s3 metrics adjust the label

fix https://github.com/chrislusf/seaweedfs/issues/1501

.travis.yml

@@ -1,9 +1,8 @@
 sudo: false
 language: go
 go:
-  - 1.12.x
-  - 1.13.x
   - 1.14.x
+  - 1.15.x
 
 before_install:
   - export PATH=/home/travis/gopath/bin:$PATH
@@ -45,4 +44,4 @@ deploy:
   on:
     tags: true
     repo: chrislusf/seaweedfs
-    go: 1.14.x
+    go: 1.15.x

README.md

@@ -90,7 +90,7 @@ There is only 40 bytes of disk storage overhead for each file's metadata. It is
 
 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)
 
-On top of the object store, optional [Filer] can support directories and POSIX attributes. Filer is a separate linearly-scalable stateless server with customizable metadata stores, e.g., MySql, Postgres, Mongodb, Redis, Etcd, Cassandra, LevelDB, MemSql, TiDB, TiKV, CockroachDB, etc.
+On top of the object store, optional [Filer] can support directories and POSIX attributes. Filer is a separate linearly-scalable stateless server with customizable metadata stores, e.g., MySql, Postgres, Mongodb, Redis, Cassandra, Elastic Search, LevelDB, MemSql, TiDB, Etcd, CockroachDB, etc.
 
 [Back to TOC](#table-of-contents)
 
@@ -115,9 +115,10 @@ On top of the object store, optional [Filer] can support directories and POSIX a
 * [Filer server][Filer] provides "normal" directories and files via http.
 * [Super Large Files][SuperLargeFiles] stores large or super large files in tens of TB.
 * [Mount filer][Mount] reads and writes files directly as a local directory via FUSE.
+* [Active-Active Replication][ActiveActiveAsyncReplication] enables asynchronous one-way or two-way cross cluster continuous replication.
 * [Amazon S3 compatible API][AmazonS3API] accesses files with S3 tooling.
-* [Hadoop Compatible File System][Hadoop] accesses 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.
+* [Hadoop Compatible File System][Hadoop] accesses files from Hadoop/Spark/Flink/etc or even runs HBase.
+* [Async Replication To Cloud][BackupToCloud] has extremely fast local access and backups to Amazon S3, Google Cloud Storage, Azure, BackBlaze.
 * [WebDAV] accesses as a mapped drive on Mac and Windows, or from mobile devices.
 * [AES256-GCM Encrypted Storage][FilerDataEncryption] safely stores the encrypted data.
 * [File TTL][FilerTTL] automatically purges file metadata and actual file data.
@@ -127,7 +128,7 @@ On top of the object store, optional [Filer] can support directories and POSIX a
 [SuperLargeFiles]: https://github.com/chrislusf/seaweedfs/wiki/Data-Structure-for-Large-Files
 [Mount]: https://github.com/chrislusf/seaweedfs/wiki/FUSE-Mount
 [AmazonS3API]: https://github.com/chrislusf/seaweedfs/wiki/Amazon-S3-API
-[BackupToCloud]: https://github.com/chrislusf/seaweedfs/wiki/Backup-to-Cloud
+[BackupToCloud]: https://github.com/chrislusf/seaweedfs/wiki/Async-Replication-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
@@ -136,6 +137,7 @@ On top of the object store, optional [Filer] can support directories and POSIX a
 [FilerTTL]: https://github.com/chrislusf/seaweedfs/wiki/Filer-Stores
 [VolumeServerTTL]: https://github.com/chrislusf/seaweedfs/wiki/Store-file-with-a-Time-To-Live
 [SeaweedFsCsiDriver]: https://github.com/seaweedfs/seaweedfs-csi-driver
+[ActiveActiveAsyncReplication]: https://github.com/chrislusf/seaweedfs/wiki/Filer-Active-Active-cross-cluster-continuous-synchronization
 
 [Back to TOC](#table-of-contents)
 
@@ -365,10 +367,10 @@ 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, Mongodb, Redis, Etcd, MySql, Postgres, MemSql, TiDB, CockroachDB, etc, and is easy to customized.
+* SeaweedFS Filer metadata store can be any well-known and proven data stores, e.g., Cassandra, Mongodb, Redis, Elastic Search, MySql, Postgres, MemSql, TiDB, CockroachDB, Etcd 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 |
+| System         | File Meta                       | File Content Read| POSIX  | REST API | Optimized for large number of small files |
 | -------------  | ------------------------------- | ---------------- | ------ | -------- | ------------------------- |
 | SeaweedFS      | lookup volume id, cacheable     | O(1) disk seek   |        | Yes      | Yes                       |
 | SeaweedFS Filer| Linearly Scalable, Customizable | O(1) disk seek   | FUSE   | Yes      | Yes                       |
@@ -406,7 +408,7 @@ Ceph uses CRUSH hashing to automatically manage the data placement. SeaweedFS pl
 
 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, Mongodb, Redis, Etcd, Cassandra, MemSql, TiDB, CockroachCB, to manage file directories. These stores are proven, scalable, and easier to manage.
+SeaweedFS Filer uses off-the-shelf stores, such as MySql, Postgres, Mongodb, Redis, Elastic Search, Cassandra, MemSql, TiDB, CockroachCB, Etcd, to manage file directories. These stores are proven, scalable, and easier to manage.
 
 | SeaweedFS         | comparable to Ceph | advantage |
 | -------------  | ------------- | ---------------- |

docker/Dockerfile.go_build_large

@@ -0,0 +1,35 @@
+FROM frolvlad/alpine-glibc as builder
+RUN apk add git go g++
+RUN mkdir -p /go/src/github.com/chrislusf/
+RUN git clone https://github.com/chrislusf/seaweedfs /go/src/github.com/chrislusf/seaweedfs
+RUN cd /go/src/github.com/chrislusf/seaweedfs/weed && go install -tags 5BytesOffset
+
+FROM alpine AS final
+LABEL author="Chris Lu"
+COPY --from=builder /root/go/bin/weed /usr/bin/
+RUN mkdir -p /etc/seaweedfs
+COPY --from=builder /go/src/github.com/chrislusf/seaweedfs/docker/filer.toml /etc/seaweedfs/filer.toml
+COPY --from=builder /go/src/github.com/chrislusf/seaweedfs/docker/entrypoint.sh /entrypoint.sh
+
+# 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 chmod +x /entrypoint.sh
+
+ENTRYPOINT ["/entrypoint.sh"]

go.mod

@@ -8,7 +8,7 @@ require (
 	github.com/Azure/azure-storage-blob-go v0.8.0
 	github.com/OneOfOne/xxhash v1.2.2
 	github.com/Shopify/sarama v1.23.1
-	github.com/aws/aws-sdk-go v1.23.13
+	github.com/aws/aws-sdk-go v1.33.5
 	github.com/buraksezer/consistent v0.0.0-20191006190839-693edf70fd72
 	github.com/cespare/xxhash v1.1.0
 	github.com/chrislusf/raft v1.0.1
@@ -24,10 +24,9 @@ require (
 	github.com/facebookgo/subset v0.0.0-20200203212716-c811ad88dec4 // indirect
 	github.com/frankban/quicktest v1.7.2 // indirect
 	github.com/go-redis/redis v6.15.7+incompatible
-	github.com/go-sql-driver/mysql v1.4.1
+	github.com/go-sql-driver/mysql v1.5.0
 	github.com/gocql/gocql v0.0.0-20190829130954-e163eff7a8c6
 	github.com/gogo/protobuf v1.2.2-0.20190730201129-28a6bbf47e48 // indirect
-	github.com/golang/groupcache v0.0.0-20190702054246-869f871628b6 // indirect
 	github.com/golang/protobuf v1.4.2
 	github.com/google/btree v1.0.0
 	github.com/google/uuid v1.1.1
@@ -36,6 +35,7 @@ require (
 	github.com/grpc-ecosystem/grpc-gateway v1.11.0 // indirect
 	github.com/hashicorp/golang-lru v0.5.3 // indirect
 	github.com/jcmturner/gofork v1.0.0 // indirect
+	github.com/json-iterator/go v1.1.10
 	github.com/karlseguin/ccache v2.0.3+incompatible
 	github.com/karlseguin/expect v1.0.1 // indirect
 	github.com/klauspost/compress v1.10.9
@@ -48,6 +48,7 @@ require (
 	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/olivere/elastic/v7 v7.0.19
 	github.com/onsi/ginkgo v1.10.1 // indirect
 	github.com/onsi/gomega v1.7.0 // indirect
 	github.com/peterh/liner v1.1.0
@@ -63,7 +64,7 @@ require (
 	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
+	github.com/stretchr/testify v1.5.1
 	github.com/syndtr/goleveldb v1.0.0
 	github.com/tidwall/gjson v1.3.2
 	github.com/tidwall/match v1.0.1
@@ -76,7 +77,7 @@ require (
 	gocloud.dev/pubsub/natspubsub v0.16.0
 	gocloud.dev/pubsub/rabbitpubsub v0.16.0
 	golang.org/x/image v0.0.0-20200119044424-58c23975cae1 // indirect
-	golang.org/x/net v0.0.0-20190909003024-a7b16738d86b
+	golang.org/x/net v0.0.0-20200202094626-16171245cfb2
 	golang.org/x/sys v0.0.0-20200202164722-d101bd2416d5
 	golang.org/x/tools v0.0.0-20191029190741-b9c20aec41a5
 	google.golang.org/api v0.9.0

go.sum

@@ -47,6 +47,8 @@ github.com/aws/aws-sdk-go v1.19.18/go.mod h1:KmX6BPdI08NWTb3/sm4ZGu5ShLoqVDhKgpi
 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/aws/aws-sdk-go v1.33.5 h1:p2fr1ryvNTU6avUWLI+/H7FGv0TBIjzVM5WDgXBBv4U=
+github.com/aws/aws-sdk-go v1.33.5/go.mod h1:5zCpMtNQVjRREroY7sYe8lOMRSxkhG6MZveU8YkpAk0=
 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=
@@ -138,6 +140,8 @@ github.com/go-redis/redis v6.15.7+incompatible h1:3skhDh95XQMpnqeqNftPkQD9jL9e5e
 github.com/go-redis/redis v6.15.7+incompatible/go.mod h1:NAIEuMOZ/fxfXJIrKDQDz8wamY7mA7PouImQ2Jvg6kA=
 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-sql-driver/mysql v1.5.0 h1:ozyZYNQW3x3HtqT1jira07DN2PArx2v7/mN66gGcHOs=
+github.com/go-sql-driver/mysql v1.5.0/go.mod h1:DCzpHaOWr8IXmIStZouvnhqoel9Qv2LBy8hT2VhHyBg=
 github.com/go-stack/stack v1.8.0 h1:5SgMzNM5HxrEjV0ww2lTmX6E2Izsfxas4+YHWRs3Lsk=
 github.com/go-stack/stack v1.8.0/go.mod h1:v0f6uXyyMGvRgIKkXu+yp6POWl0qKG85gN/melR3HDY=
 github.com/gobuffalo/attrs v0.0.0-20190224210810-a9411de4debd/go.mod h1:4duuawTqi2wkkpB4ePgWMaai6/Kc6WEz83bhFwpHzj0=
@@ -210,6 +214,8 @@ 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-cmp v0.4.0 h1:xsAVV57WRhGj6kEIi8ReJzQlHHqcBYCElAvkovg3B/4=
 github.com/google/go-cmp v0.4.0/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
+github.com/google/go-cmp v0.5.0 h1:/QaMHBdZ26BB3SSst0Iwl10Epc+xhTquomWX0oZEB6w=
+github.com/google/go-cmp v0.5.0/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
 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=
@@ -273,12 +279,16 @@ github.com/jcmturner/gofork v1.0.0/go.mod h1:MK8+TM0La+2rjBD4jE12Kj1pCCxK7d2LK/U
 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/jmespath/go-jmespath v0.3.0 h1:OS12ieG61fsCg5+qLJ+SsW9NicxNkg3b25OyT2yCeUc=
+github.com/jmespath/go-jmespath v0.3.0/go.mod h1:9QtRXoHjLGCJ5IBSaohpXITPlowMeeYCZ7fLUTSywik=
 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/json-iterator/go v1.1.10 h1:Kz6Cvnvv2wGdaG/V8yMvfkmNiXq9Ya2KUv4rouJJr68=
+github.com/json-iterator/go v1.1.10/go.mod h1:KdQUCv79m/52Kvf8AW2vK1V8akMuk1QjK/uOdHXbAo4=
 github.com/jstemmer/go-junit-report v0.0.0-20190106144839-af01ea7f8024/go.mod h1:6v2b51hI/fHJwM22ozAgKL4VKDeJcHhJFhtBdhmNjmU=
 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=
@@ -319,6 +329,8 @@ github.com/magiconair/properties v1.8.0 h1:LLgXmsheXeRoUOBOjtwPQCWIYqM/LU1ayDtDe
 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/mailru/easyjson v0.7.1 h1:mdxE1MF9o53iCb2Ghj1VfWvh7ZOwHpnVG/xwXrV90U8=
+github.com/mailru/easyjson v0.7.1/go.mod h1:KAzv3t3aY1NaHWoQz1+4F1ccyAH66Jk7yos7ldAVICs=
 github.com/markbates/oncer v0.0.0-20181203154359-bf2de49a0be2/go.mod h1:Ld9puTsIW75CHf65OeIOkyKbteujpZVXDpWK6YGZbxE=
 github.com/markbates/safe v1.0.1/go.mod h1:nAqgmRi7cY2nqMc92/bSEeQA+R4OheNU2T1kNSCBdG0=
 github.com/mattn/go-colorable v0.0.9/go.mod h1:9vuHe8Xs5qXnSaW/c/ABM9alt+Vo+STaOChaDxuIBZU=
@@ -364,6 +376,8 @@ 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/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/olivere/elastic/v7 v7.0.19 h1:w4F6JpqOISadhYf/n0NR1cNj73xHqh4pzPwD1Gkidts=
+github.com/olivere/elastic/v7 v7.0.19/go.mod h1:4Jqt5xvjqpjCqgnTcHwl3j8TLs8mvoOK8NYgo/qEOu4=
 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=
@@ -373,6 +387,7 @@ 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/opentracing-go v1.2.0/go.mod h1:GxEUsuufX4nBwe+T+Wl9TAgYrxe9dPLANfrWvHYVTgc=
 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 v1.4.0 h1:u3Z1r+oOXJIkxqw34zVhyPgjBsm6X2wn21NWs/HfSeg=
@@ -385,6 +400,8 @@ github.com/pierrec/lz4 v2.2.7+incompatible/go.mod h1:pdkljMzZIN41W+lC3N2tnIh5sFi
 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/errors v0.9.1 h1:FEBLx1zS214owpjy7qsBeixbURkuhQAwrK5UwLGTwt4=
+github.com/pkg/errors v0.9.1/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=
@@ -433,6 +450,9 @@ github.com/sirupsen/logrus v1.4.0/go.mod h1:LxeOpSwHxABJmUn/MG1IvRgCAasNZTLOkJPx
 github.com/sirupsen/logrus v1.4.1/go.mod h1:ni0Sbl8bgC9z8RoU9G6nDWqqs/fq4eDPysMBDgk/93Q=
 github.com/sirupsen/logrus v1.4.2 h1:SPIRibHv4MatM3XXNO2BJeFLZwZ2LvZgfQ5+UNI2im4=
 github.com/sirupsen/logrus v1.4.2/go.mod h1:tLMulIdttU9McNUspp0xgXVQah82FyeX6MwdIuYE2rE=
+github.com/smartystreets/assertions v1.1.1/go.mod h1:tcbTF8ujkAEcZ8TElKY+i30BzYlVhC/LOxJk7iOWnoo=
+github.com/smartystreets/go-aws-auth v0.0.0-20180515143844-0c1422d1fdb9/go.mod h1:SnhjPscd9TpLiy1LpzGSKh3bXCfxxXuqd9xmQJy3slM=
+github.com/smartystreets/gunit v1.3.4/go.mod h1:ZjM1ozSIMJlAz/ay4SG8PeKF00ckUp+zMHZXV9/bvak=
 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=
@@ -468,6 +488,8 @@ github.com/stretchr/testify v1.3.0 h1:TivCn/peBQ7UY8ooIcPgZFpTNSz0Q2U6UrFlUfqbe0
 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/stretchr/testify v1.5.1 h1:nOGnQDM7FYENwehXlg/kFVnos3rEvtKTjRvOWSzb6H4=
+github.com/stretchr/testify v1.5.1/go.mod h1:5W2xD1RspED5o8YsWQXVCued0rvSQ+mT+I5cxcmMvtA=
 github.com/syndtr/goleveldb v1.0.0 h1:fBdIW9lB4Iz0n9khmH8w27SJ3QEJ7+IgjPEwGSZiFdE=
 github.com/syndtr/goleveldb v1.0.0/go.mod h1:ZVVdQEZoIme9iO1Ch2Jdy24qqXrMMOU6lpPAyBWyWuQ=
 github.com/tidwall/gjson v1.3.2 h1:+7p3qQFaH3fOMXAJSrdZwGKcOO/lYdGS0HqGhPqDdTI=
@@ -508,6 +530,8 @@ 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.opencensus.io v0.22.4 h1:LYy1Hy3MJdrCdMwwzxA/dRok4ejH+RwNGbuoD9fCjto=
+go.opencensus.io v0.22.4/go.mod h1:yxeiOL68Rb0Xd1ddK5vPZ/oVn4vY4Ynel7k9FzqtOIw=
 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/atomic v1.6.0 h1:Ezj3JGmsOnG1MoRWQkPBsKLe9DwWD9QeXzTRzzldNVk=
@@ -575,6 +599,8 @@ golang.org/x/net v0.0.0-20190620200207-3b0461eec859/go.mod h1:z5CRVTTTmAJ677TzLL
 golang.org/x/net v0.0.0-20190813141303-74dc4d7220e7/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/net v0.0.0-20200202094626-16171245cfb2 h1:CCH4IOTTfewWjGOlSp+zGcjutRKlBEZQ6wTn8ozI/nI=
+golang.org/x/net v0.0.0-20200202094626-16171245cfb2/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=

k8s/seaweedfs/Chart.yaml

@@ -1,4 +1,4 @@
 apiVersion: v1
 description: SeaweedFS
 name: seaweedfs
-version: 1.88
+version: 2.02

k8s/seaweedfs/templates/filer-statefulset.yaml

@@ -99,6 +99,9 @@ spec:
               {{- end }}
               filer \
               -port={{ .Values.filer.port }} \
+              {{- if .Values.filer.redirectOnRead }}
+              -redirectOnRead \
+              {{- end }}
               {{- if .Values.filer.disableHttp }}
               -disableHttp \
               {{- end }}
@@ -106,7 +109,24 @@ spec:
               -disableDirListing \
               {{- end }}
               -dirListLimit={{ .Values.filer.dirListLimit }} \
+              {{- if .Values.global.enableReplication }}
+              -defaultReplicaPlacement={{ .Values.global.replicationPlacment }} \
+              {{- else }}
+              -defaultReplicaPlacement={{ .Values.filer.defaultReplicaPlacement }} \
+              {{- end }}
+              {{- if .Values.filer.disableDirListing }}
+              -disableDirListing \
+              {{- end }}
+              {{- if .Values.filer.maxMB }}
+              -maxMB={{ .Values.filer.maxMB }} \
+              {{- end }}
+              {{- if .Values.filer.encryptVolumeData }}
+              -encryptVolumeData \
+              {{- end }}
               -ip=${POD_IP} \
+              {{- if gt (.Values.filer.replicas | int) 1 }}
+              -peers={{ range $index := until (.Values.filer.replicas | int) }}${SEAWEEDFS_FULLNAME}-filer-{{ $index }}.${SEAWEEDFS_FULLNAME}-filer:{{ $.Values.filer.port }}{{ if lt $index (sub ($.Values.filer.replicas | int) 1) }},{{ end }}{{ end }}
+              {{- end }}
               -master={{ range $index := until (.Values.master.replicas | int) }}${SEAWEEDFS_FULLNAME}-master-{{ $index }}.${SEAWEEDFS_FULLNAME}-master:{{ $.Values.master.port }}{{ if lt $index (sub ($.Values.master.replicas | int) 1) }},{{ end }}{{ end }}
           {{- if or (.Values.global.enableSecurity) (.Values.filer.extraVolumeMounts) }}    
           volumeMounts:
@@ -149,6 +169,7 @@ spec:
             periodSeconds: 15
             successThreshold: 1
             failureThreshold: 100
+            timeoutSeconds: 3
           livenessProbe:
             httpGet:
               path: /
@@ -158,6 +179,7 @@ spec:
             periodSeconds: 30
             successThreshold: 1
             failureThreshold: 5
+            timeoutSeconds: 3
           {{- if .Values.filer.resources }}
           resources:
             {{ tpl .Values.filer.resources . | nindent 12 | trim }}

k8s/seaweedfs/templates/master-statefulset.yaml

@@ -70,6 +70,12 @@ spec:
                   fieldPath: metadata.namespace
             - name: SEAWEEDFS_FULLNAME
               value: "{{ template "seaweedfs.name" . }}"
+            {{- if .Values.master.extraEnvironmentVars }}
+            {{- range $key, $value := .Values.master.extraEnvironmentVars }}
+            - name: {{ $key }}
+              value: {{ $value | quote }}
+            {{- end }}
+            {{- end }}
           command:
             - "/bin/sh"
             - "-ec"
@@ -84,6 +90,11 @@ spec:
               -port={{ .Values.master.port }} \
               -mdir=/data \
               -ip.bind={{ .Values.master.ipBind }} \
+              {{- if .Values.global.enableReplication }}
+              -defaultReplication={{ .Values.global.replicationPlacment }} \
+              {{- else }}
+              -defaultReplication={{ .Values.master.defaultReplication }} \
+              {{- end }}
               {{- if .Values.master.volumePreallocate }}
               -volumePreallocate \
               {{- end }}
@@ -94,6 +105,15 @@ spec:
               {{- if .Values.master.disableHttp }}
               -disableHttp \
               {{- end }}
+              {{- if .Values.master.pulseSeconds }}
+              -pulseSeconds={{ .Values.master.pulseSeconds }} \
+              {{- end }}
+              {{- if .Values.master.garbageThreshold }}
+              -garbageThreshold={{ .Values.master.garbageThreshold }} \
+              {{- end }}
+              {{- if .Values.master.metricsIntervalSec }}
+              -metrics.intervalSeconds={{ .Values.master.metricsIntervalSec }} \
+              {{- end }}
               -ip=${POD_NAME}.${SEAWEEDFS_FULLNAME}-master \
               -peers={{ range $index := until (.Values.master.replicas | int) }}${SEAWEEDFS_FULLNAME}-master-{{ $index }}.${SEAWEEDFS_FULLNAME}-master:{{ $.Values.master.port }}{{ if lt $index (sub ($.Values.master.replicas | int) 1) }},{{ end }}{{ end }}
           volumeMounts:
@@ -133,19 +153,21 @@ spec:
               path: /cluster/status
               port: {{ .Values.master.port }}
               scheme: HTTP
-            initialDelaySeconds: 5
-            periodSeconds: 15
+            initialDelaySeconds: 10
+            periodSeconds: 45
             successThreshold: 2
             failureThreshold: 100
+            timeoutSeconds: 5
           livenessProbe:
             httpGet:
               path: /cluster/status
               port: {{ .Values.master.port }}
               scheme: HTTP
             initialDelaySeconds: 20
-            periodSeconds: 10
+            periodSeconds: 30
             successThreshold: 1
-            failureThreshold: 6
+            failureThreshold: 4
+            timeoutSeconds: 5
           {{- if .Values.master.resources }}
           resources:
             {{ tpl .Values.master.resources . | nindent 12 | trim }}

k8s/seaweedfs/templates/s3-deployment.yaml

@@ -116,6 +116,7 @@ spec:
             periodSeconds: 15
             successThreshold: 1
             failureThreshold: 100
+            timeoutSeconds: 3
           livenessProbe:
             httpGet:
               path: /
@@ -125,6 +126,7 @@ spec:
             periodSeconds: 60
             successThreshold: 1
             failureThreshold: 20
+            timeoutSeconds: 3
           {{- if .Values.s3.resources }}
           resources:
             {{ tpl .Values.s3.resources . | nindent 12 | trim }}

k8s/seaweedfs/templates/seaweefs-grafana-dashboard.yaml

@@ -91,7 +91,7 @@ data:
               "thresholds": [],
               "timeFrom": null,
               "timeShift": null,
-              "title": "Filer Request Duration 95th percentile",
+              "title": "Filer Request Duration 80th percentile",
               "tooltip": {
                 "msResolution": true,
                 "shared": true,
@@ -1349,4 +1349,4 @@ data:
       "title": "SeaweedFS",
       "version": 3
     }
-{{- end }}
+{{- end }}

k8s/seaweedfs/templates/volume-statefulset.yaml

@@ -12,6 +12,7 @@ metadata:
 spec:
   serviceName: {{ template "seaweedfs.name" . }}-volume
   replicas: {{ .Values.volume.replicas }}
+  podManagementPolicy: Parallel
   selector:
     matchLabels:
       app: {{ template "seaweedfs.name" . }}
@@ -33,7 +34,7 @@ spec:
       restartPolicy: {{ default .Values.global.restartPolicy .Values.volume.restartPolicy }}
       {{- if .Values.volume.tolerations }}
       tolerations:
-        {{ tpl .Values.volume.tolerations . | nindent 8 | trim }}
+      {{ tpl .Values.volume.tolerations . | nindent 8 | trim }}
       {{- end }}
       {{- if .Values.global.imagePullSecrets }}
       imagePullSecrets:
@@ -62,7 +63,7 @@ spec:
                 fieldRef:
                   fieldPath: status.hostIP
             - name: SEAWEEDFS_FULLNAME
-              value: "{{ template "seaweedfs.name" . }}"                  
+              value: "{{ template "seaweedfs.name" . }}"
           command:
             - "/bin/sh"
             - "-ec"
@@ -91,6 +92,16 @@ spec:
                 {{- if .Values.volume.imagesFixOrientation }}
                 -images.fix.orientation \
                 {{- end }}
+                {{- if .Values.volume.pulseSeconds }}
+                -pulseSeconds={{ .Values.volume.pulseSeconds }} \
+                {{- end }}
+                {{- if .Values.volume.index }}
+                -index={{ .Values.volume.index }} \
+                {{- end }}
+                {{- if .Values.volume.fileSizeLimitMB }}
+                -fileSizeLimitMB={{ .Values.volume.fileSizeLimitMB }} \
+                {{- end }}
+                -minFreeSpacePercent={{ .Values.volume.minFreeSpacePercent }} \
                 -ip=${POD_NAME}.${SEAWEEDFS_FULLNAME}-volume \
                 -compactionMBps={{ .Values.volume.compactionMBps }} \
                 -mserver={{ range $index := until (.Values.master.replicas | int) }}${SEAWEEDFS_FULLNAME}-master-{{ $index }}.${SEAWEEDFS_FULLNAME}-master:{{ $.Values.master.port }}{{ if lt $index (sub ($.Values.master.replicas | int) 1) }},{{ end }}{{ end }}
@@ -131,19 +142,21 @@ spec:
               path: /status
               port: {{ .Values.volume.port }}
               scheme: HTTP
-            initialDelaySeconds: 5
-            periodSeconds: 15
+            initialDelaySeconds: 15
+            periodSeconds: 90
             successThreshold: 1
             failureThreshold: 100
+            timeoutSeconds: 5
           livenessProbe:
             httpGet:
               path: /status
               port: {{ .Values.volume.port }}
               scheme: HTTP
             initialDelaySeconds: 20
-            periodSeconds: 30
+            periodSeconds: 90
             successThreshold: 1
-            failureThreshold: 10
+            failureThreshold: 4
+            timeoutSeconds: 5
           {{- if .Values.volume.resources }}
           resources:
             {{ tpl .Values.volume.resources . | nindent 12 | trim }}

k8s/seaweedfs/values.yaml

@@ -4,7 +4,7 @@ global:
   registry: ""
   repository: ""
   imageName: chrislusf/seaweedfs
-  imageTag: "1.88"
+  imageTag: "2.02"
   imagePullPolicy: IfNotPresent
   imagePullSecrets: imagepullsecret
   restartPolicy: Always
@@ -14,6 +14,13 @@ global:
     enabled: false
     gatewayHost: null
     gatewayPort: null
+  # if enabled will use global.replicationPlacment and override master & filer defaultReplicaPlacement config
+  enableReplication: false
+  #  replication type is XYZ:
+  # X number of replica in other data centers
+  # Y number of replica in other racks in the same data center
+  # Z number of replica in other servers in the same rack
+  replicationPlacment: "001"
 
 image:
   registry: ""
@@ -31,8 +38,20 @@ master:
   grpcPort: 19333
   ipBind: "0.0.0.0"
   volumePreallocate: false
+  #Master stops directing writes to oversized volumes
   volumeSizeLimitMB: 30000
   loggingOverrideLevel: null
+  #number of seconds between heartbeats, default 5
+  pulseSeconds: null
+  #threshold to vacuum and reclaim spaces, default 0.3 (30%)
+  garbageThreshold: null
+  #Prometheus push interval in seconds, default 15
+  metricsIntervalSec: 15
+  #  replication type is XYZ:
+  # X number of replica in other data centers
+  # Y number of replica in other racks in the same data center
+  # Z number of replica in other servers in the same rack
+  defaultReplication: "000"
 
   # Disable http request, only gRpc operations are allowed
   disableHttp: false
@@ -87,6 +106,11 @@ master:
   # ref: https://kubernetes.io/docs/concepts/configuration/pod-priority-preemption/
   priorityClassName: ""
 
+  extraEnvironmentVars:
+    WEED_MASTER_VOLUME_GROWTH_COPY_1: 7
+    WEED_MASTER_VOLUME_GROWTH_COPY_2: 6
+    WEED_MASTER_VOLUME_GROWTH_COPY_3: 3
+    WEED_MASTER_VOLUME_GROWTH_COPY_OTHER: 1
 
 volume:
   enabled: true
@@ -100,9 +124,18 @@ volume:
   ipBind: "0.0.0.0"
   replicas: 1
   loggingOverrideLevel: null
+  # number of seconds between heartbeats, must be smaller than or equal to the master's setting
+  pulseSeconds: null
+  # Choose [memory|leveldb|leveldbMedium|leveldbLarge] mode for memory~performance balance., default memory
+  index: null
+  # limit file size to avoid out of memory, default 256mb
+  fileSizeLimitMB: null
+  # minimum free disk space(in percents). If free disk space lower this value - all volumes marks as ReadOnly
+  minFreeSpacePercent: 1
+
 
   # limit background compaction or copying speed in mega bytes per second
-  compactionMBps: "40"
+  compactionMBps: "50"
 
   # Directories to store data files. dir[,dir]... (default "/tmp")
   dir: "/data"
@@ -177,6 +210,20 @@ filer:
   port: 8888
   grpcPort: 18888
   loggingOverrideLevel: null
+  #  replication type is XYZ:
+  # X number of replica in other data centers
+  # Y number of replica in other racks in the same data center
+  # Z number of replica in other servers in the same rack
+  defaultReplicaPlacement: "000"
+  # turn off directory listing
+  disableDirListing: false
+  # split files larger than the limit, default 32
+  maxMB: null
+  # encrypt data on volume servers
+  encryptVolumeData: false
+
+  # Whether proxy or redirect to volume server during file GET request
+  redirectOnRead: false
 
   # Limit sub dir listing size (default 100000)
   dirListLimit: 100000
@@ -237,11 +284,6 @@ filer:
   # ref: https://kubernetes.io/docs/concepts/configuration/pod-priority-preemption/
   priorityClassName: ""
 
-  dbSchema:
-    imageName: db-schema
-    imageTag: "development"
-    imageOverride: ""
-
   # extraEnvVars is a list of extra enviroment variables to set with the stateful set.
   extraEnvironmentVars:
     WEED_MYSQL_ENABLED: "true"
@@ -260,6 +302,8 @@ filer:
     WEED_FILER_BUCKETS_FOLDER: "/buckets"
     # directories under this folder will be store message queue data
     WEED_FILER_QUEUES_FOLDER: "/queues"
+    # WEED_FILER_OPTIONS_BUCKETS_FSYNC a list of buckets names with all write requests fsync=true
+    WEED_FILER_OPTIONS_BUCKETS_FSYNC: []
 
 s3:
   enabled: true

other/java/client/pom.xml

@@ -5,7 +5,7 @@
 
     <groupId>com.github.chrislusf</groupId>
     <artifactId>seaweedfs-client</artifactId>
-    <version>1.4.5</version>
+    <version>1.4.8</version>
 
     <parent>
         <groupId>org.sonatype.oss</groupId>

other/java/client/pom.xml.deploy

@@ -5,7 +5,7 @@
 
     <groupId>com.github.chrislusf</groupId>
     <artifactId>seaweedfs-client</artifactId>
-    <version>1.4.5</version>
+    <version>1.4.8</version>
 
     <parent>
         <groupId>org.sonatype.oss</groupId>

other/java/client/pom_debug.xml

@@ -5,7 +5,7 @@
 
     <groupId>com.github.chrislusf</groupId>
     <artifactId>seaweedfs-client</artifactId>
-    <version>1.4.5</version>
+    <version>1.4.8</version>
 
     <parent>
         <groupId>org.sonatype.oss</groupId>

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

@@ -23,7 +23,7 @@ public class SeaweedRead {
     // 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 {
+                            final int bufferLength, final long fileSize) throws IOException {
 
         List<ChunkView> chunkViews = viewFromVisibles(visibleIntervals, position, bufferLength);
 
@@ -42,6 +42,14 @@ public class SeaweedRead {
         long readCount = 0;
         int startOffset = bufferOffset;
         for (ChunkView chunkView : chunkViews) {
+
+            if (startOffset < chunkView.logicOffset) {
+                long gap = chunkView.logicOffset - startOffset;
+                LOG.debug("zero [{},{})", startOffset, startOffset + gap);
+                readCount += gap;
+                startOffset += gap;
+            }
+
             FilerProto.Locations locations = vid2Locations.get(parseVolumeId(chunkView.fileId));
             if (locations == null || locations.getLocationsCount() == 0) {
                 LOG.error("failed to locate {}", chunkView.fileId);
@@ -51,11 +59,22 @@ public class SeaweedRead {
 
             int len = readChunkView(position, buffer, startOffset, chunkView, locations);
 
+            LOG.debug("read [{},{}) {} size {}", startOffset, startOffset + len, chunkView.fileId, chunkView.size);
+
             readCount += len;
             startOffset += len;
 
         }
 
+        long limit = Math.min(bufferLength, fileSize);
+
+        if (startOffset < limit) {
+            long gap = limit - startOffset;
+            LOG.debug("zero2 [{},{})", startOffset, startOffset + gap);
+            readCount += gap;
+            startOffset += gap;
+        }
+
         return readCount;
     }
 
@@ -71,7 +90,7 @@ public class SeaweedRead {
         int len = (int) chunkView.size;
         LOG.debug("readChunkView fid:{} chunkData.length:{} chunkView.offset:{} buffer.length:{} startOffset:{} len:{}",
                 chunkView.fileId, chunkData.length, chunkView.offset, buffer.length, startOffset, len);
-        System.arraycopy(chunkData, (int) chunkView.offset, buffer, startOffset, len);
+        System.arraycopy(chunkData, startOffset - (int) (chunkView.logicOffset - chunkView.offset), buffer, startOffset, len);
 
         return len;
     }
@@ -93,7 +112,7 @@ public class SeaweedRead {
             Header contentEncodingHeader = entity.getContentEncoding();
 
             if (contentEncodingHeader != null) {
-                HeaderElement[] encodings =contentEncodingHeader.getElements();
+                HeaderElement[] encodings = contentEncodingHeader.getElements();
                 for (int i = 0; i < encodings.length; i++) {
                     if (encodings[i].getName().equalsIgnoreCase("gzip")) {
                         entity = new GzipDecompressingEntity(entity);
@@ -134,18 +153,19 @@ public class SeaweedRead {
 
         long stop = offset + size;
         for (VisibleInterval chunk : visibleIntervals) {
-            if (chunk.start <= offset && offset < chunk.stop && offset < stop) {
+            long chunkStart = Math.max(offset, chunk.start);
+            long chunkStop = Math.min(stop, chunk.stop);
+            if (chunkStart < chunkStop) {
                 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,
+                        chunkStart - chunk.start + chunk.chunkOffset,
+                        chunkStop - chunkStart,
+                        chunkStart,
                         isFullChunk,
                         chunk.cipherKey,
                         chunk.isCompressed
                 ));
-                offset = Math.min(chunk.stop, stop);
             }
         }
         return views;
@@ -160,7 +180,13 @@ public class SeaweedRead {
         Arrays.sort(chunks, new Comparator<FilerProto.FileChunk>() {
             @Override
             public int compare(FilerProto.FileChunk a, FilerProto.FileChunk b) {
-                return (int) (a.getMtime() - b.getMtime());
+                // if just a.getMtime() - b.getMtime(), it will overflow!
+                if (a.getMtime() < b.getMtime()) {
+                    return -1;
+                } else if (a.getMtime() > b.getMtime()) {
+                    return 1;
+                }
+                return 0;
             }
         });
 
@@ -181,6 +207,7 @@ public class SeaweedRead {
                 chunk.getOffset() + chunk.getSize(),
                 chunk.getFileId(),
                 chunk.getMtime(),
+                0,
                 true,
                 chunk.getCipherKey().toByteArray(),
                 chunk.getIsCompressed()
@@ -203,6 +230,7 @@ public class SeaweedRead {
                         chunk.getOffset(),
                         v.fileId,
                         v.modifiedTime,
+                        v.chunkOffset,
                         false,
                         v.cipherKey,
                         v.isCompressed
@@ -215,6 +243,7 @@ public class SeaweedRead {
                         v.stop,
                         v.fileId,
                         v.modifiedTime,
+                        v.chunkOffset + (chunkStop - v.start),
                         false,
                         v.cipherKey,
                         v.isCompressed
@@ -247,6 +276,10 @@ public class SeaweedRead {
         return fileId;
     }
 
+    public static long fileSize(FilerProto.Entry entry) {
+        return Math.max(totalSize(entry.getChunksList()), entry.getAttributes().getFileSize());
+    }
+
     public static long totalSize(List<FilerProto.FileChunk> chunksList) {
         long size = 0;
         for (FilerProto.FileChunk chunk : chunksList) {
@@ -263,15 +296,17 @@ public class SeaweedRead {
         public final long stop;
         public final long modifiedTime;
         public final String fileId;
+        public final long chunkOffset;
         public final boolean isFullChunk;
         public final byte[] cipherKey;
         public final boolean isCompressed;
 
-        public VisibleInterval(long start, long stop, String fileId, long modifiedTime, boolean isFullChunk, byte[] cipherKey, boolean isCompressed) {
+        public VisibleInterval(long start, long stop, String fileId, long modifiedTime, long chunkOffset, boolean isFullChunk, byte[] cipherKey, boolean isCompressed) {
             this.start = start;
             this.stop = stop;
             this.modifiedTime = modifiedTime;
             this.fileId = fileId;
+            this.chunkOffset = chunkOffset;
             this.isFullChunk = isFullChunk;
             this.cipherKey = cipherKey;
             this.isCompressed = isCompressed;

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

@@ -58,6 +58,12 @@ service SeaweedFiler {
     rpc LocateBroker (LocateBrokerRequest) returns (LocateBrokerResponse) {
     }
 
+    rpc KvGet (KvGetRequest) returns (KvGetResponse) {
+    }
+
+    rpc KvPut (KvPutRequest) returns (KvPutResponse) {
+    }
+
 }
 
 //////////////////////////////////////////////////
@@ -89,6 +95,8 @@ message Entry {
     repeated FileChunk chunks = 3;
     FuseAttributes attributes = 4;
     map<string, bytes> extended = 5;
+    bytes hard_link_id = 7;
+    int32 hard_link_counter = 8; // only exists in hard link meta data
 }
 
 message FullEntry {
@@ -102,6 +110,7 @@ message EventNotification {
     bool delete_chunks = 3;
     string new_parent_path = 4;
     bool is_from_other_cluster = 5;
+    repeated int32 signatures = 6;
 }
 
 message FileChunk {
@@ -150,6 +159,7 @@ message CreateEntryRequest {
     Entry entry = 2;
     bool o_excl = 3;
     bool is_from_other_cluster = 4;
+    repeated int32 signatures = 5;
 }
 
 message CreateEntryResponse {
@@ -160,6 +170,7 @@ message UpdateEntryRequest {
     string directory = 1;
     Entry entry = 2;
     bool is_from_other_cluster = 3;
+    repeated int32 signatures = 4;
 }
 message UpdateEntryResponse {
 }
@@ -180,6 +191,7 @@ message DeleteEntryRequest {
     bool is_recursive = 5;
     bool ignore_recursive_error = 6;
     bool is_from_other_cluster = 7;
+    repeated int32 signatures = 8;
 }
 
 message DeleteEntryResponse {
@@ -262,12 +274,16 @@ message GetFilerConfigurationResponse {
     uint32 max_mb = 4;
     string dir_buckets = 5;
     bool cipher = 7;
+    int32 signature = 8;
+    string metrics_address = 9;
+    int32 metrics_interval_sec = 10;
 }
 
 message SubscribeMetadataRequest {
     string client_name = 1;
     string path_prefix = 2;
     int64 since_ns = 3;
+    int32 signature = 4;
 }
 message SubscribeMetadataResponse {
     string directory = 1;
@@ -302,3 +318,19 @@ message LocateBrokerResponse {
     }
     repeated Resource resources = 2;
 }
+
+// Key-Value operations
+message KvGetRequest {
+    bytes key = 1;
+}
+message KvGetResponse {
+    bytes value = 1;
+    string error = 2;
+}
+message KvPutRequest {
+    bytes key = 1;
+    bytes value = 2;
+}
+message KvPutResponse {
+    string error = 1;
+}

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

@@ -301,7 +301,7 @@
     </snapshotRepository>
   </distributionManagement>
   <properties>
-    <seaweedfs.client.version>1.4.5</seaweedfs.client.version>
+    <seaweedfs.client.version>1.4.8</seaweedfs.client.version>
     <hadoop.version>2.9.2</hadoop.version>
   </properties>
 </project>

other/java/hdfs2/pom.xml

@@ -5,7 +5,7 @@
     <modelVersion>4.0.0</modelVersion>
 
     <properties>
-        <seaweedfs.client.version>1.4.5</seaweedfs.client.version>
+        <seaweedfs.client.version>1.4.8</seaweedfs.client.version>
         <hadoop.version>2.9.2</hadoop.version>
     </properties>
 

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

@@ -5,7 +5,6 @@ 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;
@@ -14,20 +13,19 @@ import seaweedfs.client.FilerProto;
 
 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";
+    public static final int FS_SEAWEED_DEFAULT_PORT = 8888;
+    public static final String FS_SEAWEED_BUFFER_SIZE = "fs.seaweed.buffer.size";
+    public static final int FS_SEAWEED_DEFAULT_BUFFER_SIZE = 4 * 1024 * 1024;
 
     private static final Logger LOG = LoggerFactory.getLogger(SeaweedFileSystem.class);
 
@@ -75,8 +73,9 @@ public class SeaweedFileSystem extends FileSystem {
         path = qualify(path);
 
         try {
-            FSInputStream inputStream = seaweedFileSystemStore.openFileForRead(path, statistics, bufferSize);
-            return new FSDataInputStream(new BufferedFSInputStream(inputStream, 16 * 1024 * 1024));
+            int seaweedBufferSize = this.getConf().getInt(FS_SEAWEED_BUFFER_SIZE, FS_SEAWEED_DEFAULT_BUFFER_SIZE);
+            FSInputStream inputStream = seaweedFileSystemStore.openFileForRead(path, statistics, seaweedBufferSize);
+            return new FSDataInputStream(new BufferedFSInputStream(inputStream, 4 * seaweedBufferSize));
         } catch (Exception ex) {
             LOG.warn("open path: {} bufferSize:{}", path, bufferSize, ex);
             return null;
@@ -93,7 +92,8 @@ public class SeaweedFileSystem extends FileSystem {
 
         try {
             String replicaPlacement = String.format("%03d", replication - 1);
-            OutputStream outputStream = seaweedFileSystemStore.createFile(path, overwrite, permission, bufferSize, replicaPlacement);
+            int seaweedBufferSize = this.getConf().getInt(FS_SEAWEED_BUFFER_SIZE, FS_SEAWEED_DEFAULT_BUFFER_SIZE);
+            OutputStream outputStream = seaweedFileSystemStore.createFile(path, overwrite, permission, seaweedBufferSize, replicaPlacement);
             return new FSDataOutputStream(outputStream, statistics);
         } catch (Exception ex) {
             LOG.warn("create path: {} bufferSize:{} blockSize:{}", path, bufferSize, blockSize, ex);
@@ -103,8 +103,9 @@ public class SeaweedFileSystem extends FileSystem {
 
     /**
      * {@inheritDoc}
+     *
      * @throws FileNotFoundException if the parent directory is not present -or
-     * is not a directory.
+     *                               is not a directory.
      */
     @Override
     public FSDataOutputStream createNonRecursive(Path path,
@@ -121,9 +122,10 @@ public class SeaweedFileSystem extends FileSystem {
                 throw new FileAlreadyExistsException("Not a directory: " + parent);
             }
         }
+        int seaweedBufferSize = this.getConf().getInt(FS_SEAWEED_BUFFER_SIZE, FS_SEAWEED_DEFAULT_BUFFER_SIZE);
         return create(path, permission,
                 flags.contains(CreateFlag.OVERWRITE), bufferSize,
-                replication, blockSize, progress);
+                replication, seaweedBufferSize, progress);
     }
 
     @Override
@@ -133,7 +135,8 @@ public class SeaweedFileSystem extends FileSystem {
 
         path = qualify(path);
         try {
-            OutputStream outputStream = seaweedFileSystemStore.createFile(path, false, null, bufferSize, "");
+            int seaweedBufferSize = this.getConf().getInt(FS_SEAWEED_BUFFER_SIZE, FS_SEAWEED_DEFAULT_BUFFER_SIZE);
+            OutputStream outputStream = seaweedFileSystemStore.createFile(path, false, null, seaweedBufferSize, "");
             return new FSDataOutputStream(outputStream, statistics);
         } catch (Exception ex) {
             LOG.warn("append path: {} bufferSize:{}", path, bufferSize, ex);
@@ -338,9 +341,7 @@ public class SeaweedFileSystem extends FileSystem {
 
     @Override
     public void createSymlink(final Path target, final Path link,
-                              final boolean createParent) throws AccessControlException,
-            FileAlreadyExistsException, FileNotFoundException,
-            ParentNotDirectoryException, UnsupportedFileSystemException,
+                              final boolean createParent) throws
             IOException {
         // Supporting filesystems should override this method
         throw new UnsupportedOperationException(

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

@@ -8,14 +8,10 @@ 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 seaweedfs.client.*;
 
 import java.io.FileNotFoundException;
 import java.io.IOException;
-import java.io.InputStream;
 import java.io.OutputStream;
 import java.util.ArrayList;
 import java.util.Arrays;
@@ -124,7 +120,7 @@ public class SeaweedFileSystemStore {
 
     private FileStatus doGetFileStatus(Path path, FilerProto.Entry entry) {
         FilerProto.FuseAttributes attributes = entry.getAttributes();
-        long length = SeaweedRead.totalSize(entry.getChunksList());
+        long length = SeaweedRead.fileSize(entry);
         boolean isDir = entry.getIsDirectory();
         int block_replication = 1;
         int blocksize = 512;
@@ -185,7 +181,7 @@ public class SeaweedFileSystemStore {
                 entry.mergeFrom(existingEntry);
                 entry.getAttributesBuilder().setMtime(now);
                 LOG.debug("createFile merged entry path:{} entry:{} from:{}", path, entry, existingEntry);
-                writePosition = SeaweedRead.totalSize(existingEntry.getChunksList());
+                writePosition = SeaweedRead.fileSize(existingEntry);
                 replication = existingEntry.getAttributes().getReplication();
             }
         }
@@ -202,6 +198,7 @@ public class SeaweedFileSystemStore {
                     .clearGroupName()
                     .addAllGroupName(Arrays.asList(userGroupInformation.getGroupNames()))
                 );
+            SeaweedWrite.writeMeta(filerGrpcClient, getParentDirectory(path), entry);
         }
 
         return new SeaweedOutputStream(filerGrpcClient, path, entry, writePosition, bufferSize, replication);

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

@@ -41,7 +41,7 @@ public class SeaweedInputStream extends FSInputStream {
         this.statistics = statistics;
         this.path = path;
         this.entry = entry;
-        this.contentLength = SeaweedRead.totalSize(entry.getChunksList());
+        this.contentLength = SeaweedRead.fileSize(entry);
         this.bufferSize = bufferSize;
 
         this.visibleIntervalList = SeaweedRead.nonOverlappingVisibleIntervals(filerGrpcClient, entry.getChunksList());
@@ -87,7 +87,7 @@ public class SeaweedInputStream extends FSInputStream {
             throw new IllegalArgumentException("requested read length is more than will fit after requested offset in buffer");
         }
 
-        long bytesRead = SeaweedRead.read(this.filerGrpcClient, this.visibleIntervalList, this.position, b, off, len);
+        long bytesRead = SeaweedRead.read(this.filerGrpcClient, this.visibleIntervalList, this.position, b, off, len, SeaweedRead.fileSize(entry));
         if (bytesRead > Integer.MAX_VALUE) {
             throw new IOException("Unexpected Content-Length");
         }

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

@@ -120,6 +120,188 @@
       </plugin>
     </plugins>
   </build>
+  <dependencies>
+    <dependency>
+      <groupId>org.apache.hadoop</groupId>
+      <artifactId>hadoop-client</artifactId>
+      <version>3.1.1</version>
+      <scope>provided</scope>
+      <exclusions>
+        <exclusion>
+          <artifactId>hadoop-hdfs-client</artifactId>
+          <groupId>org.apache.hadoop</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>hadoop-yarn-api</artifactId>
+          <groupId>org.apache.hadoop</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>hadoop-yarn-client</artifactId>
+          <groupId>org.apache.hadoop</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>hadoop-mapreduce-client-core</artifactId>
+          <groupId>org.apache.hadoop</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>hadoop-mapreduce-client-jobclient</artifactId>
+          <groupId>org.apache.hadoop</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>hadoop-annotations</artifactId>
+          <groupId>org.apache.hadoop</groupId>
+        </exclusion>
+      </exclusions>
+    </dependency>
+    <dependency>
+      <groupId>org.apache.hadoop</groupId>
+      <artifactId>hadoop-common</artifactId>
+      <version>3.1.1</version>
+      <scope>provided</scope>
+      <exclusions>
+        <exclusion>
+          <artifactId>commons-cli</artifactId>
+          <groupId>commons-cli</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>commons-math3</artifactId>
+          <groupId>org.apache.commons</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>commons-io</artifactId>
+          <groupId>commons-io</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>commons-net</artifactId>
+          <groupId>commons-net</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>commons-collections</artifactId>
+          <groupId>commons-collections</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>javax.servlet-api</artifactId>
+          <groupId>javax.servlet</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jetty-server</artifactId>
+          <groupId>org.eclipse.jetty</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jetty-util</artifactId>
+          <groupId>org.eclipse.jetty</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jetty-servlet</artifactId>
+          <groupId>org.eclipse.jetty</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jetty-webapp</artifactId>
+          <groupId>org.eclipse.jetty</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jsp-api</artifactId>
+          <groupId>javax.servlet.jsp</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jersey-core</artifactId>
+          <groupId>com.sun.jersey</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jersey-servlet</artifactId>
+          <groupId>com.sun.jersey</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jersey-json</artifactId>
+          <groupId>com.sun.jersey</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jersey-server</artifactId>
+          <groupId>com.sun.jersey</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>log4j</artifactId>
+          <groupId>log4j</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>commons-lang</artifactId>
+          <groupId>commons-lang</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>commons-beanutils</artifactId>
+          <groupId>commons-beanutils</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>commons-configuration2</artifactId>
+          <groupId>org.apache.commons</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>commons-lang3</artifactId>
+          <groupId>org.apache.commons</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>slf4j-log4j12</artifactId>
+          <groupId>org.slf4j</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>avro</artifactId>
+          <groupId>org.apache.avro</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>re2j</artifactId>
+          <groupId>com.google.re2j</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>hadoop-auth</artifactId>
+          <groupId>org.apache.hadoop</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jsch</artifactId>
+          <groupId>com.jcraft</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>curator-client</artifactId>
+          <groupId>org.apache.curator</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>curator-recipes</artifactId>
+          <groupId>org.apache.curator</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>htrace-core4</artifactId>
+          <groupId>org.apache.htrace</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>zookeeper</artifactId>
+          <groupId>org.apache.zookeeper</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>commons-compress</artifactId>
+          <groupId>org.apache.commons</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>kerb-simplekdc</artifactId>
+          <groupId>org.apache.kerby</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>jackson-databind</artifactId>
+          <groupId>com.fasterxml.jackson.core</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>stax2-api</artifactId>
+          <groupId>org.codehaus.woodstox</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>woodstox-core</artifactId>
+          <groupId>com.fasterxml.woodstox</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>hadoop-annotations</artifactId>
+          <groupId>org.apache.hadoop</groupId>
+        </exclusion>
+      </exclusions>
+    </dependency>
+  </dependencies>
   <distributionManagement>
     <snapshotRepository>
       <id>ossrh</id>
@@ -127,7 +309,7 @@
     </snapshotRepository>
   </distributionManagement>
   <properties>
-    <seaweedfs.client.version>1.4.5</seaweedfs.client.version>
+    <seaweedfs.client.version>1.4.8</seaweedfs.client.version>
     <hadoop.version>3.1.1</hadoop.version>
   </properties>
 </project>

other/java/hdfs3/pom.xml

@@ -5,7 +5,7 @@
     <modelVersion>4.0.0</modelVersion>
 
     <properties>
-        <seaweedfs.client.version>1.4.5</seaweedfs.client.version>
+        <seaweedfs.client.version>1.4.8</seaweedfs.client.version>
         <hadoop.version>3.1.1</hadoop.version>
     </properties>
 

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

@@ -5,7 +5,6 @@ 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;
@@ -14,20 +13,19 @@ import seaweedfs.client.FilerProto;
 
 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";
+    public static final int FS_SEAWEED_DEFAULT_PORT = 8888;
+    public static final String FS_SEAWEED_BUFFER_SIZE = "fs.seaweed.buffer.size";
+    public static final int FS_SEAWEED_DEFAULT_BUFFER_SIZE = 4 * 1024 * 1024;
 
     private static final Logger LOG = LoggerFactory.getLogger(SeaweedFileSystem.class);
 
@@ -75,8 +73,9 @@ public class SeaweedFileSystem extends FileSystem {
         path = qualify(path);
 
         try {
-            FSInputStream inputStream = seaweedFileSystemStore.openFileForRead(path, statistics, bufferSize);
-            return new FSDataInputStream(new BufferedFSInputStream(inputStream, 16 * 1024 * 1024));
+            int seaweedBufferSize = this.getConf().getInt(FS_SEAWEED_BUFFER_SIZE, FS_SEAWEED_DEFAULT_BUFFER_SIZE);
+            FSInputStream inputStream = seaweedFileSystemStore.openFileForRead(path, statistics, seaweedBufferSize);
+            return new FSDataInputStream(new BufferedFSInputStream(inputStream, 4 * seaweedBufferSize));
         } catch (Exception ex) {
             LOG.warn("open path: {} bufferSize:{}", path, bufferSize, ex);
             return null;
@@ -93,7 +92,8 @@ public class SeaweedFileSystem extends FileSystem {
 
         try {
             String replicaPlacement = String.format("%03d", replication - 1);
-            OutputStream outputStream = seaweedFileSystemStore.createFile(path, overwrite, permission, bufferSize, replicaPlacement);
+            int seaweedBufferSize = this.getConf().getInt(FS_SEAWEED_BUFFER_SIZE, FS_SEAWEED_DEFAULT_BUFFER_SIZE);
+            OutputStream outputStream = seaweedFileSystemStore.createFile(path, overwrite, permission, seaweedBufferSize, replicaPlacement);
             return new FSDataOutputStream(outputStream, statistics);
         } catch (Exception ex) {
             LOG.warn("create path: {} bufferSize:{} blockSize:{}", path, bufferSize, blockSize, ex);
@@ -103,8 +103,9 @@ public class SeaweedFileSystem extends FileSystem {
 
     /**
      * {@inheritDoc}
+     *
      * @throws FileNotFoundException if the parent directory is not present -or
-     * is not a directory.
+     *                               is not a directory.
      */
     @Override
     public FSDataOutputStream createNonRecursive(Path path,
@@ -121,9 +122,10 @@ public class SeaweedFileSystem extends FileSystem {
                 throw new FileAlreadyExistsException("Not a directory: " + parent);
             }
         }
+        int seaweedBufferSize = this.getConf().getInt(FS_SEAWEED_BUFFER_SIZE, FS_SEAWEED_DEFAULT_BUFFER_SIZE);
         return create(path, permission,
                 flags.contains(CreateFlag.OVERWRITE), bufferSize,
-                replication, blockSize, progress);
+                replication, seaweedBufferSize, progress);
     }
 
     @Override
@@ -133,7 +135,8 @@ public class SeaweedFileSystem extends FileSystem {
 
         path = qualify(path);
         try {
-            OutputStream outputStream = seaweedFileSystemStore.createFile(path, false, null, bufferSize, "");
+            int seaweedBufferSize = this.getConf().getInt(FS_SEAWEED_BUFFER_SIZE, FS_SEAWEED_DEFAULT_BUFFER_SIZE);
+            OutputStream outputStream = seaweedFileSystemStore.createFile(path, false, null, seaweedBufferSize, "");
             return new FSDataOutputStream(outputStream, statistics);
         } catch (Exception ex) {
             LOG.warn("append path: {} bufferSize:{}", path, bufferSize, ex);
@@ -338,9 +341,7 @@ public class SeaweedFileSystem extends FileSystem {
 
     @Override
     public void createSymlink(final Path target, final Path link,
-                              final boolean createParent) throws AccessControlException,
-            FileAlreadyExistsException, FileNotFoundException,
-            ParentNotDirectoryException, UnsupportedFileSystemException,
+                              final boolean createParent) throws
             IOException {
         // Supporting filesystems should override this method
         throw new UnsupportedOperationException(

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

@@ -8,14 +8,10 @@ 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 seaweedfs.client.*;
 
 import java.io.FileNotFoundException;
 import java.io.IOException;
-import java.io.InputStream;
 import java.io.OutputStream;
 import java.util.ArrayList;
 import java.util.Arrays;
@@ -124,7 +120,7 @@ public class SeaweedFileSystemStore {
 
     private FileStatus doGetFileStatus(Path path, FilerProto.Entry entry) {
         FilerProto.FuseAttributes attributes = entry.getAttributes();
-        long length = SeaweedRead.totalSize(entry.getChunksList());
+        long length = SeaweedRead.fileSize(entry);
         boolean isDir = entry.getIsDirectory();
         int block_replication = 1;
         int blocksize = 512;
@@ -185,7 +181,7 @@ public class SeaweedFileSystemStore {
                 entry.mergeFrom(existingEntry);
                 entry.getAttributesBuilder().setMtime(now);
                 LOG.debug("createFile merged entry path:{} entry:{} from:{}", path, entry, existingEntry);
-                writePosition = SeaweedRead.totalSize(existingEntry.getChunksList());
+                writePosition = SeaweedRead.fileSize(existingEntry);
                 replication = existingEntry.getAttributes().getReplication();
             }
         }
@@ -202,6 +198,7 @@ public class SeaweedFileSystemStore {
                     .clearGroupName()
                     .addAllGroupName(Arrays.asList(userGroupInformation.getGroupNames()))
                 );
+            SeaweedWrite.writeMeta(filerGrpcClient, getParentDirectory(path), entry);
         }
 
         return new SeaweedOutputStream(filerGrpcClient, path, entry, writePosition, bufferSize, replication);

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

@@ -41,7 +41,7 @@ public class SeaweedInputStream extends FSInputStream {
         this.statistics = statistics;
         this.path = path;
         this.entry = entry;
-        this.contentLength = SeaweedRead.totalSize(entry.getChunksList());
+        this.contentLength = SeaweedRead.fileSize(entry);
         this.bufferSize = bufferSize;
 
         this.visibleIntervalList = SeaweedRead.nonOverlappingVisibleIntervals(filerGrpcClient, entry.getChunksList());
@@ -87,7 +87,7 @@ public class SeaweedInputStream extends FSInputStream {
             throw new IllegalArgumentException("requested read length is more than will fit after requested offset in buffer");
         }
 
-        long bytesRead = SeaweedRead.read(this.filerGrpcClient, this.visibleIntervalList, this.position, b, off, len);
+        long bytesRead = SeaweedRead.read(this.filerGrpcClient, this.visibleIntervalList, this.position, b, off, len, SeaweedRead.fileSize(entry));
         if (bytesRead > Integer.MAX_VALUE) {
             throw new IOException("Unexpected Content-Length");
         }

test/random_access/src/test/java/seaweedfs/file/MmapFileTest.java

@@ -0,0 +1,143 @@
+package seaweedfs.file;
+
+import org.junit.Test;
+
+import java.io.File;
+import java.io.FileOutputStream;
+import java.io.IOException;
+import java.io.RandomAccessFile;
+import java.nio.MappedByteBuffer;
+import java.nio.channels.FileChannel;
+
+public class MmapFileTest {
+
+    static File dir = new File("/Users/chris/tmp/mm/dev");
+
+    @Test
+    public void testMmap() {
+        try {
+            System.out.println("starting ...");
+
+            File f = new File(dir, "mmap_file.txt");
+            RandomAccessFile raf = new RandomAccessFile(f, "rw");
+            FileChannel fc = raf.getChannel();
+            MappedByteBuffer mbf = fc.map(FileChannel.MapMode.READ_ONLY, 0, fc.size());
+            fc.close();
+            raf.close();
+
+            FileOutputStream fos = new FileOutputStream(f);
+            fos.write("abcdefg".getBytes());
+            fos.close();
+            System.out.println("completed!");
+        } catch (Exception e) {
+            e.printStackTrace();
+        }
+    }
+
+    @Test
+    public void testBigMmap() throws IOException {
+        /*
+
+// new file
+I0817 09:48:02 25175 dir.go:147] create /dev/mmap_big.txt: OpenReadWrite+OpenCreate
+I0817 09:48:02 25175 wfs.go:116] AcquireHandle /dev/mmap_big.txt uid=502 gid=20
+I0817 09:48:02 25175 file.go:62] file Attr /dev/mmap_big.txt, open:1, size: 0
+I0817 09:48:02 25175 meta_cache_subscribe.go:32] creating /dev/mmap_big.txt
+
+//get channel
+I0817 09:48:26 25175 file.go:62] file Attr /dev/mmap_big.txt, open:1, size: 0
+
+I0817 09:48:32 25175 file.go:62] file Attr /dev/mmap_big.txt, open:1, size: 0
+I0817 09:48:32 25175 wfs.go:116] AcquireHandle /dev/mmap_big.txt uid=0 gid=0
+I0817 09:48:32 25175 filehandle.go:160] Release /dev/mmap_big.txt fh 14968871991130164560
+
+//fileChannel.map
+I0817 09:49:18 25175 file.go:62] file Attr /dev/mmap_big.txt, open:1, size: 0
+I0817 09:49:18 25175 file.go:112] /dev/mmap_big.txt file setattr set size=262144 chunks=0
+I0817 09:49:18 25175 file.go:62] file Attr /dev/mmap_big.txt, open:1, size: 262144
+I0817 09:49:18 25175 file.go:62] file Attr /dev/mmap_big.txt, open:1, size: 262144
+I0817 09:49:18 25175 file.go:62] file Attr /dev/mmap_big.txt, open:1, size: 262144
+
+// buffer.put
+I0817 09:49:49 25175 filehandle.go:57] /dev/mmap_big.txt read fh 14968871991130164560: [0,32768) size 32768 resp.Data len=0 cap=32768
+I0817 09:49:49 25175 reader_at.go:113] zero2 [0,32768)
+I0817 09:49:50 25175 file.go:62] file Attr /dev/mmap_big.txt, open:1, size: 262144
+
+I0817 09:49:53 25175 file.go:233] /dev/mmap_big.txt fsync file Fsync [ID=0x4 Node=0xe Uid=0 Gid=0 Pid=0] Handle 0x2 Flags 1
+
+//close
+I0817 09:50:14 25175 file.go:62] file Attr /dev/mmap_big.txt, open:1, size: 262144
+I0817 09:50:14 25175 dirty_page.go:130] saveToStorage /dev/mmap_big.txt 1,315b69812039e5 [0,4096) of 262144 bytes
+I0817 09:50:14 25175 file.go:274] /dev/mmap_big.txt existing 0 chunks adds 1 more
+I0817 09:50:14 25175 filehandle.go:218] /dev/mmap_big.txt set chunks: 1
+I0817 09:50:14 25175 filehandle.go:220] /dev/mmap_big.txt chunks 0: 1,315b69812039e5 [0,4096)
+I0817 09:50:14 25175 meta_cache_subscribe.go:23] deleting /dev/mmap_big.txt
+I0817 09:50:14 25175 meta_cache_subscribe.go:32] creating /dev/mmap_big.txt
+
+// end of test
+I0817 09:50:41 25175 file.go:62] file Attr /dev/mmap_big.txt, open:1, size: 262144
+I0817 09:50:41 25175 filehandle.go:160] Release /dev/mmap_big.txt fh 14968871991130164560
+
+         */
+        // Create file object
+        File file = new File(dir, "mmap_big.txt");
+
+        try (RandomAccessFile randomAccessFile = new RandomAccessFile(file, "rw")) {
+            // Get file channel in read-write mode
+            FileChannel fileChannel = randomAccessFile.getChannel();
+
+            // Get direct byte buffer access using channel.map() operation
+            MappedByteBuffer buffer = fileChannel.map(FileChannel.MapMode.READ_WRITE, 0, 4096 * 8 * 8);
+
+            //Write the content using put methods
+            buffer.put("howtodoinjava.com".getBytes());
+        }
+
+/*
+> meta.cat /dev/mmap_big.txt
+{
+  "name": "mmap_big.txt",
+  "isDirectory": false,
+  "chunks": [
+    {
+      "fileId": "1,315b69812039e5",
+      "offset": "0",
+      "size": "4096",
+      "mtime": "1597683014026365000",
+      "eTag": "985ab0ac",
+      "sourceFileId": "",
+      "fid": {
+        "volumeId": 1,
+        "fileKey": "3234665",
+        "cookie": 2166372837
+      },
+      "sourceFid": null,
+      "cipherKey": null,
+      "isCompressed": true,
+      "isChunkManifest": false
+    }
+  ],
+  "attributes": {
+    "fileSize": "262144",
+    "mtime": "1597683014",
+    "fileMode": 420,
+    "uid": 502,
+    "gid": 20,
+    "crtime": "1597682882",
+    "mime": "application/octet-stream",
+    "replication": "",
+    "collection": "",
+    "ttlSec": 0,
+    "userName": "",
+    "groupName": [
+    ],
+    "symlinkTarget": "",
+    "md5": null
+  },
+  "extended": {
+  }
+}
+ */
+
+    }
+}

test/random_access/src/test/java/seaweedfs/file/RandomeAccessFileTest.java

@@ -0,0 +1,70 @@
+package seaweedfs.file;
+
+import org.junit.Assert;
+import org.junit.Test;
+
+import java.io.File;
+import java.io.IOException;
+import java.io.RandomAccessFile;
+import java.nio.ByteBuffer;
+import java.util.Random;
+
+public class RandomeAccessFileTest {
+
+    @Test
+    public void testRandomWriteAndRead() throws IOException {
+
+        File f = new File(MmapFileTest.dir, "mmap_file.txt");
+
+        RandomAccessFile af = new RandomAccessFile(f, "rw");
+        af.setLength(0);
+        af.close();
+
+        Random r = new Random();
+
+        int maxLength = 5000;
+
+        byte[] data = new byte[maxLength];
+        byte[] readData = new byte[maxLength];
+
+        for (int i = 4096; i < maxLength; i++) {
+
+            RandomAccessFile raf = new RandomAccessFile(f, "rw");
+            long fileSize = raf.length();
+
+            raf.readFully(readData, 0, (int)fileSize);
+
+            for (int x=0;x<fileSize;x++){
+                Assert.assertEquals(data[x], readData[x]);
+            }
+
+            int start = r.nextInt(i);
+            int stop = r.nextInt(i);
+            if (start > stop) {
+                int t = stop;
+                stop = start;
+                start = t;
+            }
+            if (stop > fileSize) {
+                fileSize = stop;
+                raf.setLength(fileSize);
+            }
+
+            randomize(r, data, start, stop);
+            raf.seek(start);
+            raf.write(data, start, stop-start);
+
+            raf.close();
+        }
+
+    }
+
+    private static void randomize(Random r, byte[] bytes, int start, int stop) {
+        for (int i = start; i < stop; i++) {
+            int rnd = r.nextInt();
+            bytes[i] = (byte) rnd;
+        }
+    }
+
+
+}

test/s3/basic/basic_test.go

@@ -61,7 +61,7 @@ func TestCreateBucket(t *testing.T) {
 
 }
 
-func TestListBuckets(t *testing.T) {
+func TestPutObject(t *testing.T) {
 
 	input := &s3.PutObjectInput{
 		ACL:    aws.String("authenticated-read"),
@@ -89,7 +89,7 @@ func TestListBuckets(t *testing.T) {
 
 }
 
-func TestPutObject(t *testing.T) {
+func TestListBucket(t *testing.T) {
 
 	result, err := svc.ListBuckets(nil)
 	if err != nil {
@@ -105,6 +105,23 @@ func TestPutObject(t *testing.T) {
 
 }
 
+func TestListObjectV2(t *testing.T) {
+
+	listObj, err := svc.ListObjectsV2(&s3.ListObjectsV2Input{
+		Bucket: aws.String(Bucket),
+		Prefix: aws.String("foo"),
+		Delimiter: aws.String("/"),
+	})
+	if err != nil {
+		exitErrorf("Unable to list objects, %v", err)
+	}
+	for _, content := range listObj.Contents {
+		fmt.Println(aws.StringValue(content.Key))
+	}
+	fmt.Printf("list: %s\n", listObj)
+
+}
+
 func exitErrorf(msg string, args ...interface{}) {
 	fmt.Fprintf(os.Stderr, msg+"\n", args...)
 	os.Exit(1)

test/s3/multipart/aws_upload.go

@@ -0,0 +1,175 @@
+package main
+
+// copied from https://github.com/apoorvam/aws-s3-multipart-upload
+
+import (
+	"bytes"
+	"flag"
+	"fmt"
+	"net/http"
+	"os"
+
+	"github.com/aws/aws-sdk-go/aws"
+	"github.com/aws/aws-sdk-go/aws/awserr"
+	"github.com/aws/aws-sdk-go/aws/credentials"
+	"github.com/aws/aws-sdk-go/aws/session"
+	"github.com/aws/aws-sdk-go/service/s3"
+)
+
+const (
+	maxPartSize        = int64(5 * 1024 * 1024)
+	maxRetries         = 3
+	awsAccessKeyID     = "Your access key"
+	awsSecretAccessKey = "Your secret key"
+	awsBucketRegion    = "S3 bucket region"
+	awsBucketName      = "newBucket"
+)
+
+var (
+	filename = flag.String("f", "", "the file name")
+)
+
+func main() {
+	flag.Parse()
+
+	creds := credentials.NewStaticCredentials(awsAccessKeyID, awsSecretAccessKey, "")
+	_, err := creds.Get()
+	if err != nil {
+		fmt.Printf("bad credentials: %s", err)
+	}
+	cfg := aws.NewConfig().WithRegion(awsBucketRegion).WithCredentials(creds).WithDisableSSL(true).WithEndpoint("localhost:8333")
+	svc := s3.New(session.New(), cfg)
+
+	file, err := os.Open(*filename)
+	if err != nil {
+		fmt.Printf("err opening file: %s", err)
+		return
+	}
+	defer file.Close()
+	fileInfo, _ := file.Stat()
+	size := fileInfo.Size()
+	buffer := make([]byte, size)
+	fileType := http.DetectContentType(buffer)
+	file.Read(buffer)
+
+	path := "/media/" + file.Name()
+	input := &s3.CreateMultipartUploadInput{
+		Bucket:      aws.String(awsBucketName),
+		Key:         aws.String(path),
+		ContentType: aws.String(fileType),
+	}
+
+	resp, err := svc.CreateMultipartUpload(input)
+	if err != nil {
+		fmt.Println(err.Error())
+		return
+	}
+	fmt.Println("Created multipart upload request")
+
+	var curr, partLength int64
+	var remaining = size
+	var completedParts []*s3.CompletedPart
+	partNumber := 1
+	for curr = 0; remaining != 0; curr += partLength {
+		if remaining < maxPartSize {
+			partLength = remaining
+		} else {
+			partLength = maxPartSize
+		}
+		completedPart, err := uploadPart(svc, resp, buffer[curr:curr+partLength], partNumber)
+		if err != nil {
+			fmt.Println(err.Error())
+			err := abortMultipartUpload(svc, resp)
+			if err != nil {
+				fmt.Println(err.Error())
+			}
+			return
+		}
+		remaining -= partLength
+		partNumber++
+		completedParts = append(completedParts, completedPart)
+	}
+
+	// list parts
+	parts, err := svc.ListParts(&s3.ListPartsInput{
+		Bucket:           input.Bucket,
+		Key:              input.Key,
+		MaxParts:         nil,
+		PartNumberMarker: nil,
+		RequestPayer:     nil,
+		UploadId:         resp.UploadId,
+	})
+	if err != nil {
+		fmt.Println(err.Error())
+		return
+	}
+	fmt.Printf("list parts: %d\n", len(parts.Parts))
+	for i, part := range parts.Parts {
+		fmt.Printf("part %d: %v\n", i, part)
+	}
+
+
+	completeResponse, err := completeMultipartUpload(svc, resp, completedParts)
+	if err != nil {
+		fmt.Println(err.Error())
+		return
+	}
+
+	fmt.Printf("Successfully uploaded file: %s\n", completeResponse.String())
+}
+
+func completeMultipartUpload(svc *s3.S3, resp *s3.CreateMultipartUploadOutput, completedParts []*s3.CompletedPart) (*s3.CompleteMultipartUploadOutput, error) {
+	completeInput := &s3.CompleteMultipartUploadInput{
+		Bucket:   resp.Bucket,
+		Key:      resp.Key,
+		UploadId: resp.UploadId,
+		MultipartUpload: &s3.CompletedMultipartUpload{
+			Parts: completedParts,
+		},
+	}
+	return svc.CompleteMultipartUpload(completeInput)
+}
+
+func uploadPart(svc *s3.S3, resp *s3.CreateMultipartUploadOutput, fileBytes []byte, partNumber int) (*s3.CompletedPart, error) {
+	tryNum := 1
+	partInput := &s3.UploadPartInput{
+		Body:          bytes.NewReader(fileBytes),
+		Bucket:        resp.Bucket,
+		Key:           resp.Key,
+		PartNumber:    aws.Int64(int64(partNumber)),
+		UploadId:      resp.UploadId,
+		ContentLength: aws.Int64(int64(len(fileBytes))),
+	}
+
+	for tryNum <= maxRetries {
+		uploadResult, err := svc.UploadPart(partInput)
+		if err != nil {
+			if tryNum == maxRetries {
+				if aerr, ok := err.(awserr.Error); ok {
+					return nil, aerr
+				}
+				return nil, err
+			}
+			fmt.Printf("Retrying to upload part #%v\n", partNumber)
+			tryNum++
+		} else {
+			fmt.Printf("Uploaded part #%v\n", partNumber)
+			return &s3.CompletedPart{
+				ETag:       uploadResult.ETag,
+				PartNumber: aws.Int64(int64(partNumber)),
+			}, nil
+		}
+	}
+	return nil, nil
+}
+
+func abortMultipartUpload(svc *s3.S3, resp *s3.CreateMultipartUploadOutput) error {
+	fmt.Println("Aborting multipart upload for UploadId#" + *resp.UploadId)
+	abortInput := &s3.AbortMultipartUploadInput{
+		Bucket:   resp.Bucket,
+		Key:      resp.Key,
+		UploadId: resp.UploadId,
+	}
+	_, err := svc.AbortMultipartUpload(abortInput)
+	return err
+}

unmaintained/diff_volume_servers/diff_volume_servers.go

@@ -118,13 +118,15 @@ const (
 
 type needleState struct {
 	state uint8
-	size  uint32
+	size  types.Size
 }
 
 func getVolumeFiles(v uint32, addr string) (map[types.NeedleId]needleState, int64, error) {
 	var idxFile *bytes.Reader
 	err := operation.WithVolumeServerClient(addr, grpcDialOption, func(vs volume_server_pb.VolumeServerClient) error {
-		copyFileClient, err := vs.CopyFile(context.Background(), &volume_server_pb.CopyFileRequest{
+		ctx, cancel := context.WithCancel(context.Background())
+		defer cancel()
+		copyFileClient, err := vs.CopyFile(ctx, &volume_server_pb.CopyFileRequest{
 			VolumeId:           v,
 			Ext:                ".idx",
 			CompactionRevision: math.MaxUint32,
@@ -154,8 +156,8 @@ func getVolumeFiles(v uint32, addr string) (map[types.NeedleId]needleState, int6
 
 	var maxOffset int64
 	files := map[types.NeedleId]needleState{}
-	err = idx.WalkIndexFile(idxFile, func(key types.NeedleId, offset types.Offset, size uint32) error {
-		if offset.IsZero() || size == types.TombstoneFileSize {
+	err = idx.WalkIndexFile(idxFile, func(key types.NeedleId, offset types.Offset, size types.Size) error {
+		if offset.IsZero() || size.IsDeleted() {
 			files[key] = needleState{
 				state: stateDeleted,
 				size:  size,

unmaintained/fix_dat/fix_dat.go

@@ -98,7 +98,7 @@ func iterateEntries(datBackend backend.BackendStorageFile, idxFile *os.File, vis
 		// parse index file entry
 		key := util.BytesToUint64(bytes[0:8])
 		offsetFromIndex := util.BytesToUint32(bytes[8:12])
-		sizeFromIndex := util.BytesToUint32(bytes[12:16])
+		sizeFromIndex := types.BytesToSize(bytes[12:16])
 		count, _ = idxFile.ReadAt(bytes, readerOffset)
 		readerOffset += int64(count)
 
@@ -123,7 +123,7 @@ func iterateEntries(datBackend backend.BackendStorageFile, idxFile *os.File, vis
 			}
 		}()
 
-		if n.Size <= n.DataSize {
+		if n.Size <= types.Size(n.DataSize) {
 			continue
 		}
 		visitNeedle(n, offset)

unmaintained/see_dat/see_dat_gzip.go

@@ -1,83 +0,0 @@
-package main
-
-import (
-	"bytes"
-	"compress/gzip"
-	"crypto/md5"
-	"flag"
-	"io"
-	"io/ioutil"
-	"net/http"
-	"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"
-	"github.com/chrislusf/seaweedfs/weed/util"
-)
-
-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
-}
-
-var (
-	files     = int64(0)
-	filebytes = int64(0)
-	diffbytes = int64(0)
-)
-
-func Compresssion(data []byte) float64 {
-	if len(data) <= 128 {
-		return 100.0
-	}
-	compressed, _ := util.GzipData(data[0:128])
-	return float64(len(compressed)*10) / 1280.0
-}
-
-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).Info("----------------------------------------------------------------------------------")
-	glog.V(0).Infof("%d,%s%x offset %d size %d(%s) cookie %x appendedAt %v  hasmime[%t] mime[%s] (len: %d)",
-		*volumeId, n.Id, n.Cookie, offset, n.Size, util.BytesToHumanReadable(uint64(n.Size)), n.Cookie, t, n.HasMime(), string(n.Mime), len(n.Mime))
-	r, err := gzip.NewReader(bytes.NewReader(n.Data))
-	if err == nil {
-		buf := bytes.Buffer{}
-		h := md5.New()
-		c, _ := io.Copy(&buf, r)
-		d := buf.Bytes()
-		io.Copy(h, bytes.NewReader(d))
-		diff := (int64(n.DataSize) - int64(c))
-		diffbytes += diff
-		glog.V(0).Infof("was gzip! stored_size: %d orig_size: %d diff: %d(%d) mime:%s compression-of-128: %.2f md5: %x", n.DataSize, c, diff, diffbytes, http.DetectContentType(d), Compresssion(d), h.Sum(nil))
-	} else {
-		glog.V(0).Infof("no gzip!")
-	}
-	return nil
-}
-
-var (
-	_                = ioutil.ReadAll
-	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 main() {
-	flag.Parse()
-	vid := needle.VolumeId(*volumeId)
-	glog.V(0).Info("Starting")
-	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

@@ -36,7 +36,7 @@ func main() {
 	}
 	defer indexFile.Close()
 
-	idx.WalkIndexFile(indexFile, func(key types.NeedleId, offset types.Offset, size uint32) error {
+	idx.WalkIndexFile(indexFile, func(key types.NeedleId, offset types.Offset, size types.Size) error {
 		fmt.Printf("key:%v offset:%v size:%v(%v)\n", key, offset, size, util.BytesToHumanReadable(uint64(size)))
 		return nil
 	})

unmaintained/see_log_entry/see_log_entry.go

@@ -9,13 +9,13 @@ import (
 
 	"github.com/golang/protobuf/proto"
 
-	"github.com/chrislusf/seaweedfs/weed/filer2"
+	"github.com/chrislusf/seaweedfs/weed/filer"
 	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
 	"github.com/chrislusf/seaweedfs/weed/util"
 )
 
 var (
-	logdataFile = flag.String("logdata", "", "log data file saved under "+ filer2.SystemLogDir)
+	logdataFile = flag.String("logdata", "", "log data file saved under "+ filer.SystemLogDir)
 )
 
 func main() {

weed/Makefile

@@ -0,0 +1,27 @@
+BINARY = weed
+
+SOURCE_DIR = .
+
+all: debug_mount
+
+.PHONY : clean debug_mount
+
+clean:
+	go clean $(SOURCE_DIR)
+	rm -f $(BINARY)
+
+debug_shell:
+	go build -gcflags="all=-N -l"
+	dlv --listen=:2345 --headless=true --api-version=2 --accept-multiclient exec weed -- shell
+
+debug_mount:
+	go build -gcflags="all=-N -l"
+	dlv --listen=:2345 --headless=true --api-version=2 --accept-multiclient exec weed -- mount -dir=~/tmp/mm
+
+debug_server:
+	go build -gcflags="all=-N -l"
+	dlv --listen=:2345 --headless=true --api-version=2 --accept-multiclient exec weed -- server -dir=/Volumes/mobile_disk/99 -filer -volume.port=8343 -s3 -volume.max=0
+
+debug_volume:
+	go build -gcflags="all=-N -l"
+	dlv --listen=:2345 --headless=true --api-version=2 --accept-multiclient exec weed -- volume -dir=/Volumes/mobile_disk/100 -port 8564 -max=30 -preStopSeconds=2

weed/command/command.go

@@ -16,6 +16,7 @@ var Commands = []*Command{
 	cmdExport,
 	cmdFiler,
 	cmdFilerReplicate,
+	cmdFilerSynchronize,
 	cmdFix,
 	cmdMaster,
 	cmdMount,

weed/command/download.go

@@ -4,6 +4,7 @@ import (
 	"fmt"
 	"io"
 	"io/ioutil"
+	"net/http"
 	"os"
 	"path"
 	"strings"
@@ -59,7 +60,7 @@ func downloadToFile(server, fileId, saveDir string) error {
 	if err != nil {
 		return err
 	}
-	defer rc.Close()
+	defer util.CloseResponse(rc)
 	if filename == "" {
 		filename = fileId
 	}
@@ -71,12 +72,11 @@ func downloadToFile(server, fileId, saveDir string) error {
 	}
 	f, err := os.OpenFile(path.Join(saveDir, filename), os.O_WRONLY|os.O_CREATE|os.O_TRUNC, os.ModePerm)
 	if err != nil {
-		io.Copy(ioutil.Discard, rc)
 		return err
 	}
 	defer f.Close()
 	if isFileList {
-		content, err := ioutil.ReadAll(rc)
+		content, err := ioutil.ReadAll(rc.Body)
 		if err != nil {
 			return err
 		}
@@ -95,7 +95,7 @@ func downloadToFile(server, fileId, saveDir string) error {
 			}
 		}
 	} else {
-		if _, err = io.Copy(f, rc); err != nil {
+		if _, err = io.Copy(f, rc.Body); err != nil {
 			return err
 		}
 
@@ -108,12 +108,12 @@ func fetchContent(server string, fileId string) (filename string, content []byte
 	if lookupError != nil {
 		return "", nil, lookupError
 	}
-	var rc io.ReadCloser
+	var rc *http.Response
 	if filename, _, rc, e = util.DownloadFile(fileUrl); e != nil {
 		return "", nil, e
 	}
-	content, e = ioutil.ReadAll(rc)
-	rc.Close()
+	defer util.CloseResponse(rc)
+	content, e = ioutil.ReadAll(rc.Body)
 	return
 }
 

weed/command/export.go

@@ -23,7 +23,7 @@ import (
 )
 
 const (
-	defaultFnFormat = `{{.Mime}}/{{.Id}}:{{.Name}}`
+	defaultFnFormat = `{{.Id}}_{{.Name}}{{.Ext}}`
 	timeFormat      = "2006-01-02T15:04:05"
 )
 
@@ -56,7 +56,7 @@ func init() {
 
 var (
 	output      = cmdExport.Flag.String("o", "", "output tar file name, must ends with .tar, or just a \"-\" for stdout")
-	format      = cmdExport.Flag.String("fileNameFormat", defaultFnFormat, "filename formatted with {{.Mime}} {{.Id}} {{.Name}} {{.Ext}}")
+	format      = cmdExport.Flag.String("fileNameFormat", defaultFnFormat, "filename formatted with {{.Id}} {{.Name}} {{.Ext}}")
 	newer       = cmdExport.Flag.String("newer", "", "export only files newer than this time, default is all files. Must be specified in RFC3339 without timezone, e.g. 2006-01-02T15:04:05")
 	showDeleted = cmdExport.Flag.Bool("deleted", false, "export deleted files. only applies if -o is not specified")
 	limit       = cmdExport.Flag.Int("limit", 0, "only show first n entries if specified")
@@ -70,13 +70,13 @@ var (
 	localLocation, _             = time.LoadLocation("Local")
 )
 
-func printNeedle(vid needle.VolumeId, n *needle.Needle, version needle.Version, deleted bool) {
+func printNeedle(vid needle.VolumeId, n *needle.Needle, version needle.Version, deleted bool, offset int64, onDiskSize int64) {
 	key := needle.NewFileIdFromNeedle(vid, n).String()
-	size := n.DataSize
+	size := int32(n.DataSize)
 	if version == needle.Version1 {
-		size = n.Size
+		size = int32(n.Size)
 	}
-	fmt.Printf("%s\t%s\t%d\t%t\t%s\t%s\t%s\t%t\n",
+	fmt.Printf("%s\t%s\t%d\t%t\t%s\t%s\t%s\t%t\t%d\t%d\n",
 		key,
 		n.Name,
 		size,
@@ -85,6 +85,8 @@ func printNeedle(vid needle.VolumeId, n *needle.Needle, version needle.Version,
 		n.LastModifiedString(),
 		n.Ttl.String(),
 		deleted,
+		offset,
+		offset+onDiskSize,
 	)
 }
 
@@ -111,7 +113,7 @@ func (scanner *VolumeFileScanner4Export) VisitNeedle(n *needle.Needle, offset in
 	nv, ok := needleMap.Get(n.Id)
 	glog.V(3).Infof("key %d offset %d size %d disk_size %d compressed %v ok %v nv %+v",
 		n.Id, offset, n.Size, n.DiskSize(scanner.version), n.IsCompressed(), ok, nv)
-	if ok && nv.Size > 0 && nv.Size != types.TombstoneFileSize && nv.Offset.ToAcutalOffset() == offset {
+	if *showDeleted && n.Size > 0 || ok && nv.Size.IsValid() && 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)
@@ -124,17 +126,17 @@ func (scanner *VolumeFileScanner4Export) VisitNeedle(n *needle.Needle, offset in
 		if tarOutputFile != nil {
 			return writeFile(vid, n)
 		} else {
-			printNeedle(vid, n, scanner.version, false)
+			printNeedle(vid, n, scanner.version, false, offset, n.DiskSize(scanner.version))
 			return nil
 		}
 	}
 	if !ok {
 		if *showDeleted && tarOutputFile == nil {
 			if n.DataSize > 0 {
-				printNeedle(vid, n, scanner.version, true)
+				printNeedle(vid, n, scanner.version, true, offset, n.DiskSize(scanner.version))
 			} else {
 				n.Name = []byte("*tombstone")
-				printNeedle(vid, n, scanner.version, true)
+				printNeedle(vid, n, scanner.version, true, offset, n.DiskSize(scanner.version))
 			}
 		}
 		glog.V(2).Infof("This seems deleted %d size %d", n.Id, n.Size)
@@ -208,7 +210,7 @@ func runExport(cmd *Command, args []string) bool {
 	}
 
 	if tarOutputFile == nil {
-		fmt.Printf("key\tname\tsize\tgzip\tmime\tmodified\tttl\tdeleted\n")
+		fmt.Printf("key\tname\tsize\tgzip\tmime\tmodified\tttl\tdeleted\tstart\tstop\n")
 	}
 
 	err = storage.ScanVolumeFile(util.ResolvePath(*export.dir), *export.collection, vid, storage.NeedleMapInMemory, volumeFileScanner)

weed/command/filer.go

@@ -1,6 +1,7 @@
 package command
 
 import (
+	"fmt"
 	"net/http"
 	"strconv"
 	"strings"
@@ -13,11 +14,14 @@ import (
 	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
 	"github.com/chrislusf/seaweedfs/weed/security"
 	"github.com/chrislusf/seaweedfs/weed/server"
+	stats_collect "github.com/chrislusf/seaweedfs/weed/stats"
 	"github.com/chrislusf/seaweedfs/weed/util"
 )
 
 var (
-	f FilerOptions
+	f              FilerOptions
+	filerStartS3   *bool
+	filerS3Options S3Options
 )
 
 type FilerOptions struct {
@@ -36,6 +40,7 @@ type FilerOptions struct {
 	disableHttp             *bool
 	cipher                  *bool
 	peers                   *string
+	metricsHttpPort         *int
 
 	// default leveldb directory, used in "weed server" mode
 	defaultLevelDbDirectory *string
@@ -49,7 +54,7 @@ func init() {
 	f.bindIp = cmdFiler.Flag.String("ip.bind", "0.0.0.0", "ip address to bind to")
 	f.port = cmdFiler.Flag.Int("port", 8888, "filer server http listen port")
 	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.defaultReplicaPlacement = cmdFiler.Flag.String("defaultReplicaPlacement", "", "default replication type. If not specified, use master setting.")
 	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.dirListingLimit = cmdFiler.Flag.Int("dirListLimit", 100000, "limit sub dir listing size")
@@ -57,6 +62,15 @@ func init() {
 	f.disableHttp = cmdFiler.Flag.Bool("disableHttp", false, "disable http request, only gRpc operations are allowed")
 	f.cipher = cmdFiler.Flag.Bool("encryptVolumeData", false, "encrypt data on volume servers")
 	f.peers = cmdFiler.Flag.String("peers", "", "all filers sharing the same filer store in comma separated ip:port list")
+	f.metricsHttpPort = cmdFiler.Flag.Int("metricsPort", 0, "Prometheus metrics listen port")
+
+	// start s3 on filer
+	filerStartS3 = cmdFiler.Flag.Bool("s3", false, "whether to start S3 gateway")
+	filerS3Options.port = cmdFiler.Flag.Int("s3.port", 8333, "s3 server http listen port")
+	filerS3Options.domainName = cmdFiler.Flag.String("s3.domainName", "", "suffix of the host name, {bucket}.{domainName}")
+	filerS3Options.tlsPrivateKey = cmdFiler.Flag.String("s3.key.file", "", "path to the TLS private key file")
+	filerS3Options.tlsCertificate = cmdFiler.Flag.String("s3.cert.file", "", "path to the TLS certificate file")
+	filerS3Options.config = cmdFiler.Flag.String("s3.config", "", "path to the config file")
 }
 
 var cmdFiler = &Command{
@@ -84,6 +98,17 @@ func runFiler(cmd *Command, args []string) bool {
 
 	util.LoadConfiguration("security", false)
 
+	go stats_collect.StartMetricsServer(*f.metricsHttpPort)
+
+	if *filerStartS3 {
+		filerAddress := fmt.Sprintf("%s:%d", *f.ip, *f.port)
+		filerS3Options.filer = &filerAddress
+		go func() {
+			time.Sleep(2 * time.Second)
+			filerS3Options.startS3Server()
+		}()
+	}
+
 	f.startFiler()
 
 	return true
@@ -152,7 +177,7 @@ func (fo *FilerOptions) startFiler() {
 
 	// starting grpc server
 	grpcPort := *fo.port + 10000
-	grpcL, err := util.NewListener(":"+strconv.Itoa(grpcPort), 0)
+	grpcL, err := util.NewListener(*fo.bindIp+":"+strconv.Itoa(grpcPort), 0)
 	if err != nil {
 		glog.Fatalf("failed to listen on grpc port %d: %v", grpcPort, err)
 	}

weed/command/filer_copy.go

@@ -72,7 +72,7 @@ var cmdCopy = &Command{
 
   If "maxMB" is set to a positive number, files larger than it would be split into chunks.
 
-  `,
+`,
 }
 
 func runCopy(cmd *Command, args []string) bool {

weed/command/filer_sync.go

@@ -0,0 +1,337 @@
+package command
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"github.com/chrislusf/seaweedfs/weed/glog"
+	"github.com/chrislusf/seaweedfs/weed/pb"
+	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
+	"github.com/chrislusf/seaweedfs/weed/replication"
+	"github.com/chrislusf/seaweedfs/weed/replication/sink/filersink"
+	"github.com/chrislusf/seaweedfs/weed/replication/source"
+	"github.com/chrislusf/seaweedfs/weed/security"
+	"github.com/chrislusf/seaweedfs/weed/util"
+	"github.com/chrislusf/seaweedfs/weed/util/grace"
+	"google.golang.org/grpc"
+	"io"
+	"strings"
+	"time"
+)
+
+type SyncOptions struct {
+	isActivePassive *bool
+	filerA          *string
+	filerB          *string
+	aPath           *string
+	bPath           *string
+	aReplication    *string
+	bReplication    *string
+	aCollection     *string
+	bCollection     *string
+	aTtlSec         *int
+	bTtlSec         *int
+	aDebug          *bool
+	bDebug          *bool
+}
+
+var (
+	syncOptions    SyncOptions
+	syncCpuProfile *string
+	syncMemProfile *string
+)
+
+func init() {
+	cmdFilerSynchronize.Run = runFilerSynchronize // break init cycle
+	syncOptions.isActivePassive = cmdFilerSynchronize.Flag.Bool("isActivePassive", false, "one directional follow if true")
+	syncOptions.filerA = cmdFilerSynchronize.Flag.String("a", "", "filer A in one SeaweedFS cluster")
+	syncOptions.filerB = cmdFilerSynchronize.Flag.String("b", "", "filer B in the other SeaweedFS cluster")
+	syncOptions.aPath = cmdFilerSynchronize.Flag.String("a.path", "/", "directory to sync on filer A")
+	syncOptions.bPath = cmdFilerSynchronize.Flag.String("b.path", "/", "directory to sync on filer B")
+	syncOptions.aReplication = cmdFilerSynchronize.Flag.String("a.replication", "", "replication on filer A")
+	syncOptions.bReplication = cmdFilerSynchronize.Flag.String("b.replication", "", "replication on filer B")
+	syncOptions.aCollection = cmdFilerSynchronize.Flag.String("a.collection", "", "collection on filer A")
+	syncOptions.bCollection = cmdFilerSynchronize.Flag.String("b.collection", "", "collection on filer B")
+	syncOptions.aTtlSec = cmdFilerSynchronize.Flag.Int("a.ttlSec", 0, "ttl in seconds on filer A")
+	syncOptions.bTtlSec = cmdFilerSynchronize.Flag.Int("b.ttlSec", 0, "ttl in seconds on filer B")
+	syncOptions.aDebug = cmdFilerSynchronize.Flag.Bool("a.debug", false, "debug mode to print out filer A received files")
+	syncOptions.bDebug = cmdFilerSynchronize.Flag.Bool("b.debug", false, "debug mode to print out filer B received files")
+	syncCpuProfile = cmdFilerSynchronize.Flag.String("cpuprofile", "", "cpu profile output file")
+	syncMemProfile = cmdFilerSynchronize.Flag.String("memprofile", "", "memory profile output file")
+}
+
+var cmdFilerSynchronize = &Command{
+	UsageLine: "filer.sync -a=<oneFilerHost>:<oneFilerPort> -b=<otherFilerHost>:<otherFilerPort>",
+	Short:     "continuously synchronize between two active-active or active-passive SeaweedFS clusters",
+	Long: `continuously synchronize file changes between two active-active or active-passive filers
+
+	filer.sync listens on filer notifications. If any file is updated, it will fetch the updated content,
+	and write to the other destination. Different from filer.replicate:
+
+	* filer.sync only works between two filers.
+	* filer.sync does not need any special message queue setup.
+	* filer.sync supports both active-active and active-passive modes.
+	
+	If restarted, the synchronization will resume from the previous checkpoints, persisted every minute.
+	A fresh sync will start from the earliest metadata logs.
+
+`,
+}
+
+func runFilerSynchronize(cmd *Command, args []string) bool {
+
+	grpcDialOption := security.LoadClientTLS(util.GetViper(), "grpc.client")
+
+	grace.SetupProfiling(*syncCpuProfile, *syncMemProfile)
+
+	go func() {
+		for {
+			err := doSubscribeFilerMetaChanges(grpcDialOption, *syncOptions.filerA, *syncOptions.aPath, *syncOptions.filerB,
+				*syncOptions.bPath, *syncOptions.bReplication, *syncOptions.bCollection, *syncOptions.bTtlSec, *syncOptions.bDebug)
+			if err != nil {
+				glog.Errorf("sync from %s to %s: %v", *syncOptions.filerA, *syncOptions.filerB, err)
+				time.Sleep(1747 * time.Millisecond)
+			}
+		}
+	}()
+
+	if !*syncOptions.isActivePassive {
+		go func() {
+			for {
+				err := doSubscribeFilerMetaChanges(grpcDialOption, *syncOptions.filerB, *syncOptions.bPath, *syncOptions.filerA,
+					*syncOptions.aPath, *syncOptions.aReplication, *syncOptions.aCollection, *syncOptions.aTtlSec, *syncOptions.aDebug)
+				if err != nil {
+					glog.Errorf("sync from %s to %s: %v", *syncOptions.filerB, *syncOptions.filerA, err)
+					time.Sleep(2147 * time.Millisecond)
+				}
+			}
+		}()
+	}
+
+	select {}
+
+	return true
+}
+
+func doSubscribeFilerMetaChanges(grpcDialOption grpc.DialOption, sourceFiler, sourcePath, targetFiler, targetPath string,
+	replicationStr, collection string, ttlSec int, debug bool) error {
+
+	// read source filer signature
+	sourceFilerSignature, sourceErr := replication.ReadFilerSignature(grpcDialOption, sourceFiler)
+	if sourceErr != nil {
+		return sourceErr
+	}
+	// read target filer signature
+	targetFilerSignature, targetErr := replication.ReadFilerSignature(grpcDialOption, targetFiler)
+	if targetErr != nil {
+		return targetErr
+	}
+
+	// if first time, start from now
+	// if has previously synced, resume from that point of time
+	sourceFilerOffsetTsNs, err := readSyncOffset(grpcDialOption, targetFiler, sourceFilerSignature)
+	if err != nil {
+		return err
+	}
+
+	glog.V(0).Infof("start sync %s(%d) => %s(%d) from %v(%d)", sourceFiler, sourceFilerSignature, targetFiler, targetFilerSignature, time.Unix(0, sourceFilerOffsetTsNs), sourceFilerOffsetTsNs)
+
+	// create filer sink
+	filerSource := &source.FilerSource{}
+	filerSource.DoInitialize(pb.ServerToGrpcAddress(sourceFiler), sourcePath)
+	filerSink := &filersink.FilerSink{}
+	filerSink.DoInitialize(pb.ServerToGrpcAddress(targetFiler), targetPath, replicationStr, collection, ttlSec, grpcDialOption)
+	filerSink.SetSourceFiler(filerSource)
+
+	processEventFn := func(resp *filer_pb.SubscribeMetadataResponse) error {
+		message := resp.EventNotification
+
+		var sourceOldKey, sourceNewKey util.FullPath
+		if message.OldEntry != nil {
+			sourceOldKey = util.FullPath(resp.Directory).Child(message.OldEntry.Name)
+		}
+		if message.NewEntry != nil {
+			sourceNewKey = util.FullPath(message.NewParentPath).Child(message.NewEntry.Name)
+		}
+
+		for _, sig := range message.Signatures {
+			if sig == targetFilerSignature && targetFilerSignature != 0 {
+				fmt.Printf("%s skipping %s change to %v\n", targetFiler, sourceFiler, message)
+				return nil
+			}
+		}
+		if debug {
+			fmt.Printf("%s check %s change %s,%s sig %v, target sig: %v\n", targetFiler, sourceFiler, sourceOldKey, sourceNewKey, message.Signatures, targetFilerSignature)
+		}
+
+		if !strings.HasPrefix(resp.Directory, sourcePath) {
+			return nil
+		}
+
+		// handle deletions
+		if message.OldEntry != nil && message.NewEntry == nil {
+			if !strings.HasPrefix(string(sourceOldKey), sourcePath) {
+				return nil
+			}
+			key := util.Join(targetPath, string(sourceOldKey)[len(sourcePath):])
+			return filerSink.DeleteEntry(key, message.OldEntry.IsDirectory, message.DeleteChunks, message.Signatures)
+		}
+
+		// handle new entries
+		if message.OldEntry == nil && message.NewEntry != nil {
+			if !strings.HasPrefix(string(sourceNewKey), sourcePath) {
+				return nil
+			}
+			key := util.Join(targetPath, string(sourceNewKey)[len(sourcePath):])
+			return filerSink.CreateEntry(key, message.NewEntry, message.Signatures)
+		}
+
+		// this is something special?
+		if message.OldEntry == nil && message.NewEntry == nil {
+			return nil
+		}
+
+		// handle updates
+		if strings.HasPrefix(string(sourceOldKey), sourcePath) {
+			// old key is in the watched directory
+			if strings.HasPrefix(string(sourceNewKey), sourcePath) {
+				// new key is also in the watched directory
+				oldKey := util.Join(targetPath, string(sourceOldKey)[len(sourcePath):])
+				message.NewParentPath = util.Join(targetPath, message.NewParentPath[len(sourcePath):])
+				foundExisting, err := filerSink.UpdateEntry(string(oldKey), message.OldEntry, message.NewParentPath, message.NewEntry, message.DeleteChunks, message.Signatures)
+				if foundExisting {
+					return err
+				}
+
+				// not able to find old entry
+				if err = filerSink.DeleteEntry(string(oldKey), message.OldEntry.IsDirectory, false, message.Signatures); err != nil {
+					return fmt.Errorf("delete old entry %v: %v", oldKey, err)
+				}
+
+				// create the new entry
+				newKey := util.Join(targetPath, string(sourceNewKey)[len(sourcePath):])
+				return filerSink.CreateEntry(newKey, message.NewEntry, message.Signatures)
+
+			} else {
+				// new key is outside of the watched directory
+				key := util.Join(targetPath, string(sourceOldKey)[len(sourcePath):])
+				return filerSink.DeleteEntry(key, message.OldEntry.IsDirectory, message.DeleteChunks, message.Signatures)
+			}
+		} else {
+			// old key is outside of the watched directory
+			if strings.HasPrefix(string(sourceNewKey), sourcePath) {
+				// new key is in the watched directory
+				key := util.Join(targetPath, string(sourceNewKey)[len(sourcePath):])
+				return filerSink.CreateEntry(key, message.NewEntry, message.Signatures)
+			} else {
+				// new key is also outside of the watched directory
+				// skip
+			}
+		}
+
+		return nil
+	}
+
+	return pb.WithFilerClient(sourceFiler, grpcDialOption, func(client filer_pb.SeaweedFilerClient) error {
+
+		ctx, cancel := context.WithCancel(context.Background())
+		defer cancel()
+
+		stream, err := client.SubscribeMetadata(ctx, &filer_pb.SubscribeMetadataRequest{
+			ClientName: "syncTo_" + targetFiler,
+			PathPrefix: sourcePath,
+			SinceNs:    sourceFilerOffsetTsNs,
+			Signature:  targetFilerSignature,
+		})
+		if err != nil {
+			return fmt.Errorf("listen: %v", err)
+		}
+
+		var counter int64
+		var lastWriteTime time.Time
+		for {
+			resp, listenErr := stream.Recv()
+			if listenErr == io.EOF {
+				return nil
+			}
+			if listenErr != nil {
+				return listenErr
+			}
+
+			if err := processEventFn(resp); err != nil {
+				return err
+			}
+
+			counter++
+			if lastWriteTime.Add(3 * time.Second).Before(time.Now()) {
+				glog.V(0).Infof("sync %s => %s progressed to %v %0.2f/sec", sourceFiler, targetFiler, time.Unix(0, resp.TsNs), float64(counter)/float64(3))
+				counter = 0
+				lastWriteTime = time.Now()
+				if err := writeSyncOffset(grpcDialOption, targetFiler, sourceFilerSignature, resp.TsNs); err != nil {
+					return err
+				}
+			}
+
+		}
+
+	})
+
+}
+
+const (
+	SyncKeyPrefix = "sync."
+)
+
+func readSyncOffset(grpcDialOption grpc.DialOption, filer string, filerSignature int32) (lastOffsetTsNs int64, readErr error) {
+
+	readErr = pb.WithFilerClient(filer, grpcDialOption, func(client filer_pb.SeaweedFilerClient) error {
+		syncKey := []byte(SyncKeyPrefix + "____")
+		util.Uint32toBytes(syncKey[len(SyncKeyPrefix):len(SyncKeyPrefix)+4], uint32(filerSignature))
+
+		resp, err := client.KvGet(context.Background(), &filer_pb.KvGetRequest{Key: syncKey})
+		if err != nil {
+			return err
+		}
+
+		if len(resp.Error) != 0 {
+			return errors.New(resp.Error)
+		}
+		if len(resp.Value) < 8 {
+			return nil
+		}
+
+		lastOffsetTsNs = int64(util.BytesToUint64(resp.Value))
+
+		return nil
+	})
+
+	return
+
+}
+
+func writeSyncOffset(grpcDialOption grpc.DialOption, filer string, filerSignature int32, offsetTsNs int64) error {
+	return pb.WithFilerClient(filer, grpcDialOption, func(client filer_pb.SeaweedFilerClient) error {
+
+		syncKey := []byte(SyncKeyPrefix + "____")
+		util.Uint32toBytes(syncKey[len(SyncKeyPrefix):len(SyncKeyPrefix)+4], uint32(filerSignature))
+
+		valueBuf := make([]byte, 8)
+		util.Uint64toBytes(valueBuf, uint64(offsetTsNs))
+
+		resp, err := client.KvPut(context.Background(), &filer_pb.KvPutRequest{
+			Key:   syncKey,
+			Value: valueBuf,
+		})
+		if err != nil {
+			return err
+		}
+
+		if len(resp.Error) != 0 {
+			return errors.New(resp.Error)
+		}
+
+		return nil
+
+	})
+
+}

weed/command/fix.go

@@ -48,7 +48,7 @@ func (scanner *VolumeFileScanner4Fix) ReadNeedleBody() bool {
 
 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 compressed %v", n.Id, offset, n.Size, n.DiskSize(scanner.version), n.IsCompressed())
-	if n.Size > 0 && n.Size != types.TombstoneFileSize {
+	if n.Size.IsValid() {
 		pe := scanner.nm.Set(n.Id, types.ToOffset(offset), n.Size)
 		glog.V(2).Infof("saved %d with error %v", n.Size, pe)
 	} else {

weed/command/master.go

@@ -57,7 +57,7 @@ func init() {
 	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.metricsAddress = cmdMaster.Flag.String("metrics.address", "", "Prometheus gateway address <host>:<port>")
 	m.metricsIntervalSec = cmdMaster.Flag.Int("metrics.intervalSeconds", 15, "Prometheus push interval in seconds")
 }
 

weed/command/mount.go

@@ -20,6 +20,8 @@ type MountOptions struct {
 	umaskString                 *string
 	nonempty                    *bool
 	outsideContainerClusterMode *bool
+	uidMap                      *string
+	gidMap                      *string
 }
 
 var (
@@ -47,6 +49,8 @@ func init() {
 	mountCpuProfile = cmdMount.Flag.String("cpuprofile", "", "cpu profile output file")
 	mountMemProfile = cmdMount.Flag.String("memprofile", "", "memory profile output file")
 	mountOptions.outsideContainerClusterMode = cmdMount.Flag.Bool("outsideContainerClusterMode", false, "allows other users to access the file system")
+	mountOptions.uidMap = cmdMount.Flag.String("map.uid", "", "map local uid to uid on filer, comma-separated <local_uid>:<filer_uid>")
+	mountOptions.gidMap = cmdMount.Flag.String("map.gid", "", "map local gid to gid on filer, comma-separated <local_gid>:<filer_gid>")
 }
 
 var cmdMount = &Command{

weed/command/mount_std.go

@@ -5,8 +5,8 @@ package command
 import (
 	"context"
 	"fmt"
+	"github.com/chrislusf/seaweedfs/weed/filesys/meta_cache"
 	"os"
-	"os/user"
 	"path"
 	"runtime"
 	"strconv"
@@ -86,33 +86,17 @@ func RunMount(option *MountOptions, umask os.FileMode) bool {
 
 	fuse.Unmount(dir)
 
-	uid, gid := uint32(0), uint32(0)
-
 	// detect mount folder mode
 	if *option.dirAutoCreate {
-		os.MkdirAll(dir, 0755)
+		os.MkdirAll(dir, os.FileMode(0777)&^umask)
 	}
-	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())
-	} else {
-		fmt.Printf("can not stat %s\n", dir)
-		return false
-	}
 
-	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)
-		}
+	// mapping uid, gid
+	uidGidMapper, err := meta_cache.NewUidGidMapper(*option.uidMap, *option.gidMap)
+	if err != nil {
+		fmt.Printf("failed to parse %s %s: %v\n", *option.uidMap, *option.gidMap, err)
+		return false
 	}
 
 	// Ensure target mount point availability
@@ -166,14 +150,12 @@ func RunMount(option *MountOptions, umask os.FileMode) bool {
 		CacheSizeMB:                 *option.cacheSizeMB,
 		DataCenter:                  *option.dataCenter,
 		EntryCacheTtl:               3 * time.Second,
-		MountUid:                    uid,
-		MountGid:                    gid,
-		MountMode:                   mountMode,
 		MountCtime:                  fileInfo.ModTime(),
 		MountMtime:                  time.Now(),
 		Umask:                       umask,
 		OutsideContainerClusterMode: *mountOptions.outsideContainerClusterMode,
 		Cipher:                      cipher,
+		UidGidMapper:                uidGidMapper,
 	})
 
 	// mount

weed/command/s3.go

@@ -14,6 +14,7 @@ import (
 
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/s3api"
+	stats_collect "github.com/chrislusf/seaweedfs/weed/stats"
 	"github.com/chrislusf/seaweedfs/weed/util"
 )
 
@@ -22,12 +23,13 @@ var (
 )
 
 type S3Options struct {
-	filer          *string
-	port           *int
-	config         *string
-	domainName     *string
-	tlsPrivateKey  *string
-	tlsCertificate *string
+	filer           *string
+	port            *int
+	config          *string
+	domainName      *string
+	tlsPrivateKey   *string
+	tlsCertificate  *string
+	metricsHttpPort *int
 }
 
 func init() {
@@ -38,6 +40,7 @@ func init() {
 	s3StandaloneOptions.config = cmdS3.Flag.String("config", "", "path to the config file")
 	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")
+	s3StandaloneOptions.metricsHttpPort = cmdS3.Flag.Int("metricsPort", 0, "Prometheus metrics listen port")
 }
 
 var cmdS3 = &Command{
@@ -112,6 +115,8 @@ func runS3(cmd *Command, args []string) bool {
 
 	util.LoadConfiguration("security", false)
 
+	go stats_collect.StartMetricsServer(*s3StandaloneOptions.metricsHttpPort)
+
 	return s3StandaloneOptions.startS3Server()
 
 }
@@ -128,6 +133,10 @@ func (s3opt *S3Options) startS3Server() bool {
 
 	grpcDialOption := security.LoadClientTLS(util.GetViper(), "grpc.client")
 
+	// metrics read from the filer
+	var metricsAddress string
+	var metricsIntervalSec int
+
 	for {
 		err = pb.WithGrpcFilerClient(filerGrpcAddress, grpcDialOption, func(client filer_pb.SeaweedFilerClient) error {
 			resp, err := client.GetFilerConfiguration(context.Background(), &filer_pb.GetFilerConfigurationRequest{})
@@ -135,6 +144,7 @@ func (s3opt *S3Options) startS3Server() bool {
 				return fmt.Errorf("get filer %s configuration: %v", filerGrpcAddress, err)
 			}
 			filerBucketsPath = resp.DirBuckets
+			metricsAddress, metricsIntervalSec = resp.MetricsAddress, int(resp.MetricsIntervalSec)
 			glog.V(0).Infof("S3 read filer buckets dir: %s", filerBucketsPath)
 			return nil
 		})
@@ -147,6 +157,8 @@ func (s3opt *S3Options) startS3Server() bool {
 		}
 	}
 
+	go stats_collect.LoopPushingMetric("s3", stats_collect.SourceName(uint32(*s3opt.port)), metricsAddress, metricsIntervalSec)
+
 	router := mux.NewRouter().SkipClean(true)
 
 	_, s3ApiServer_err := s3api.NewS3ApiServer(router, &s3api.S3ApiServerOption{

weed/command/scaffold.go

@@ -140,6 +140,8 @@ keyspace="seaweedfs"
 hosts=[
 	"localhost:9042",
 ]
+username=""
+password=""
 
 [redis2]
 enabled = false
@@ -173,6 +175,20 @@ enabled = false
 uri = "mongodb://localhost:27017"
 option_pool_size = 0
 database = "seaweedfs"
+
+[elastic7]
+enabled = false
+servers = [
+    "http://localhost1:9200",
+    "http://localhost2:9200",
+    "http://localhost3:9200",
+]
+username = ""
+password = ""
+sniff_enabled = false
+healthcheck_enabled = false
+# increase the value is recommend, be sure the value in Elastic is greater or equal here
+index.max_result_window = 10000
 `
 
 	NOTIFICATION_TOML_EXAMPLE = `
@@ -377,7 +393,7 @@ default = "localhost:8888"    # used by maintenance scripts if the scripts needs
 
 
 [master.sequencer]
-type = "memory"     # Choose [memory|etcd] type for storing the file id sequence
+type = "raft"     # Choose [raft|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"

weed/command/server.go

@@ -2,6 +2,7 @@ package command
 
 import (
 	"fmt"
+	stats_collect "github.com/chrislusf/seaweedfs/weed/stats"
 	"os"
 	"runtime"
 	"runtime/pprof"
@@ -56,6 +57,7 @@ var (
 	volumeDataFolders         = cmdServer.Flag.String("dir", os.TempDir(), "directories to store data files. dir[,dir]...")
 	volumeMaxDataVolumeCounts = cmdServer.Flag.String("volume.max", "8", "maximum numbers of volumes, count[,count]... If set to zero, the limit will be auto configured.")
 	volumeMinFreeSpacePercent = cmdServer.Flag.String("volume.minFreeSpacePercent", "1", "minimum free disk space (default to 1%). Low disk space will mark all volumes as ReadOnly.")
+	serverMetricsHttpPort     = cmdServer.Flag.Int("metricsPort", 0, "Prometheus metrics listen port")
 
 	// pulseSeconds              = cmdServer.Flag.Int("pulseSeconds", 5, "number of seconds between heartbeats")
 	isStartingFiler     = cmdServer.Flag.Bool("filer", false, "whether to start filer")
@@ -83,7 +85,7 @@ func init() {
 	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.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.defaultReplicaPlacement = cmdServer.Flag.String("filer.defaultReplicaPlacement", "", "default replication type. If not specified, use master setting.")
 	filerOptions.disableDirListing = cmdServer.Flag.Bool("filer.disableDirListing", false, "turn off directory listing")
 	filerOptions.maxMB = cmdServer.Flag.Int("filer.maxMB", 32, "split files larger than the limit")
 	filerOptions.dirListingLimit = cmdServer.Flag.Int("filer.dirListLimit", 1000, "limit sub dir listing size")
@@ -98,7 +100,8 @@ func init() {
 	serverOptions.v.compactionMBPerSecond = cmdServer.Flag.Int("volume.compactionMBps", 0, "limit compaction speed in mega bytes per second")
 	serverOptions.v.fileSizeLimitMB = cmdServer.Flag.Int("volume.fileSizeLimitMB", 1024, "limit file size to avoid out of memory")
 	serverOptions.v.publicUrl = cmdServer.Flag.String("volume.publicUrl", "", "publicly accessible address")
-	serverOptions.v.pprof = &False
+	serverOptions.v.preStopSeconds = cmdServer.Flag.Int("volume.preStopSeconds", 10, "number of seconds between stop send heartbeats and stop volume server")
+	serverOptions.v.pprof = cmdServer.Flag.Bool("volume.pprof", false, "enable pprof http handlers. precludes --memprofile and --cpuprofile")
 
 	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}")
@@ -135,6 +138,7 @@ func runServer(cmd *Command, args []string) bool {
 	peers := strings.Join(peerList, ",")
 	masterOptions.peers = &peers
 
+	// ip address
 	masterOptions.ip = serverIp
 	masterOptions.ipBind = serverBindIp
 	filerOptions.masters = &peers
@@ -161,11 +165,8 @@ func runServer(cmd *Command, args []string) bool {
 	s3Options.filer = &filerAddress
 	msgBrokerOptions.filer = &filerAddress
 
-	if *filerOptions.defaultReplicaPlacement == "" {
-		*filerOptions.defaultReplicaPlacement = *masterOptions.defaultReplication
-	}
-
 	runtime.GOMAXPROCS(runtime.NumCPU())
+	go stats_collect.StartMetricsServer(*serverMetricsHttpPort)
 
 	folders := strings.Split(*volumeDataFolders, ",")
 

weed/command/volume.go

@@ -25,6 +25,7 @@ import (
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/pb/volume_server_pb"
 	"github.com/chrislusf/seaweedfs/weed/server"
+	stats_collect "github.com/chrislusf/seaweedfs/weed/stats"
 	"github.com/chrislusf/seaweedfs/weed/storage"
 	"github.com/chrislusf/seaweedfs/weed/util"
 )
@@ -55,6 +56,8 @@ type VolumeServerOptions struct {
 	fileSizeLimitMB       *int
 	minFreeSpacePercents  []float32
 	pprof                 *bool
+	preStopSeconds        *int
+	metricsHttpPort       *int
 	// pulseSeconds          *int
 }
 
@@ -66,6 +69,7 @@ func init() {
 	v.publicUrl = cmdVolume.Flag.String("publicUrl", "", "Publicly accessible address")
 	v.bindIp = cmdVolume.Flag.String("ip.bind", "0.0.0.0", "ip address to bind to")
 	v.masters = cmdVolume.Flag.String("mserver", "localhost:9333", "comma-separated master servers")
+	v.preStopSeconds = cmdVolume.Flag.Int("preStopSeconds", 10, "number of seconds between stop send heartbeats and stop volume server")
 	// 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.dataCenter = cmdVolume.Flag.String("dataCenter", "", "current volume server's data center name")
@@ -78,6 +82,7 @@ func init() {
 	v.compactionMBPerSecond = cmdVolume.Flag.Int("compactionMBps", 0, "limit background compaction or copying speed in mega bytes per second")
 	v.fileSizeLimitMB = cmdVolume.Flag.Int("fileSizeLimitMB", 1024, "limit file size to avoid out of memory")
 	v.pprof = cmdVolume.Flag.Bool("pprof", false, "enable pprof http handlers. precludes --memprofile and --cpuprofile")
+	v.metricsHttpPort = cmdVolume.Flag.Int("metricsPort", 0, "Prometheus metrics listen port")
 }
 
 var cmdVolume = &Command{
@@ -107,6 +112,8 @@ func runVolume(cmd *Command, args []string) bool {
 		grace.SetupProfiling(*v.cpuProfile, *v.memProfile)
 	}
 
+	go stats_collect.StartMetricsServer(*v.metricsHttpPort)
+
 	v.startVolumeServer(*volumeFolders, *maxVolumeCounts, *volumeWhiteListOption, *minFreeSpacePercent)
 
 	return true
@@ -206,7 +213,6 @@ func (v VolumeServerOptions) startVolumeServer(volumeFolders, maxVolumeCounts, v
 		*v.compactionMBPerSecond,
 		*v.fileSizeLimitMB,
 	)
-
 	// starting grpc server
 	grpcS := v.startGrpcService(volumeServer)
 
@@ -222,47 +228,48 @@ func (v VolumeServerOptions) startVolumeServer(volumeFolders, maxVolumeCounts, v
 	// starting the cluster http server
 	clusterHttpServer := v.startClusterHttpService(volumeMux)
 
-	stopChain := make(chan struct{})
+	stopChan := make(chan bool)
 	grace.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)
+		// Stop heartbeats
+		if !volumeServer.StopHeartbeat() {
+			glog.V(0).Infof("stop send heartbeat and wait %d seconds until shutdown ...", *v.preStopSeconds)
+			time.Sleep(time.Duration(*v.preStopSeconds) * time.Second)
 		}
 
-		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
+		shutdown(publicHttpDown, clusterHttpServer, grpcS, volumeServer)
+		stopChan <- true
 	})
 
 	select {
-	case <-stopChain:
+	case <-stopChan:
 	}
-	glog.Warningf("the volume server exit.")
+
+}
+
+func shutdown(publicHttpDown httpdown.Server, clusterHttpServer httpdown.Server, grpcS *grpc.Server, volumeServer *weed_server.VolumeServer) {
+
+	// firstly, stop the public http service to prevent from receiving new user request
+	if nil != publicHttpDown {
+		glog.V(0).Infof("stop public http server ... ")
+		if err := publicHttpDown.Stop(); err != nil {
+			glog.Warningf("stop the public http server failed, %v", err)
+		}
+	}
+
+	glog.V(0).Infof("graceful stop cluster http server ... ")
+	if err := clusterHttpServer.Stop(); err != nil {
+		glog.Warningf("stop the cluster http server failed, %v", err)
+	}
+
+	glog.V(0).Infof("graceful stop gRPC ...")
+	grpcS.GracefulStop()
+
+	volumeServer.Shutdown()
+
+	pprof.StopCPUProfile()
+
 }
 
 // check whether configure the public port

weed/command/watch.go

@@ -4,6 +4,8 @@ import (
 	"context"
 	"fmt"
 	"io"
+	"path/filepath"
+	"strings"
 	"time"
 
 	"github.com/chrislusf/seaweedfs/weed/pb"
@@ -17,7 +19,7 @@ func init() {
 }
 
 var cmdWatch = &Command{
-	UsageLine: "watch <wip> [-filer=localhost:8888] [-target=/]",
+	UsageLine: "watch [-filer=localhost:8888] [-target=/]",
 	Short:     "see recent changes on a filer",
 	Long: `See recent changes on a filer.
 
@@ -25,18 +27,61 @@ var cmdWatch = &Command{
 }
 
 var (
-	watchFiler  = cmdWatch.Flag.String("filer", "localhost:8888", "filer hostname:port")
-	watchTarget = cmdWatch.Flag.String("pathPrefix", "/", "path to a folder or file, or common prefix for the folders or files on filer")
-	watchStart  = cmdWatch.Flag.Duration("timeAgo", 0, "start time before now. \"300ms\", \"1.5h\" or \"2h45m\". Valid time units are \"ns\", \"us\" (or \"µs\"), \"ms\", \"s\", \"m\", \"h\"")
+	watchFiler   = cmdWatch.Flag.String("filer", "localhost:8888", "filer hostname:port")
+	watchTarget  = cmdWatch.Flag.String("pathPrefix", "/", "path to a folder or file, or common prefix for the folders or files on filer")
+	watchStart   = cmdWatch.Flag.Duration("timeAgo", 0, "start time before now. \"300ms\", \"1.5h\" or \"2h45m\". Valid time units are \"ns\", \"us\" (or \"µs\"), \"ms\", \"s\", \"m\", \"h\"")
+	watchPattern = cmdWatch.Flag.String("pattern", "", "full path or just filename pattern, ex: \"/home/?opher\", \"*.pdf\", see https://golang.org/pkg/path/filepath/#Match ")
 )
 
 func runWatch(cmd *Command, args []string) bool {
 
 	grpcDialOption := security.LoadClientTLS(util.GetViper(), "grpc.client")
 
+	var filterFunc func(dir, fname string) bool
+	if *watchPattern != "" {
+		if strings.Contains(*watchPattern, "/") {
+			println("watch path pattern", *watchPattern)
+			filterFunc = func(dir, fname string) bool {
+				matched, err := filepath.Match(*watchPattern, dir+"/"+fname)
+				if err != nil {
+					fmt.Printf("error: %v", err)
+				}
+				return matched
+			}
+		} else {
+			println("watch file pattern", *watchPattern)
+			filterFunc = func(dir, fname string) bool {
+				matched, err := filepath.Match(*watchPattern, fname)
+				if err != nil {
+					fmt.Printf("error: %v", err)
+				}
+				return matched
+			}
+		}
+	}
+
+	shouldPrint := func(resp *filer_pb.SubscribeMetadataResponse) bool {
+		if filterFunc == nil {
+			return true
+		}
+		if resp.EventNotification.OldEntry == nil && resp.EventNotification.NewEntry == nil {
+			return false
+		}
+		if resp.EventNotification.OldEntry != nil && filterFunc(resp.Directory, resp.EventNotification.OldEntry.Name) {
+			return true
+		}
+		if resp.EventNotification.NewEntry != nil && filterFunc(resp.EventNotification.NewParentPath, resp.EventNotification.NewEntry.Name) {
+			return true
+		}
+		return false
+	}
+
 	watchErr := pb.WithFilerClient(*watchFiler, grpcDialOption, func(client filer_pb.SeaweedFilerClient) error {
 
-		stream, err := client.SubscribeMetadata(context.Background(), &filer_pb.SubscribeMetadataRequest{
+		ctx, cancel := context.WithCancel(context.Background())
+		defer cancel()
+
+		stream, err := client.SubscribeMetadata(ctx, &filer_pb.SubscribeMetadataRequest{
 			ClientName: "watch",
 			PathPrefix: *watchTarget,
 			SinceNs:    time.Now().Add(-*watchStart).UnixNano(),
@@ -53,7 +98,10 @@ func runWatch(cmd *Command, args []string) bool {
 			if listenErr != nil {
 				return listenErr
 			}
-			fmt.Printf("events: %+v\n", resp.EventNotification)
+			if !shouldPrint(resp) {
+				continue
+			}
+			fmt.Printf("dir:%s %+v\n", resp.Directory, resp.EventNotification)
 		}
 
 	})

weed/filer/abstract_sql/abstract_sql_store.go

@@ -4,11 +4,11 @@ import (
 	"context"
 	"database/sql"
 	"fmt"
-
-	"github.com/chrislusf/seaweedfs/weed/filer2"
+	"github.com/chrislusf/seaweedfs/weed/filer"
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
 	"github.com/chrislusf/seaweedfs/weed/util"
+	"strings"
 )
 
 type AbstractSqlStore struct {
@@ -59,7 +59,7 @@ func (store *AbstractSqlStore) getTxOrDB(ctx context.Context) TxOrDB {
 	return store.DB
 }
 
-func (store *AbstractSqlStore) InsertEntry(ctx context.Context, entry *filer2.Entry) (err error) {
+func (store *AbstractSqlStore) InsertEntry(ctx context.Context, entry *filer.Entry) (err error) {
 
 	dir, name := entry.FullPath.DirAndName()
 	meta, err := entry.EncodeAttributesAndChunks()
@@ -67,19 +67,36 @@ func (store *AbstractSqlStore) InsertEntry(ctx context.Context, entry *filer2.En
 		return fmt.Errorf("encode %s: %s", entry.FullPath, err)
 	}
 
+	if len(entry.Chunks) > 50 {
+		meta = util.MaybeGzipData(meta)
+	}
+
 	res, err := store.getTxOrDB(ctx).ExecContext(ctx, store.SqlInsert, util.HashStringToLong(dir), name, dir, meta)
+	if err == nil {
+		return
+	}
+
+	if !strings.Contains(strings.ToLower(err.Error()), "duplicate") {
+		return fmt.Errorf("kv insert: %s", err)
+	}
+
+	// now the insert failed possibly due to duplication constraints
+	glog.V(1).Infof("insert %s falls back to update: %v", entry.FullPath, err)
+
+	res, err = store.getTxOrDB(ctx).ExecContext(ctx, store.SqlUpdate, meta, util.HashStringToLong(dir), name, dir)
 	if err != nil {
-		return fmt.Errorf("insert %s: %s", entry.FullPath, err)
+		return fmt.Errorf("upsert %s: %s", entry.FullPath, err)
 	}
 
 	_, err = res.RowsAffected()
 	if err != nil {
-		return fmt.Errorf("insert %s but no rows affected: %s", entry.FullPath, err)
+		return fmt.Errorf("upsert %s but no rows affected: %s", entry.FullPath, err)
 	}
 	return nil
+
 }
 
-func (store *AbstractSqlStore) UpdateEntry(ctx context.Context, entry *filer2.Entry) (err error) {
+func (store *AbstractSqlStore) UpdateEntry(ctx context.Context, entry *filer.Entry) (err error) {
 
 	dir, name := entry.FullPath.DirAndName()
 	meta, err := entry.EncodeAttributesAndChunks()
@@ -99,19 +116,23 @@ func (store *AbstractSqlStore) UpdateEntry(ctx context.Context, entry *filer2.En
 	return nil
 }
 
-func (store *AbstractSqlStore) FindEntry(ctx context.Context, fullpath util.FullPath) (*filer2.Entry, error) {
+func (store *AbstractSqlStore) FindEntry(ctx context.Context, fullpath util.FullPath) (*filer.Entry, error) {
 
 	dir, name := fullpath.DirAndName()
 	row := store.getTxOrDB(ctx).QueryRowContext(ctx, store.SqlFind, util.HashStringToLong(dir), name, dir)
+
 	var data []byte
 	if err := row.Scan(&data); err != nil {
-		return nil, filer_pb.ErrNotFound
+		if err == sql.ErrNoRows {
+			return nil, filer_pb.ErrNotFound
+		}
+		return nil, fmt.Errorf("find %s: %v", fullpath, err)
 	}
 
-	entry := &filer2.Entry{
+	entry := &filer.Entry{
 		FullPath: fullpath,
 	}
-	if err := entry.DecodeAttributesAndChunks(data); err != nil {
+	if err := entry.DecodeAttributesAndChunks(util.MaybeDecompressData(data)); err != nil {
 		return entry, fmt.Errorf("decode %s : %v", entry.FullPath, err)
 	}
 
@@ -150,14 +171,13 @@ func (store *AbstractSqlStore) DeleteFolderChildren(ctx context.Context, fullpat
 	return nil
 }
 
-func (store *AbstractSqlStore) ListDirectoryEntries(ctx context.Context, fullpath util.FullPath, startFileName string, inclusive bool, limit int) (entries []*filer2.Entry, err error) {
-
+func (store *AbstractSqlStore) ListDirectoryPrefixedEntries(ctx context.Context, fullpath util.FullPath, startFileName string, inclusive bool, limit int, prefix string) (entries []*filer.Entry, err error) {
 	sqlText := store.SqlListExclusive
 	if inclusive {
 		sqlText = store.SqlListInclusive
 	}
 
-	rows, err := store.getTxOrDB(ctx).QueryContext(ctx, sqlText, util.HashStringToLong(string(fullpath)), startFileName, string(fullpath), limit)
+	rows, err := store.getTxOrDB(ctx).QueryContext(ctx, sqlText, util.HashStringToLong(string(fullpath)), startFileName, string(fullpath), prefix+"%", limit)
 	if err != nil {
 		return nil, fmt.Errorf("list %s : %v", fullpath, err)
 	}
@@ -171,10 +191,10 @@ func (store *AbstractSqlStore) ListDirectoryEntries(ctx context.Context, fullpat
 			return nil, fmt.Errorf("scan %s: %v", fullpath, err)
 		}
 
-		entry := &filer2.Entry{
+		entry := &filer.Entry{
 			FullPath: util.NewFullPath(string(fullpath), name),
 		}
-		if err = entry.DecodeAttributesAndChunks(data); err != nil {
+		if err = entry.DecodeAttributesAndChunks(util.MaybeDecompressData(data)); err != nil {
 			glog.V(0).Infof("scan decode %s : %v", entry.FullPath, err)
 			return nil, fmt.Errorf("scan decode %s : %v", entry.FullPath, err)
 		}
@@ -185,6 +205,10 @@ func (store *AbstractSqlStore) ListDirectoryEntries(ctx context.Context, fullpat
 	return entries, nil
 }
 
+func (store *AbstractSqlStore) ListDirectoryEntries(ctx context.Context, fullpath util.FullPath, startFileName string, inclusive bool, limit int) (entries []*filer.Entry, err error) {
+	return store.ListDirectoryPrefixedEntries(ctx, fullpath, startFileName, inclusive, limit, "")
+}
+
 func (store *AbstractSqlStore) Shutdown() {
 	store.DB.Close()
 }

weed/filer/abstract_sql/abstract_sql_store_kv.go

@@ -0,0 +1,87 @@
+package abstract_sql
+
+import (
+	"context"
+	"database/sql"
+	"fmt"
+	"strings"
+
+	"github.com/chrislusf/seaweedfs/weed/filer"
+	"github.com/chrislusf/seaweedfs/weed/glog"
+	"github.com/chrislusf/seaweedfs/weed/util"
+)
+
+func (store *AbstractSqlStore) KvPut(ctx context.Context, key []byte, value []byte) (err error) {
+
+	dirStr, dirHash, name := genDirAndName(key)
+
+	res, err := store.getTxOrDB(ctx).ExecContext(ctx, store.SqlInsert, dirHash, name, dirStr, value)
+	if err != nil {
+		if !strings.Contains(strings.ToLower(err.Error()), "duplicate") {
+			return fmt.Errorf("kv insert: %s", err)
+		}
+	}
+
+	// now the insert failed possibly due to duplication constraints
+	glog.V(1).Infof("kv insert falls back to update: %s", err)
+
+	res, err = store.getTxOrDB(ctx).ExecContext(ctx, store.SqlUpdate, value, dirHash, name, dirStr)
+	if err != nil {
+		return fmt.Errorf("kv upsert: %s", err)
+	}
+
+	_, err = res.RowsAffected()
+	if err != nil {
+		return fmt.Errorf("kv upsert no rows affected: %s", err)
+	}
+	return nil
+
+}
+
+func (store *AbstractSqlStore) KvGet(ctx context.Context, key []byte) (value []byte, err error) {
+
+	dirStr, dirHash, name := genDirAndName(key)
+	row := store.getTxOrDB(ctx).QueryRowContext(ctx, store.SqlFind, dirHash, name, dirStr)
+
+	err = row.Scan(&value)
+
+	if err == sql.ErrNoRows {
+		return nil, filer.ErrKvNotFound
+	}
+
+	if err != nil {
+		return nil, fmt.Errorf("kv get: %v", err)
+	}
+
+	return
+}
+
+func (store *AbstractSqlStore) KvDelete(ctx context.Context, key []byte) (err error) {
+
+	dirStr, dirHash, name := genDirAndName(key)
+
+	res, err := store.getTxOrDB(ctx).ExecContext(ctx, store.SqlDelete, dirHash, name, dirStr)
+	if err != nil {
+		return fmt.Errorf("kv delete: %s", err)
+	}
+
+	_, err = res.RowsAffected()
+	if err != nil {
+		return fmt.Errorf("kv delete no rows affected: %s", err)
+	}
+
+	return nil
+
+}
+
+func genDirAndName(key []byte) (dirStr string, dirHash int64, name string) {
+	for len(key) < 8 {
+		key = append(key, 0)
+	}
+
+	dirHash = int64(util.BytesToUint64(key[:8]))
+	dirStr = string(key[:8])
+	name = string(key[8:])
+
+	return
+}

weed/filer/cassandra/cassandra_store.go

@@ -3,17 +3,16 @@ package cassandra
 import (
 	"context"
 	"fmt"
-
 	"github.com/gocql/gocql"
 
-	"github.com/chrislusf/seaweedfs/weed/filer2"
+	"github.com/chrislusf/seaweedfs/weed/filer"
 	"github.com/chrislusf/seaweedfs/weed/glog"
 	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
 	"github.com/chrislusf/seaweedfs/weed/util"
 )
 
 func init() {
-	filer2.Stores = append(filer2.Stores, &CassandraStore{})
+	filer.Stores = append(filer.Stores, &CassandraStore{})
 }
 
 type CassandraStore struct {
@@ -29,11 +28,16 @@ func (store *CassandraStore) Initialize(configuration util.Configuration, prefix
 	return store.initialize(
 		configuration.GetString(prefix+"keyspace"),
 		configuration.GetStringSlice(prefix+"hosts"),
+		configuration.GetString(prefix+"username"),
+		configuration.GetString(prefix+"password"),
 	)
 }
 
-func (store *CassandraStore) initialize(keyspace string, hosts []string) (err error) {
+func (store *CassandraStore) initialize(keyspace string, hosts []string, username string, password string) (err error) {
 	store.cluster = gocql.NewCluster(hosts...)
+	if username != "" && password != "" {
+		store.cluster.Authenticator = gocql.PasswordAuthenticator{Username: username, Password: password}
+	}
 	store.cluster.Keyspace = keyspace
 	store.cluster.Consistency = gocql.LocalQuorum
 	store.session, err = store.cluster.CreateSession()
@@ -53,7 +57,7 @@ func (store *CassandraStore) RollbackTransaction(ctx context.Context) error {
 	return nil
 }
 
-func (store *CassandraStore) InsertEntry(ctx context.Context, entry *filer2.Entry) (err error) {
+func (store *CassandraStore) InsertEntry(ctx context.Context, entry *filer.Entry) (err error) {
 
 	dir, name := entry.FullPath.DirAndName()
 	meta, err := entry.EncodeAttributesAndChunks()
@@ -61,6 +65,10 @@ func (store *CassandraStore) InsertEntry(ctx context.Context, entry *filer2.Entr
 		return fmt.Errorf("encode %s: %s", entry.FullPath, err)
 	}
 
+	if len(entry.Chunks) > 50 {
+		meta = util.MaybeGzipData(meta)
+	}
+
 	if err := store.session.Query(
 		"INSERT INTO filemeta (directory,name,meta) VALUES(?,?,?) USING TTL ? ",
 		dir, name, meta, entry.TtlSec).Exec(); err != nil {
@@ -70,12 +78,12 @@ func (store *CassandraStore) InsertEntry(ctx context.Context, entry *filer2.Entr
 	return nil
 }
 
-func (store *CassandraStore) UpdateEntry(ctx context.Context, entry *filer2.Entry) (err error) {
+func (store *CassandraStore) UpdateEntry(ctx context.Context, entry *filer.Entry) (err error) {
 
 	return store.InsertEntry(ctx, entry)
 }
 
-func (store *CassandraStore) FindEntry(ctx context.Context, fullpath util.FullPath) (entry *filer2.Entry, err error) {
+func (store *CassandraStore) FindEntry(ctx context.Context, fullpath util.FullPath) (entry *filer.Entry, err error) {
 
 	dir, name := fullpath.DirAndName()
 	var data []byte
@@ -91,10 +99,10 @@ func (store *CassandraStore) FindEntry(ctx context.Context, fullpath util.FullPa
 		return nil, filer_pb.ErrNotFound
 	}
 
-	entry = &filer2.Entry{
+	entry = &filer.Entry{
 		FullPath: fullpath,
 	}
-	err = entry.DecodeAttributesAndChunks(data)
+	err = entry.DecodeAttributesAndChunks(util.MaybeDecompressData(data))
 	if err != nil {
 		return entry, fmt.Errorf("decode %s : %v", entry.FullPath, err)
 	}
@@ -126,8 +134,12 @@ func (store *CassandraStore) DeleteFolderChildren(ctx context.Context, fullpath
 	return nil
 }
 
+func (store *CassandraStore) ListDirectoryPrefixedEntries(ctx context.Context, fullpath util.FullPath, startFileName string, inclusive bool, limit int, prefix string) (entries []*filer.Entry, err error) {
+	return nil, filer.ErrUnsupportedListDirectoryPrefixed
+}
+
 func (store *CassandraStore) ListDirectoryEntries(ctx context.Context, fullpath util.FullPath, startFileName string, inclusive bool,
-	limit int) (entries []*filer2.Entry, err error) {
+	limit int) (entries []*filer.Entry, err error) {
 
 	cqlStr := "SELECT NAME, meta FROM filemeta WHERE directory=? AND name>? ORDER BY NAME ASC LIMIT ?"
 	if inclusive {
@@ -138,10 +150,10 @@ func (store *CassandraStore) ListDirectoryEntries(ctx context.Context, fullpath
 	var name string
 	iter := store.session.Query(cqlStr, string(fullpath), startFileName, limit).Iter()
 	for iter.Scan(&name, &data) {
-		entry := &filer2.Entry{
+		entry := &filer.Entry{
 			FullPath: util.NewFullPath(string(fullpath), name),
 		}
-		if decodeErr := entry.DecodeAttributesAndChunks(data); decodeErr != nil {
+		if decodeErr := entry.DecodeAttributesAndChunks(util.MaybeDecompressData(data)); decodeErr != nil {
 			err = decodeErr
 			glog.V(0).Infof("list %s : %v", entry.FullPath, err)
 			break

weed/filer/cassandra/cassandra_store_kv.go

@@ -0,0 +1,61 @@
+package cassandra
+
+import (
+	"context"
+	"fmt"
+	"github.com/chrislusf/seaweedfs/weed/filer"
+	"github.com/gocql/gocql"
+)
+
+func (store *CassandraStore) KvPut(ctx context.Context, key []byte, value []byte) (err error) {
+	dir, name := genDirAndName(key)
+
+	if err := store.session.Query(
+		"INSERT INTO filemeta (directory,name,meta) VALUES(?,?,?) USING TTL ? ",
+		dir, name, value, 0).Exec(); err != nil {
+		return fmt.Errorf("kv insert: %s", err)
+	}
+
+	return nil
+}
+
+func (store *CassandraStore) KvGet(ctx context.Context, key []byte) (data []byte, err error) {
+	dir, name := genDirAndName(key)
+
+	if err := store.session.Query(
+		"SELECT meta FROM filemeta WHERE directory=? AND name=?",
+		dir, name).Consistency(gocql.One).Scan(&data); err != nil {
+		if err != gocql.ErrNotFound {
+			return nil, filer.ErrKvNotFound
+		}
+	}
+
+	if len(data) == 0 {
+		return nil, filer.ErrKvNotFound
+	}
+
+	return data, nil
+}
+
+func (store *CassandraStore) KvDelete(ctx context.Context, key []byte) (err error) {
+	dir, name := genDirAndName(key)
+
+	if err := store.session.Query(
+		"DELETE FROM filemeta WHERE directory=? AND name=?",
+		dir, name).Exec(); err != nil {
+		return fmt.Errorf("kv delete: %v", err)
+	}
+
+	return nil
+}
+
+func genDirAndName(key []byte) (dir string, name string) {
+	for len(key) < 8 {
+		key = append(key, 0)
+	}
+
+	dir = string(key[:8])
+	name = string(key[8:])
+
+	return
+}

weed/filer/configuration.go

@@ -1,4 +1,4 @@
-package filer2
+package filer
 
 import (
 	"os"

weed/filer/elastic/v7/elastic_store.go

@@ -0,0 +1,338 @@
+package elastic
+
+import (
+	"context"
+	"fmt"
+	"math"
+	"strings"
+
+	"github.com/chrislusf/seaweedfs/weed/filer"
+	"github.com/chrislusf/seaweedfs/weed/glog"
+	"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
+	weed_util "github.com/chrislusf/seaweedfs/weed/util"
+	jsoniter "github.com/json-iterator/go"
+	elastic "github.com/olivere/elastic/v7"
+)
+
+var (
+	indexType   = "_doc"
+	indexPrefix = ".seaweedfs_"
+	indexKV     = ".seaweedfs_kv_entries"
+	kvMappings  = ` {
+		     "mappings": {
+		   	"enabled": false,
+		       "properties": {
+		         "Value":{
+		           "type": "binary"
+		         }
+		       }
+		     }
+		   }`
+)
+
+type ESEntry struct {
+	ParentId string `json:"ParentId"`
+	Entry    *filer.Entry
+}
+
+type ESKVEntry struct {
+	Value []byte `json:"Value"`
+}
+
+func init() {
+	filer.Stores = append(filer.Stores, &ElasticStore{})
+}
+
+type ElasticStore struct {
+	client      *elastic.Client
+	maxPageSize int
+}
+
+func (store *ElasticStore) GetName() string {
+	return "elastic7"
+}
+
+func (store *ElasticStore) Initialize(configuration weed_util.Configuration, prefix string) (err error) {
+	options := []elastic.ClientOptionFunc{}
+	servers := configuration.GetStringSlice(prefix + "servers")
+	options = append(options, elastic.SetURL(servers...))
+	username := configuration.GetString(prefix + "username")
+	password := configuration.GetString(prefix + "password")
+	if username != "" && password != "" {
+		options = append(options, elastic.SetBasicAuth(username, password))
+	}
+	options = append(options, elastic.SetSniff(configuration.GetBool(prefix+"sniff_enabled")))
+	options = append(options, elastic.SetHealthcheck(configuration.GetBool(prefix+"healthcheck_enabled")))
+	store.maxPageSize = configuration.GetInt(prefix + "index.max_result_window")
+	if store.maxPageSize <= 0 {
+		store.maxPageSize = 10000
+	}
+	glog.Infof("filer store elastic endpoints: %v.", servers)
+	return store.initialize(options)
+}
+
+func (store *ElasticStore) initialize(options []elastic.ClientOptionFunc) (err error) {
+	ctx := context.Background()
+	store.client, err = elastic.NewClient(options...)
+	if err != nil {
+		return fmt.Errorf("init elastic %v.", err)
+	}
+	if ok, err := store.client.IndexExists(indexKV).Do(ctx); err == nil && !ok {
+		_, err = store.client.CreateIndex(indexKV).Body(kvMappings).Do(ctx)
+		if err != nil {
+			return fmt.Errorf("create index(%s) %v.", indexKV, err)
+		}
+	}
+	return nil
+}
+
+func (store *ElasticStore) BeginTransaction(ctx context.Context) (context.Context, error) {
+	return ctx, nil
+}
+func (store *ElasticStore) CommitTransaction(ctx context.Context) error {
+	return nil
+}
+func (store *ElasticStore) RollbackTransaction(ctx context.Context) error {
+	return nil
+}
+
+func (store *ElasticStore) ListDirectoryPrefixedEntries(ctx context.Context, fullpath weed_util.FullPath, startFileName string, inclusive bool, limit int, prefix string) (entries []*filer.Entry, err error) {
+	return nil, filer.ErrUnsupportedListDirectoryPrefixed
+}
+
+func (store *ElasticStore) InsertEntry(ctx context.Context, entry *filer.Entry) (err error) {
+	index := getIndex(entry.FullPath)
+	dir, _ := entry.FullPath.DirAndName()
+	id := weed_util.Md5String([]byte(entry.FullPath))
+	esEntry := &ESEntry{
+		ParentId: weed_util.Md5String([]byte(dir)),
+		Entry:    entry,
+	}
+	value, err := jsoniter.Marshal(esEntry)
+	if err != nil {
+		glog.Errorf("insert entry(%s) %v.", string(entry.FullPath), err)
+		return fmt.Errorf("insert entry %v.", err)
+	}
+	_, err = store.client.Index().
+		Index(index).
+		Type(indexType).
+		Id(id).
+		BodyJson(string(value)).
+		Do(ctx)
+	if err != nil {
+		glog.Errorf("insert entry(%s) %v.", string(entry.FullPath), err)
+		return fmt.Errorf("insert entry %v.", err)
+	}
+	return nil
+}
+
+func (store *ElasticStore) UpdateEntry(ctx context.Context, entry *filer.Entry) (err error) {
+	return store.InsertEntry(ctx, entry)
+}
+
+func (store *ElasticStore) FindEntry(ctx context.Context, fullpath weed_util.FullPath) (entry *filer.Entry, err error) {
+	index := getIndex(fullpath)
+	id := weed_util.Md5String([]byte(fullpath))
+	searchResult, err := store.client.Get().
+		Index(index).
+		Type(indexType).
+		Id(id).
+		Do(ctx)
+	if elastic.IsNotFound(err) {
+		return nil, filer_pb.ErrNotFound
+	}
+	if searchResult != nil && searchResult.Found {
+		esEntry := &ESEntry{
+			ParentId: "",
+			Entry:    &filer.Entry{},
+		}
+		err := jsoniter.Unmarshal(searchResult.Source, esEntry)
+		return esEntry.Entry, err
+	}
+	glog.Errorf("find entry(%s),%v.", string(fullpath), err)
+	return nil, filer_pb.ErrNotFound
+}
+
+func (store *ElasticStore) DeleteEntry(ctx context.Context, fullpath weed_util.FullPath) (err error) {
+	index := getIndex(fullpath)
+	id := weed_util.Md5String([]byte(fullpath))
+	if strings.Count(string(fullpath), "/") == 1 {
+		return store.deleteIndex(ctx, index)
+	}
+	return store.deleteEntry(ctx, index, id)
+}
+
+func (store *ElasticStore) deleteIndex(ctx context.Context, index string) (err error) {
+	deleteResult, err := store.client.DeleteIndex(index).Do(ctx)
+	if elastic.IsNotFound(err) || (err == nil && deleteResult.Acknowledged) {
+		return nil
+	}
+	glog.Errorf("delete index(%s) %v.", index, err)
+	return err
+}
+
+func (store *ElasticStore) deleteEntry(ctx context.Context, index, id string) (err error) {
+	deleteResult, err := store.client.Delete().
+		Index(index).
+		Type(indexType).
+		Id(id).
+		Do(ctx)
+	if err == nil {
+		if deleteResult.Result == "deleted" || deleteResult.Result == "not_found" {
+			return nil
+		}
+	}
+	glog.Errorf("delete entry(index:%s,_id:%s) %v.", index, id, err)
+	return fmt.Errorf("delete entry %v.", err)
+}
+
+func (store *ElasticStore) DeleteFolderChildren(ctx context.Context, fullpath weed_util.FullPath) (err error) {
+	if entries, err := store.ListDirectoryEntries(ctx, fullpath, "", false, math.MaxInt32); err == nil {
+		for _, entry := range entries {
+			store.DeleteEntry(ctx, entry.FullPath)
+		}
+	}
+	return nil
+}
+
+func (store *ElasticStore) ListDirectoryEntries(
+	ctx context.Context, fullpath weed_util.FullPath, startFileName string, inclusive bool, limit int,
+) (entries []*filer.Entry, err error) {
+	if string(fullpath) == "/" {
+		return store.listRootDirectoryEntries(ctx, startFileName, inclusive, limit)
+	}
+	return store.listDirectoryEntries(ctx, fullpath, startFileName, inclusive, limit)
+}
+
+func (store *ElasticStore) listRootDirectoryEntries(ctx context.Context, startFileName string, inclusive bool, limit int) (entries []*filer.Entry, err error) {
+	indexResult, err := store.client.CatIndices().Do(ctx)
+	if err != nil {
+		glog.Errorf("list indices %v.", err)
+		return entries, err
+	}
+	for _, index := range indexResult {
+		if index.Index == indexKV {
+			continue
+		}
+		if strings.HasPrefix(index.Index, indexPrefix) {
+			if entry, err := store.FindEntry(ctx,
+				weed_util.FullPath("/"+strings.Replace(index.Index, indexPrefix, "", 1))); err == nil {
+				fileName := getFileName(entry.FullPath)
+				if fileName == startFileName && !inclusive {
+					continue
+				}
+				limit--
+				if limit < 0 {
+					break
+				}
+				entries = append(entries, entry)
+			}
+		}
+	}
+	return entries, nil
+}
+
+func (store *ElasticStore) listDirectoryEntries(
+	ctx context.Context, fullpath weed_util.FullPath, startFileName string, inclusive bool, limit int,
+) (entries []*filer.Entry, err error) {
+	first := true
+	index := getIndex(fullpath)
+	nextStart := ""
+	parentId := weed_util.Md5String([]byte(fullpath))
+	if _, err := store.client.Refresh(index).Do(ctx); err != nil {
+		if elastic.IsNotFound(err) {
+			store.client.CreateIndex(index).Do(ctx)
+			return entries, nil
+		}
+	}
+	for {
+		result := &elastic.SearchResult{}
+		if (startFileName == "" && first) || inclusive {
+			if result, err = store.search(ctx, index, parentId); err != nil {
+				glog.Errorf("search (%s,%s,%t,%d) %v.", string(fullpath), startFileName, inclusive, limit, err)
+				return entries, err
+			}
+		} else {
+			fullPath := string(fullpath) + "/" + startFileName
+			if !first {
+				fullPath = nextStart
+			}
+			after := weed_util.Md5String([]byte(fullPath))
+			if result, err = store.searchAfter(ctx, index, parentId, after); err != nil {
+				glog.Errorf("searchAfter (%s,%s,%t,%d) %v.", string(fullpath), startFileName, inclusive, limit, err)
+				return entries, err
+			}
+		}
+		first = false
+		for _, hit := range result.Hits.Hits {
+			esEntry := &ESEntry{
+				ParentId: "",
+				Entry:    &filer.Entry{},
+			}
+			if err := jsoniter.Unmarshal(hit.Source, esEntry); err == nil {
+				limit--
+				if limit < 0 {
+					return entries, nil
+				}
+				nextStart = string(esEntry.Entry.FullPath)
+				fileName := getFileName(esEntry.Entry.FullPath)
+				if fileName == startFileName && !inclusive {
+					continue
+				}
+				entries = append(entries, esEntry.Entry)
+			}
+		}
+		if len(result.Hits.Hits) < store.maxPageSize {
+			break
+		}
+	}
+	return entries, nil
+}
+
+func (store *ElasticStore) search(ctx context.Context, index, parentId string) (result *elastic.SearchResult, err error) {
+	if count, err := store.client.Count(index).Do(ctx); err == nil && count == 0 {
+		return &elastic.SearchResult{
+			Hits: &elastic.SearchHits{
+				Hits: make([]*elastic.SearchHit, 0)},
+		}, nil
+	}
+	queryResult, err := store.client.Search().
+		Index(index).
+		Query(elastic.NewMatchQuery("ParentId", parentId)).
+		Size(store.maxPageSize).
+		Sort("_id", false).
+		Do(ctx)
+	return queryResult, err
+}
+
+func (store *ElasticStore) searchAfter(ctx context.Context, index, parentId, after string) (result *elastic.SearchResult, err error) {
+	queryResult, err := store.client.Search().
+		Index(index).
+		Query(elastic.NewMatchQuery("ParentId", parentId)).
+		SearchAfter(after).
+		Size(store.maxPageSize).
+		Sort("_id", false).
+		Do(ctx)
+	return queryResult, err
+
+}
+
+func (store *ElasticStore) Shutdown() {
+	store.client.Stop()
+}
+
+func getIndex(fullpath weed_util.FullPath) string {
+	path := strings.Split(string(fullpath), "/")
+	if len(path) > 1 {
+		return indexPrefix + path[1]
+	}
+	return ""
+}
+
+func getFileName(fullpath weed_util.FullPath) string {
+	path := strings.Split(string(fullpath), "/")
+	if len(path) > 1 {
+		return path[len(path)-1]
+	}
+	return ""
+}

weed/filer/elastic/v7/elastic_store_kv.go

@@ -0,0 +1,65 @@
+package elastic
+
+import (
+	"context"
+	"fmt"
+
+	"github.com/chrislusf/seaweedfs/weed/filer"
+
+	"github.com/chrislusf/seaweedfs/weed/glog"
+	jsoniter "github.com/json-iterator/go"
+	elastic "github.com/olivere/elastic/v7"
+)
+
+func (store *ElasticStore) KvDelete(ctx context.Context, key []byte) (err error) {
+	deleteResult, err := store.client.Delete().
+		Index(indexKV).
+		Type(indexType).
+		Id(string(key)).
+		Do(ctx)
+	if err == nil {
+		if deleteResult.Result == "deleted" || deleteResult.Result == "not_found" {
+			return nil
+		}
+	}
+	glog.Errorf("delete key(id:%s) %v.", string(key), err)
+	return fmt.Errorf("delete key %v.", err)
+}
+
+func (store *ElasticStore) KvGet(ctx context.Context, key []byte) (value []byte, err error) {
+	searchResult, err := store.client.Get().
+		Index(indexKV).
+		Type(indexType).
+		Id(string(key)).
+		Do(ctx)
+	if elastic.IsNotFound(err) {
+		return value, filer.ErrKvNotFound
+	}
+	if searchResult != nil && searchResult.Found {
+		esEntry := &ESKVEntry{}
+		if err := jsoniter.Unmarshal(searchResult.Source, esEntry); err == nil {
+			return esEntry.Value, nil
+		}
+	}
+	glog.Errorf("find key(%s),%v.", string(key), err)
+	return value, filer.ErrKvNotFound
+}
+
+func (store *ElasticStore) KvPut(ctx context.Context, key []byte, value []byte) (err error) {
+	esEntry := &ESKVEntry{value}
+	val, err := jsoniter.Marshal(esEntry)
+	if err != nil {
+		glog.Errorf("insert key(%s) %v.", string(key), err)
+		return fmt.Errorf("insert key %v.", err)
+	}
+	_, err = store.client.Index().
+		Index(indexKV).
+		Type(indexType).
+		Id(string(key)).
+		BodyJson(string(val)).
+		Do(ctx)
+	if err != nil {
+		return fmt.Errorf("kv put: %v", err)
+	}
+	return nil
+}