(); err != nil {
+		return fmt.Errorf("delete %s : %v", fullpath, err)
+	}
+
+	return nil
+}
+
+func (store *CassandraStore) ListDirectoryEntries(ctx context.Context, fullpath filer2.FullPath, startFileName string, inclusive bool,
 	limit int) (entries []*filer2.Entry, err error) {
 
 	cqlStr := "SELECT NAME, meta FROM filemeta WHERE directory=? AND name>? ORDER BY NAME ASC LIMIT ?"

weed/filer2/configuration.go

@@ -13,6 +13,8 @@ var (
 
 func (f *Filer) LoadConfiguration(config *viper.Viper) {
 
+	validateOneEnabledStore(config)
+
 	for _, store := range Stores {
 		if config.GetBool(store.GetName() + ".enabled") {
 			viperSub := config.Sub(store.GetName())
@@ -34,3 +36,16 @@ func (f *Filer) LoadConfiguration(config *viper.Viper) {
 
 	os.Exit(-1)
 }
+
+func validateOneEnabledStore(config *viper.Viper) {
+	enabledStore := ""
+	for _, store := range Stores {
+		if config.GetBool(store.GetName() + ".enabled") {
+			if enabledStore == "" {
+				enabledStore = store.GetName()
+			} else {
+				glog.Fatalf("Filer store is enabled for both %s and %s", enabledStore, store.GetName())
+			}
+		}
+	}
+}

weed/filer2/entry.go

@@ -8,15 +8,18 @@ import (
 )
 
 type Attr struct {
-	Mtime       time.Time   // time of last modification
-	Crtime      time.Time   // time of creation (OS X only)
-	Mode        os.FileMode // file mode
-	Uid         uint32      // owner uid
-	Gid         uint32      // group gid
-	Mime        string      // mime type
-	Replication string      // replication
-	Collection  string      // collection name
-	TtlSec      int32       // ttl in seconds
+	Mtime         time.Time   // time of last modification
+	Crtime        time.Time   // time of creation (OS X only)
+	Mode          os.FileMode // file mode
+	Uid           uint32      // owner uid
+	Gid           uint32      // group gid
+	Mime          string      // mime type
+	Replication   string      // replication
+	Collection    string      // collection name
+	TtlSec        int32       // ttl in seconds
+	UserName      string
+	GroupNames    []string
+	SymlinkTarget string
 }
 
 func (attr Attr) IsDirectory() bool {
@@ -27,6 +30,7 @@ type Entry struct {
 	FullPath
 
 	Attr
+	Extended map[string][]byte
 
 	// the following is for files
 	Chunks []*filer_pb.FileChunk `json:"chunks,omitempty"`
@@ -49,9 +53,21 @@ func (entry *Entry) ToProtoEntry() *filer_pb.Entry {
 		return nil
 	}
 	return &filer_pb.Entry{
-		Name:        string(entry.FullPath),
+		Name:        entry.FullPath.Name(),
 		IsDirectory: entry.IsDirectory(),
 		Attributes:  EntryAttributeToPb(entry),
 		Chunks:      entry.Chunks,
+		Extended:    entry.Extended,
+	}
+}
+
+func (entry *Entry) ToProtoFullEntry() *filer_pb.FullEntry {
+	if entry == nil {
+		return nil
+	}
+	dir, _ := entry.FullPath.DirAndName()
+	return &filer_pb.FullEntry{
+		Dir:   dir,
+		Entry: entry.ToProtoEntry(),
 	}
 }

weed/filer2/entry_codec.go

@@ -1,18 +1,21 @@
 package filer2
 
 import (
+	"bytes"
+	"fmt"
 	"os"
 	"time"
 
-	"fmt"
+	"github.com/golang/protobuf/proto"
+
 	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
-	"github.com/gogo/protobuf/proto"
 )
 
 func (entry *Entry) EncodeAttributesAndChunks() ([]byte, error) {
 	message := &filer_pb.Entry{
 		Attributes: EntryAttributeToPb(entry),
 		Chunks:     entry.Chunks,
+		Extended:   entry.Extended,
 	}
 	return proto.Marshal(message)
 }
@@ -27,6 +30,8 @@ func (entry *Entry) DecodeAttributesAndChunks(blob []byte) error {
 
 	entry.Attr = PbToEntryAttribute(message.Attributes)
 
+	entry.Extended = message.Extended
+
 	entry.Chunks = message.Chunks
 
 	return nil
@@ -35,15 +40,18 @@ func (entry *Entry) DecodeAttributesAndChunks(blob []byte) error {
 func EntryAttributeToPb(entry *Entry) *filer_pb.FuseAttributes {
 
 	return &filer_pb.FuseAttributes{
-		Crtime:      entry.Attr.Crtime.Unix(),
-		Mtime:       entry.Attr.Mtime.Unix(),
-		FileMode:    uint32(entry.Attr.Mode),
-		Uid:         entry.Uid,
-		Gid:         entry.Gid,
-		Mime:        entry.Mime,
-		Collection:  entry.Attr.Collection,
-		Replication: entry.Attr.Replication,
-		TtlSec:      entry.Attr.TtlSec,
+		Crtime:        entry.Attr.Crtime.Unix(),
+		Mtime:         entry.Attr.Mtime.Unix(),
+		FileMode:      uint32(entry.Attr.Mode),
+		Uid:           entry.Uid,
+		Gid:           entry.Gid,
+		Mime:          entry.Mime,
+		Collection:    entry.Attr.Collection,
+		Replication:   entry.Attr.Replication,
+		TtlSec:        entry.Attr.TtlSec,
+		UserName:      entry.Attr.UserName,
+		GroupName:     entry.Attr.GroupNames,
+		SymlinkTarget: entry.Attr.SymlinkTarget,
 	}
 }
 
@@ -60,6 +68,9 @@ func PbToEntryAttribute(attr *filer_pb.FuseAttributes) Attr {
 	t.Collection = attr.Collection
 	t.Replication = attr.Replication
 	t.TtlSec = attr.TtlSec
+	t.UserName = attr.UserName
+	t.GroupNames = attr.GroupName
+	t.SymlinkTarget = attr.SymlinkTarget
 
 	return t
 }
@@ -78,6 +89,10 @@ func EqualEntry(a, b *Entry) bool {
 		return false
 	}
 
+	if !eq(a.Extended, b.Extended) {
+		return false
+	}
+
 	for i := 0; i < len(a.Chunks); i++ {
 		if !proto.Equal(a.Chunks[i], b.Chunks[i]) {
 			return false
@@ -85,3 +100,17 @@ func EqualEntry(a, b *Entry) bool {
 	}
 	return true
 }
+
+func eq(a, b map[string][]byte) bool {
+	if len(a) != len(b) {
+		return false
+	}
+
+	for k, v := range a {
+		if w, ok := b[k]; !ok || !bytes.Equal(v, w) {
+			return false
+		}
+	}
+
+	return true
+}

weed/filer2/etcd/etcd_store.go

@@ -0,0 +1,196 @@
+package etcd
+
+import (
+	"context"
+	"fmt"
+	"strings"
+	"time"
+
+	"github.com/chrislusf/seaweedfs/weed/filer2"
+	"github.com/chrislusf/seaweedfs/weed/glog"
+	weed_util "github.com/chrislusf/seaweedfs/weed/util"
+	"go.etcd.io/etcd/clientv3"
+)
+
+const (
+	DIR_FILE_SEPARATOR = byte(0x00)
+)
+
+func init() {
+	filer2.Stores = append(filer2.Stores, &EtcdStore{})
+}
+
+type EtcdStore struct {
+	client *clientv3.Client
+}
+
+func (store *EtcdStore) GetName() string {
+	return "etcd"
+}
+
+func (store *EtcdStore) Initialize(configuration weed_util.Configuration) (err error) {
+	servers := configuration.GetString("servers")
+	if servers == "" {
+		servers = "localhost:2379"
+	}
+
+	timeout := configuration.GetString("timeout")
+	if timeout == "" {
+		timeout = "3s"
+	}
+
+	return store.initialize(servers, timeout)
+}
+
+func (store *EtcdStore) initialize(servers string, timeout string) (err error) {
+	glog.Infof("filer store etcd: %s", servers)
+
+	to, err := time.ParseDuration(timeout)
+	if err != nil {
+		return fmt.Errorf("parse timeout %s: %s", timeout, err)
+	}
+
+	store.client, err = clientv3.New(clientv3.Config{
+		Endpoints:   strings.Split(servers, ","),
+		DialTimeout: to,
+	})
+	if err != nil {
+		return fmt.Errorf("connect to etcd %s: %s", servers, err)
+	}
+
+	return
+}
+
+func (store *EtcdStore) BeginTransaction(ctx context.Context) (context.Context, error) {
+	return ctx, nil
+}
+func (store *EtcdStore) CommitTransaction(ctx context.Context) error {
+	return nil
+}
+func (store *EtcdStore) RollbackTransaction(ctx context.Context) error {
+	return nil
+}
+
+func (store *EtcdStore) InsertEntry(ctx context.Context, entry *filer2.Entry) (err error) {
+	key := genKey(entry.DirAndName())
+
+	value, err := entry.EncodeAttributesAndChunks()
+	if err != nil {
+		return fmt.Errorf("encoding %s %+v: %v", entry.FullPath, entry.Attr, err)
+	}
+
+	if _, err := store.client.Put(ctx, string(key), string(value)); err != nil {
+		return fmt.Errorf("persisting %s : %v", entry.FullPath, err)
+	}
+
+	return nil
+}
+
+func (store *EtcdStore) UpdateEntry(ctx context.Context, entry *filer2.Entry) (err error) {
+	return store.InsertEntry(ctx, entry)
+}
+
+func (store *EtcdStore) FindEntry(ctx context.Context, fullpath filer2.FullPath) (entry *filer2.Entry, err error) {
+	key := genKey(fullpath.DirAndName())
+
+	resp, err := store.client.Get(ctx, string(key))
+	if err != nil {
+		return nil, fmt.Errorf("get %s : %v", entry.FullPath, err)
+	}
+
+	if len(resp.Kvs) == 0 {
+		return nil, filer2.ErrNotFound
+	}
+
+	entry = &filer2.Entry{
+		FullPath: fullpath,
+	}
+	err = entry.DecodeAttributesAndChunks(resp.Kvs[0].Value)
+	if err != nil {
+		return entry, fmt.Errorf("decode %s : %v", entry.FullPath, err)
+	}
+
+	return entry, nil
+}
+
+func (store *EtcdStore) DeleteEntry(ctx context.Context, fullpath filer2.FullPath) (err error) {
+	key := genKey(fullpath.DirAndName())
+
+	if _, err := store.client.Delete(ctx, string(key)); err != nil {
+		return fmt.Errorf("delete %s : %v", fullpath, err)
+	}
+
+	return nil
+}
+
+func (store *EtcdStore) DeleteFolderChildren(ctx context.Context, fullpath filer2.FullPath) (err error) {
+	directoryPrefix := genDirectoryKeyPrefix(fullpath, "")
+
+	if _, err := store.client.Delete(ctx, string(directoryPrefix), clientv3.WithPrefix()); err != nil {
+		return fmt.Errorf("deleteFolderChildren %s : %v", fullpath, err)
+	}
+
+	return nil
+}
+
+func (store *EtcdStore) ListDirectoryEntries(
+	ctx context.Context, fullpath filer2.FullPath, startFileName string, inclusive bool, limit int,
+) (entries []*filer2.Entry, err error) {
+	directoryPrefix := genDirectoryKeyPrefix(fullpath, "")
+
+	resp, err := store.client.Get(ctx, string(directoryPrefix),
+		clientv3.WithPrefix(), clientv3.WithSort(clientv3.SortByKey, clientv3.SortDescend))
+	if err != nil {
+		return nil, fmt.Errorf("list %s : %v", fullpath, err)
+	}
+
+	for _, kv := range resp.Kvs {
+		fileName := getNameFromKey(kv.Key)
+		if fileName == "" {
+			continue
+		}
+		if fileName == startFileName && !inclusive {
+			continue
+		}
+		limit--
+		if limit < 0 {
+			break
+		}
+		entry := &filer2.Entry{
+			FullPath: filer2.NewFullPath(string(fullpath), fileName),
+		}
+		if decodeErr := entry.DecodeAttributesAndChunks(kv.Value); decodeErr != nil {
+			err = decodeErr
+			glog.V(0).Infof("list %s : %v", entry.FullPath, err)
+			break
+		}
+		entries = append(entries, entry)
+	}
+
+	return entries, err
+}
+
+func genKey(dirPath, fileName string) (key []byte) {
+	key = []byte(dirPath)
+	key = append(key, DIR_FILE_SEPARATOR)
+	key = append(key, []byte(fileName)...)
+	return key
+}
+
+func genDirectoryKeyPrefix(fullpath filer2.FullPath, startFileName string) (keyPrefix []byte) {
+	keyPrefix = []byte(string(fullpath))
+	keyPrefix = append(keyPrefix, DIR_FILE_SEPARATOR)
+	if len(startFileName) > 0 {
+		keyPrefix = append(keyPrefix, []byte(startFileName)...)
+	}
+	return keyPrefix
+}
+
+func getNameFromKey(key []byte) string {
+	sepIndex := len(key) - 1
+	for sepIndex >= 0 && key[sepIndex] != DIR_FILE_SEPARATOR {
+		sepIndex--
+	}
+
+	return string(key[sepIndex+1:])
+}