Files
seaweedfs/weed/command/volume.go
Chris Lu 9ae905e456 feat(security): hot-reload HTTPS certs without restart (k8s cert-manager) (#9181)
* feat(security): hot-reload HTTPS certs for master/volume/filer/webdav/admin

S3 and filer already use a refreshing pemfile provider for their HTTPS
cert, so rotated certificates (e.g. from k8s cert-manager) are picked up
without a restart. Master, volume, webdav, and admin, however, passed
cert/key paths straight to ServeTLS/ListenAndServeTLS and loaded once at
startup — rotating those certs required a pod restart.

Add a small helper NewReloadingServerCertificate in weed/security that
wraps pemfile.Provider and returns a tls.Config.GetCertificate closure,
then wire it into the four remaining HTTPS entry points. httpdown now
also calls ServeTLS when TLSConfig carries a GetCertificate/Certificates
but CertFile/KeyFile are empty, so volume server can pre-populate
TLSConfig.

A unit test exercises the rotation path (write cert, rotate on disk,
assert the callback returns the new cert) with a short refresh window.

* refactor(security): route filer/s3 HTTPS through the shared cert reloader

Before: filer.go and s3.go each kept a *certprovider.Provider on the
options struct plus a duplicated GetCertificateWithUpdate method. Both
were loading pemfile themselves. Behaviorally they already reloaded, but
the logic was duplicated two ways and neither path was shared with the
newly-added master/volume/webdav/admin wiring.

After: both use security.NewReloadingServerCertificate like the other
servers. The per-struct certProvider field and GetCertificateWithUpdate
method are removed, along with the now-unused certprovider and pemfile
imports. Net: -32 lines, one code path for all HTTPS cert reloading.

No behavior change — the refresh window, cache, and handshake contract
are identical (the helper wraps the same pemfile.NewProvider).

* feat(security): hot-reload HTTPS client certs for mount/backup/upload/etc

The HTTP client in weed/util/http/client loaded the mTLS client cert
once at startup via tls.LoadX509KeyPair. That left every long-lived
HTTPS client process (weed mount, backup, filer.copy, filer→volume,
s3→filer/volume) unable to pick up a rotated client cert without a
restart — even though the same cert-manager setup was already rotating
the server side fine.

Swap the client cert loader for a tls.Config.GetClientCertificate
callback backed by the same refreshing pemfile provider. New TLS
handshakes pick up the rotated cert; in-flight pooled connections keep
their old cert and drop as normal transport churn happens.

To keep this reusable from both server and client TLS code without an
import cycle (weed/security already imports weed/util/http/client for
LoadHTTPClientFromFile), extract the pemfile wrapper into a new
weed/security/certreload subpackage. weed/security keeps its thin
NewReloadingServerCertificate wrapper. The existing unit test moves
with the implementation.

gRPC mTLS was already handled by security.LoadServerTLS /
LoadClientTLS; this PR does not change any gRPC paths. MQ broker, MQ
agent, Kafka gateway, and FUSE mount control plane are gRPC-only and
therefore already rotate.

CA bundles (ClientCAs / RootCAs / grpc.ca) are still loaded once — noted
as a known limitation in the wiki.

* fix(security): address PR review feedback on cert reloader

Bots (gemini-code-assist + coderabbit) flagged three real issues and a
couple of nits. Addressing them here:

1. KeyMaterial used context.Background(). The grpc pemfile provider's
   KeyMaterial blocks until material arrives or the context deadline
   expires; with Background() a slow disk could hang the TLS handshake
   indefinitely. Switched both the server and client callbacks to use
   hello.Context() / cri.Context() so a stuck read is bounded by the
   handshake timeout.

2. Admin server loaded TLS inside the serve goroutine. If the cert was
   bad, the goroutine returned but startAdminServer kept blocking on
   <-ctx.Done() with no listener, making the process look healthy with
   nothing bound. Moved TLS setup to run before the goroutine starts
   and propagate errors via fmt.Errorf; also captures the provider and
   defers Close().

3. HTTP client discarded the certprovider.Provider from
   NewClientGetCertificate. That leaked the refresh goroutine, and
   NewHttpClientWithTLS had a worse case where a CA-file failure after
   provider creation orphaned the provider entirely. Added a
   certProvider field and a Close() method on HTTPClient, and made
   the constructors close the provider on subsequent error paths.

4. Server-side paths (master/volume/filer/s3/webdav/admin) now retain
   the provider. filer and webdav run ServeTLS synchronously, so a
   plain defer works. master/volume/s3 dispatch goroutines and return
   while the server keeps running, so they hook Close() into
   grace.OnInterrupt.

5. Test: certreload_test now tolerates transient read/parse errors
   during file rotation (writeSelfSigned rewrites cert before key) and
   reports the last error only if the deadline expires.

No user-visible behavior change for the happy path.

* test(tls): add end-to-end HTTPS cert rotation integration test

Boots a real `weed master` with HTTPS enabled, captures the leaf cert
served at TLS handshake time, atomically rewrites the cert/key files
on disk (the same rename-in-place pattern kubelet does when it swaps
a cert-manager Secret), and asserts that a subsequent TLS handshake
observes the rotated leaf — with no process restart, no SIGHUP, no
reloader sidecar. Verifies the full path: on-disk change → pemfile
refresh tick → provider.KeyMaterial → tls.Config.GetCertificate →
server TLS handshake.

Runtime is ~1s by exposing the reloader's refresh window as an env
var (WEED_TLS_CERT_REFRESH_INTERVAL) and setting it to 500ms for the
test. The same env var is user-facing — documented in the wiki — so
operators running short-lived certs (Vault, cert-manager with
duration: 24h, etc.) can tighten the rotation-pickup window without a
rebuild. Defaults to 5h to preserve prior behavior.

security.CredRefreshingInterval is kept for API compatibility but now
aliases certreload.DefaultRefreshInterval so the same env controls
both gRPC mTLS and HTTPS reload.

* ci(tls): wire the TLS rotation integration test into GitHub Actions

Mirrors the existing vacuum-integration-tests.yml shape: Ubuntu runner,
Go 1.25, build weed, run `go test` in test/tls_rotation, upload master
logs on failure. 10-minute job timeout; the test itself finishes in
about a second because WEED_TLS_CERT_REFRESH_INTERVAL is set to 500ms
inside the test.

Runs on every push to master and on every PR to master.

* fix(tls): address follow-up PR review comments

Three new comments on the integration test + volume shutdown path:

1. Test: peekServerCert was swallowing every dial/handshake error,
   which meant waitForCert's "last err: <nil>" fatal message lost all
   diagnostic value. Thread errors back through: peekServerCert now
   returns (*x509.Certificate, error), and waitForCert records the
   latest error so a CI flake points at the actual cause (master
   didn't come up, handshake rejected, CA pool mismatch, etc.).

2. Test: set HOME=<tempdir> on the master subprocess. Viper today
   registers the literal path "$HOME/.seaweedfs" without env
   expansion, so a developer's ~/.seaweedfs/security.toml is
   accidentally invisible — the test was relying on that. Pinning
   HOME is belt-and-braces against a future viper upgrade that does
   expand env vars.

3. volume.go: startClusterHttpService's provider close was registered
   via grace.OnInterrupt, which fires on SIGTERM but NOT on the
   v.shutdownCtx.Done() path used by mini / integration tests. The
   pemfile refresh goroutine leaked in that shutdown path. Now the
   helper returns a close func and the caller invokes it on BOTH
   shutdown paths for parity.

Also add MinVersion: TLS 1.2 to the test's tls.Config to quiet the
ast-grep static-analysis nit — zero-risk since the pool only trusts
our in-memory CA.

Test runs clean 3/3.
2026-04-21 20:20:11 -07:00

507 lines
19 KiB
Go

package command
import (
"context"
"crypto/tls"
"fmt"
"net/http"
httppprof "net/http/pprof"
"os"
"runtime/pprof"
"strconv"
"strings"
"time"
"github.com/spf13/viper"
"google.golang.org/grpc"
"google.golang.org/grpc/reflection"
"github.com/seaweedfs/seaweedfs/weed/glog"
"github.com/seaweedfs/seaweedfs/weed/pb"
"github.com/seaweedfs/seaweedfs/weed/pb/volume_server_pb"
"github.com/seaweedfs/seaweedfs/weed/security"
weed_server "github.com/seaweedfs/seaweedfs/weed/server"
"github.com/seaweedfs/seaweedfs/weed/server/constants"
stats_collect "github.com/seaweedfs/seaweedfs/weed/stats"
"github.com/seaweedfs/seaweedfs/weed/storage"
"github.com/seaweedfs/seaweedfs/weed/storage/types"
"github.com/seaweedfs/seaweedfs/weed/util"
"github.com/seaweedfs/seaweedfs/weed/util/grace"
"github.com/seaweedfs/seaweedfs/weed/util/httpdown"
"github.com/seaweedfs/seaweedfs/weed/util/version"
)
var (
v VolumeServerOptions
)
type VolumeServerOptions struct {
port *int
portGrpc *int
publicPort *int
folders []string
folderMaxLimits []int32
idxFolder *string
ip *string
id *string
publicUrl *string
bindIp *string
mastersString *string
mserverString *string // deprecated, for backward compatibility
masters []pb.ServerAddress
idleConnectionTimeout *int
dataCenter *string
rack *string
whiteList []string
indexType *string
diskType *string
tags *string
fixJpgOrientation *bool
readMode *string
cpuProfile *string
memProfile *string
compactionMBPerSecond *int
maintenanceMBPerSecond *int
fileSizeLimitMB *int
concurrentUploadLimitMB *int
concurrentDownloadLimitMB *int
pprof *bool
preStopSeconds *int
metricsHttpPort *int
metricsHttpIp *string
// pulseSeconds *int
inflightUploadDataTimeout *time.Duration
inflightDownloadDataTimeout *time.Duration
hasSlowRead *bool
readBufferSizeMB *int
ldbTimeout *int64
debug *bool
debugPort *int
// shutdownCtx, when non-nil, tells startVolumeServer to shut down once the
// ctx is cancelled. Used by integration tests and by weed mini; nil for
// standalone weed volume.
shutdownCtx context.Context
}
func init() {
cmdVolume.Run = runVolume // break init cycle
v.port = cmdVolume.Flag.Int("port", 8080, "http listen port")
v.portGrpc = cmdVolume.Flag.Int("port.grpc", 0, "grpc listen port")
v.publicPort = cmdVolume.Flag.Int("port.public", 0, "port opened to public")
v.ip = cmdVolume.Flag.String("ip", util.DetectedHostAddress(), "ip or server name, also used as identifier")
v.id = cmdVolume.Flag.String("id", "", "volume server id. If empty, default to ip:port")
v.publicUrl = cmdVolume.Flag.String("publicUrl", "", "Publicly accessible address")
v.bindIp = cmdVolume.Flag.String("ip.bind", "", "ip address to bind to. If empty, default to same as -ip option.")
v.mastersString = cmdVolume.Flag.String("master", "localhost:9333", "comma-separated master servers")
v.mserverString = cmdVolume.Flag.String("mserver", "", "comma-separated master servers (deprecated, use -master instead)")
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")
v.rack = cmdVolume.Flag.String("rack", "", "current volume server's rack name")
v.indexType = cmdVolume.Flag.String("index", "memory", "Choose [memory|leveldb|leveldbMedium|leveldbLarge] mode for memory~performance balance.")
v.diskType = cmdVolume.Flag.String("disk", "", "[hdd|ssd|<tag>] hard drive or solid state drive or any tag")
v.tags = cmdVolume.Flag.String("tags", "", "comma-separated tag groups per data dir; each group uses ':' (e.g. fast:ssd,archive)")
v.fixJpgOrientation = cmdVolume.Flag.Bool("images.fix.orientation", false, "Adjust jpg orientation when uploading.")
v.readMode = cmdVolume.Flag.String("readMode", "proxy", "[local|proxy|redirect] how to deal with non-local volume: 'not found|proxy to remote node|redirect volume location'.")
v.cpuProfile = cmdVolume.Flag.String("cpuprofile", "", "cpu profile output file")
v.memProfile = cmdVolume.Flag.String("memprofile", "", "memory profile output file")
v.compactionMBPerSecond = cmdVolume.Flag.Int("compactionMBps", 0, "limit background compaction or copying speed in mega bytes per second")
v.maintenanceMBPerSecond = cmdVolume.Flag.Int("maintenanceMBps", 0, "limit maintenance (replication / balance) IO rate in MB/s. Unset is 0, no limitation.")
v.fileSizeLimitMB = cmdVolume.Flag.Int("fileSizeLimitMB", 256, "limit file size to avoid out of memory")
v.ldbTimeout = cmdVolume.Flag.Int64("index.leveldbTimeout", 0, "alive time for leveldb (default to 0). If leveldb of volume is not accessed in ldbTimeout hours, it will be off loaded to reduce opened files and memory consumption.")
v.concurrentUploadLimitMB = cmdVolume.Flag.Int("concurrentUploadLimitMB", 0, "limit total concurrent upload size, 0 means unlimited")
v.concurrentDownloadLimitMB = cmdVolume.Flag.Int("concurrentDownloadLimitMB", 0, "limit total concurrent download size, 0 means unlimited")
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")
v.metricsHttpIp = cmdVolume.Flag.String("metricsIp", "", "metrics listen ip. If empty, default to same as -ip.bind option.")
v.idxFolder = cmdVolume.Flag.String("dir.idx", "", "directory to store .idx files")
v.inflightUploadDataTimeout = cmdVolume.Flag.Duration("inflightUploadDataTimeout", 60*time.Second, "inflight upload data wait timeout of volume servers")
v.inflightDownloadDataTimeout = cmdVolume.Flag.Duration("inflightDownloadDataTimeout", 60*time.Second, "inflight download data wait timeout of volume servers")
v.hasSlowRead = cmdVolume.Flag.Bool("hasSlowRead", true, "<experimental> if true, this prevents slow reads from blocking other requests, but large file read P99 latency will increase.")
v.readBufferSizeMB = cmdVolume.Flag.Int("readBufferSizeMB", 4, "<experimental> larger values can optimize query performance but will increase some memory usage,Use with hasSlowRead normally.")
v.debug = cmdVolume.Flag.Bool("debug", false, "serves runtime profiling data via pprof on the port specified by -debug.port")
v.debugPort = cmdVolume.Flag.Int("debug.port", 6060, "http port for debugging")
}
var cmdVolume = &Command{
UsageLine: "volume -port=8080 -dir=/tmp -max=5 -ip=server_name -master=localhost:9333",
Short: "start a volume server",
Long: `start a volume server to provide storage spaces
`,
}
var (
volumeFolders = cmdVolume.Flag.String("dir", os.TempDir(), "directories to store data files. dir[,dir]...")
maxVolumeCounts = cmdVolume.Flag.String("max", "8", "maximum numbers of volumes, count[,count]... If set to zero, the limit will be auto configured as free disk space divided by volume size.")
volumeWhiteListOption = cmdVolume.Flag.String("whiteList", "", "comma separated Ip addresses having write permission. No limit if empty.")
minFreeSpacePercent = cmdVolume.Flag.String("minFreeSpacePercent", "1", "minimum free disk space (default to 1%). Low disk space will mark all volumes as ReadOnly (deprecated, use minFreeSpace instead).")
minFreeSpace = cmdVolume.Flag.String("minFreeSpace", "", "min free disk space (value<=100 as percentage like 1, other as human readable bytes, like 10GiB). Low disk space will mark all volumes as ReadOnly.")
)
func runVolume(cmd *Command, args []string) bool {
if *v.debug {
grace.StartDebugServer(*v.debugPort)
}
util.LoadSecurityConfiguration()
// If --pprof is set we assume the caller wants to be able to collect
// cpu and memory profiles via go tool pprof
if !*v.pprof {
grace.SetupProfiling(*v.cpuProfile, *v.memProfile)
}
switch {
case *v.metricsHttpIp != "":
// noting to do, use v.metricsHttpIp
case *v.bindIp != "":
*v.metricsHttpIp = *v.bindIp
case *v.ip != "":
*v.metricsHttpIp = *v.ip
}
go stats_collect.StartMetricsServer(*v.metricsHttpIp, *v.metricsHttpPort)
// Backward compatibility: if -mserver is provided, use it
if *v.mserverString != "" {
*v.mastersString = *v.mserverString
}
minFreeSpaces := util.MustParseMinFreeSpace(*minFreeSpace, *minFreeSpacePercent)
v.masters = pb.ServerAddresses(*v.mastersString).ToAddresses()
v.startVolumeServer(*volumeFolders, *maxVolumeCounts, *volumeWhiteListOption, minFreeSpaces)
return true
}
func (v VolumeServerOptions) startVolumeServer(volumeFolders, maxVolumeCounts, volumeWhiteListOption string, minFreeSpaces []util.MinFreeSpace) {
// Set multiple folders and each folder's max volume count limit'
v.folders = strings.Split(volumeFolders, ",")
for _, folder := range v.folders {
if err := util.TestFolderWritable(util.ResolvePath(folder)); err != nil {
glog.Fatalf("Check Data Folder(-dir) Writable %s : %s", folder, err)
}
}
// set max
maxCountStrings := strings.Split(maxVolumeCounts, ",")
for _, maxString := range maxCountStrings {
if max, e := strconv.ParseInt(maxString, 10, 64); e == nil {
v.folderMaxLimits = append(v.folderMaxLimits, int32(max))
} else {
glog.Fatalf("The max specified in -max not a valid number %s", maxString)
}
}
if len(v.folderMaxLimits) == 1 && len(v.folders) > 1 {
for i := 0; i < len(v.folders)-1; i++ {
v.folderMaxLimits = append(v.folderMaxLimits, v.folderMaxLimits[0])
}
}
if len(v.folders) != len(v.folderMaxLimits) {
glog.Fatalf("%d directories by -dir, but only %d max is set by -max", len(v.folders), len(v.folderMaxLimits))
}
if len(minFreeSpaces) == 1 && len(v.folders) > 1 {
for i := 0; i < len(v.folders)-1; i++ {
minFreeSpaces = append(minFreeSpaces, minFreeSpaces[0])
}
}
if len(v.folders) != len(minFreeSpaces) {
glog.Fatalf("%d directories by -dir, but only %d minFreeSpacePercent is set by -minFreeSpacePercent", len(v.folders), len(minFreeSpaces))
}
// set disk types
var diskTypes []types.DiskType
diskTypeStrings := strings.Split(*v.diskType, ",")
for _, diskTypeString := range diskTypeStrings {
diskTypes = append(diskTypes, types.ToDiskType(diskTypeString))
}
if len(diskTypes) == 1 && len(v.folders) > 1 {
for i := 0; i < len(v.folders)-1; i++ {
diskTypes = append(diskTypes, diskTypes[0])
}
}
if len(v.folders) != len(diskTypes) {
glog.Fatalf("%d directories by -dir, but only %d disk types is set by -disk", len(v.folders), len(diskTypes))
}
var tagsArg string
if v.tags != nil {
tagsArg = *v.tags
}
folderTags := parseVolumeTags(tagsArg, len(v.folders))
// security related white list configuration
v.whiteList = util.StringSplit(volumeWhiteListOption, ",")
if *v.ip == "" {
*v.ip = util.DetectedHostAddress()
glog.V(0).Infof("detected volume server ip address: %v", *v.ip)
}
if *v.bindIp == "" {
*v.bindIp = *v.ip
}
if *v.publicPort == 0 {
*v.publicPort = *v.port
}
if *v.portGrpc == 0 {
*v.portGrpc = 10000 + *v.port
}
if *v.publicUrl == "" {
*v.publicUrl = util.JoinHostPort(*v.ip, *v.publicPort)
}
volumeMux := http.NewServeMux()
publicVolumeMux := volumeMux
if v.isSeparatedPublicPort() {
publicVolumeMux = http.NewServeMux()
}
if *v.pprof {
volumeMux.HandleFunc("/debug/pprof/", httppprof.Index)
volumeMux.HandleFunc("/debug/pprof/cmdline", httppprof.Cmdline)
volumeMux.HandleFunc("/debug/pprof/profile", httppprof.Profile)
volumeMux.HandleFunc("/debug/pprof/symbol", httppprof.Symbol)
volumeMux.HandleFunc("/debug/pprof/trace", httppprof.Trace)
}
volumeNeedleMapKind := storage.NeedleMapInMemory
switch *v.indexType {
case "leveldb":
volumeNeedleMapKind = storage.NeedleMapLevelDb
case "leveldbMedium":
volumeNeedleMapKind = storage.NeedleMapLevelDbMedium
case "leveldbLarge":
volumeNeedleMapKind = storage.NeedleMapLevelDbLarge
}
// Determine volume server ID: if not specified, use ip:port
volumeServerId := util.GetVolumeServerId(*v.id, *v.ip, *v.port)
volumeServer := weed_server.NewVolumeServer(volumeMux, publicVolumeMux,
*v.ip, *v.port, *v.portGrpc, *v.publicUrl, volumeServerId,
v.folders, v.folderMaxLimits, minFreeSpaces, diskTypes, folderTags,
*v.idxFolder,
volumeNeedleMapKind,
v.masters, constants.VolumePulsePeriod, *v.dataCenter, *v.rack,
v.whiteList,
*v.fixJpgOrientation, *v.readMode,
*v.compactionMBPerSecond,
*v.maintenanceMBPerSecond,
*v.fileSizeLimitMB,
int64(*v.concurrentUploadLimitMB)*1024*1024,
int64(*v.concurrentDownloadLimitMB)*1024*1024,
*v.inflightUploadDataTimeout,
*v.inflightDownloadDataTimeout,
*v.hasSlowRead,
*v.readBufferSizeMB,
*v.ldbTimeout,
)
// starting grpc server
grpcS := v.startGrpcService(volumeServer)
// starting public http server
var publicHttpDown httpdown.Server
if v.isSeparatedPublicPort() {
publicHttpDown = v.startPublicHttpService(publicVolumeMux)
if nil == publicHttpDown {
glog.Fatalf("start public http service failed")
}
}
// starting the cluster http server
clusterHttpServer, closeCert := v.startClusterHttpService(volumeMux)
grace.OnReload(volumeServer.LoadNewVolumes)
grace.OnReload(volumeServer.Reload)
stopChan := make(chan bool)
grace.OnInterrupt(func() {
fmt.Println("volume server has been killed")
// Stop heartbeats
if !volumeServer.StopHeartbeat() {
volumeServer.SetStopping()
glog.V(0).Infof("stop send heartbeat and wait %d seconds until shutdown ...", *v.preStopSeconds)
time.Sleep(time.Duration(*v.preStopSeconds) * time.Second)
}
shutdown(publicHttpDown, clusterHttpServer, grpcS, volumeServer)
if closeCert != nil {
closeCert()
}
stopChan <- true
})
if v.shutdownCtx != nil {
select {
case <-stopChan:
case <-v.shutdownCtx.Done():
shutdown(publicHttpDown, clusterHttpServer, grpcS, volumeServer)
if closeCert != nil {
closeCert()
}
}
} else {
select {
case <-stopChan:
}
}
}
func parseVolumeTags(tagsArg string, folderCount int) [][]string {
if folderCount <= 0 {
return nil
}
tagEntries := []string{}
if strings.TrimSpace(tagsArg) != "" {
tagEntries = strings.Split(tagsArg, ",")
}
folderTags := make([][]string, folderCount)
// If exactly one tag entry provided, replicate it to all folders
if len(tagEntries) == 1 {
normalized := util.NormalizeTagList(strings.Split(tagEntries[0], ":"))
for i := 0; i < folderCount; i++ {
folderTags[i] = append([]string(nil), normalized...)
}
} else {
// Otherwise, assign tags to folders that have explicit entries
for i := 0; i < folderCount; i++ {
if i < len(tagEntries) {
folderTags[i] = util.NormalizeTagList(strings.Split(tagEntries[i], ":"))
} else {
// Initialize remaining folders with empty tag slice
folderTags[i] = []string{}
}
}
}
return folderTags
}
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
func (v VolumeServerOptions) isSeparatedPublicPort() bool {
return *v.publicPort != *v.port
}
func (v VolumeServerOptions) startGrpcService(vs volume_server_pb.VolumeServerServer) *grpc.Server {
grpcPort := *v.portGrpc
grpcL, err := util.NewListener(util.JoinHostPort(*v.bindIp, grpcPort), 0)
if err != nil {
glog.Fatalf("failed to listen on grpc port %d: %v", grpcPort, err)
}
grpcS := pb.NewGrpcServer(security.LoadServerTLS(util.GetViper(), "grpc.volume"))
volume_server_pb.RegisterVolumeServerServer(grpcS, vs)
reflection.Register(grpcS)
go func() {
if err := grpcS.Serve(grpcL); err != nil {
glog.Fatalf("start gRPC service failed, %s", err)
}
}()
pb.ServeGrpcOnLocalSocket(grpcS, grpcPort)
return grpcS
}
func (v VolumeServerOptions) startPublicHttpService(handler http.Handler) httpdown.Server {
publicListeningAddress := util.JoinHostPort(*v.bindIp, *v.publicPort)
glog.V(0).Infoln("Start Seaweed volume server", version.Version(), "public at", publicListeningAddress)
publicListener, e := util.NewListener(publicListeningAddress, time.Duration(*v.idleConnectionTimeout)*time.Second)
if e != nil {
glog.Fatalf("Volume server listener error:%v", e)
}
pubHttp := httpdown.HTTP{StopTimeout: 5 * time.Minute, KillTimeout: 5 * time.Minute}
publicHttpDown := pubHttp.Serve(&http.Server{Handler: handler}, publicListener)
go func() {
if err := publicHttpDown.Wait(); err != nil {
glog.Errorf("public http down wait failed, %v", err)
}
}()
return publicHttpDown
}
// startClusterHttpService starts the volume cluster HTTP server and
// returns it along with a close func for the cert reloader's refresh
// goroutine (nil when HTTPS is disabled). The caller is responsible
// for invoking the close func on every shutdown path — both the
// SIGTERM/grace.OnInterrupt path and the shutdownCtx path used by
// mini/integration tests.
func (v VolumeServerOptions) startClusterHttpService(handler http.Handler) (httpdown.Server, func()) {
var (
certFile, keyFile string
)
if viper.GetString("https.volume.key") != "" {
certFile = viper.GetString("https.volume.cert")
keyFile = viper.GetString("https.volume.key")
}
listeningAddress := util.JoinHostPort(*v.bindIp, *v.port)
glog.V(0).Infof("Start Seaweed volume server %s at %s", version.Version(), listeningAddress)
listener, e := util.NewListener(listeningAddress, time.Duration(*v.idleConnectionTimeout)*time.Second)
if e != nil {
glog.Fatalf("Volume server listener error:%v", e)
}
httpDown := httpdown.HTTP{
KillTimeout: time.Minute,
StopTimeout: 30 * time.Second,
}
httpS := &http.Server{Handler: handler}
if viper.GetString("https.volume.ca") != "" {
clientCertFile := viper.GetString("https.volume.ca")
httpS.TLSConfig = security.LoadClientTLSHTTP(clientCertFile)
security.FixTlsConfig(util.GetViper(), httpS.TLSConfig)
}
var closeCert func()
if certFile != "" && keyFile != "" {
getCert, certProvider, err := security.NewReloadingServerCertificate(certFile, keyFile)
if err != nil {
glog.Fatalf("Volume server failed to load TLS certificate: %v", err)
}
closeCert = certProvider.Close
if httpS.TLSConfig == nil {
httpS.TLSConfig = &tls.Config{}
}
httpS.TLSConfig.GetCertificate = getCert
}
clusterHttpServer := httpDown.Serve(httpS, listener)
go func() {
if e := clusterHttpServer.Wait(); e != nil {
glog.Fatalf("Volume server fail to serve: %v", e)
}
}()
return clusterHttpServer, closeCert
}