aboutsummaryrefslogtreecommitdiff
path: root/weed/command
diff options
context:
space:
mode:
Diffstat (limited to 'weed/command')
-rw-r--r--weed/command/command.go1
-rw-r--r--weed/command/filer.go1
-rw-r--r--weed/command/filer_remote_sync.go219
-rw-r--r--weed/command/filer_replication.go6
-rw-r--r--weed/command/imports.go31
5 files changed, 251 insertions, 7 deletions
diff --git a/weed/command/command.go b/weed/command/command.go
index 9ae93fe61..02de2bd35 100644
--- a/weed/command/command.go
+++ b/weed/command/command.go
@@ -21,6 +21,7 @@ var Commands = []*Command{
cmdFilerCopy,
cmdFilerMetaBackup,
cmdFilerMetaTail,
+ cmdFilerRemoteSynchronize,
cmdFilerReplicate,
cmdFilerSynchronize,
cmdFix,
diff --git a/weed/command/filer.go b/weed/command/filer.go
index 4fd2f9c72..ddee0852c 100644
--- a/weed/command/filer.go
+++ b/weed/command/filer.go
@@ -3,7 +3,6 @@ package command
import (
"fmt"
"net/http"
- _ "net/http/pprof"
"os"
"strconv"
"strings"
diff --git a/weed/command/filer_remote_sync.go b/weed/command/filer_remote_sync.go
new file mode 100644
index 000000000..8d176ce2a
--- /dev/null
+++ b/weed/command/filer_remote_sync.go
@@ -0,0 +1,219 @@
+package command
+
+import (
+ "fmt"
+ "github.com/chrislusf/seaweedfs/weed/filer"
+ "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/remote_storage"
+ "github.com/chrislusf/seaweedfs/weed/replication/source"
+ "github.com/chrislusf/seaweedfs/weed/security"
+ "github.com/chrislusf/seaweedfs/weed/util"
+ "github.com/golang/protobuf/proto"
+ "google.golang.org/grpc"
+ "strings"
+ "time"
+)
+
+type RemoteSyncOptions struct {
+ filerAddress *string
+ grpcDialOption grpc.DialOption
+ readChunkFromFiler *bool
+ debug *bool
+ timeAgo *time.Duration
+ dir *string
+}
+
+const (
+ RemoteSyncKeyPrefix = "remote.sync."
+)
+
+var _ = filer_pb.FilerClient(&RemoteSyncOptions{})
+
+func (option *RemoteSyncOptions) WithFilerClient(fn func(filer_pb.SeaweedFilerClient) error) error {
+ return pb.WithFilerClient(*option.filerAddress, option.grpcDialOption, func(client filer_pb.SeaweedFilerClient) error {
+ return fn(client)
+ })
+}
+func (option *RemoteSyncOptions) AdjustedUrl(location *filer_pb.Location) string {
+ return location.Url
+}
+
+var (
+ remoteSyncOptions RemoteSyncOptions
+)
+
+func init() {
+ cmdFilerRemoteSynchronize.Run = runFilerRemoteSynchronize // break init cycle
+ remoteSyncOptions.filerAddress = cmdFilerRemoteSynchronize.Flag.String("filer", "localhost:8888", "filer of the SeaweedFS cluster")
+ remoteSyncOptions.dir = cmdFilerRemoteSynchronize.Flag.String("dir", "/", "a mounted directory on filer")
+ remoteSyncOptions.readChunkFromFiler = cmdFilerRemoteSynchronize.Flag.Bool("filerProxy", false, "read file chunks from filer instead of volume servers")
+ remoteSyncOptions.debug = cmdFilerRemoteSynchronize.Flag.Bool("debug", false, "debug mode to print out filer updated remote files")
+ remoteSyncOptions.timeAgo = cmdFilerRemoteSynchronize.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\"")
+}
+
+var cmdFilerRemoteSynchronize = &Command{
+ UsageLine: "filer.remote.sync -filer=<filerHost>:<filerPort> -dir=/mount/s3_on_cloud",
+ Short: "resumeable continuously write back updates to remote storage if the directory is mounted to the remote storage",
+ Long: `resumeable continuously write back updates to remote storage if the directory is mounted to the remote storage
+
+ filer.remote.sync listens on filer update events.
+ If any mounted remote file is updated, it will fetch the updated content,
+ and write to the remote storage.
+`,
+}
+
+func runFilerRemoteSynchronize(cmd *Command, args []string) bool {
+
+ util.LoadConfiguration("security", false)
+ grpcDialOption := security.LoadClientTLS(util.GetViper(), "grpc.client")
+ remoteSyncOptions.grpcDialOption = grpcDialOption
+
+ // read filer remote storage mount mappings
+ mappings, readErr := filer.ReadMountMappings(grpcDialOption, *remoteSyncOptions.filerAddress)
+ if readErr != nil {
+ fmt.Printf("read mount mapping: %v", readErr)
+ return false
+ }
+
+ filerSource := &source.FilerSource{}
+ filerSource.DoInitialize(
+ *remoteSyncOptions.filerAddress,
+ pb.ServerToGrpcAddress(*remoteSyncOptions.filerAddress),
+ "/", // does not matter
+ *remoteSyncOptions.readChunkFromFiler,
+ )
+
+ var found bool
+ for dir, remoteStorageMountLocation := range mappings.Mappings {
+ if *remoteSyncOptions.dir == dir {
+ found = true
+ storageConf, readErr := filer.ReadRemoteStorageConf(grpcDialOption, *remoteSyncOptions.filerAddress, remoteStorageMountLocation.Name)
+ if readErr != nil {
+ fmt.Printf("read remote storage configuration for %s: %v", dir, readErr)
+ continue
+ }
+ fmt.Printf("synchronize %s to remote storage...\n", *remoteSyncOptions.dir)
+ if err := util.Retry("filer.remote.sync "+dir, func() error {
+ return followUpdatesAndUploadToRemote(&remoteSyncOptions, filerSource, dir, storageConf, remoteStorageMountLocation)
+ }); err != nil {
+ fmt.Printf("synchronize %s: %v\n", *remoteSyncOptions.dir, err)
+ }
+ break
+ }
+ }
+ if !found {
+ fmt.Printf("directory %s is not mounted to any remote storage\n", *remoteSyncOptions.dir)
+ return false
+ }
+
+ return true
+}
+
+func followUpdatesAndUploadToRemote(option *RemoteSyncOptions, filerSource *source.FilerSource, mountedDir string, remoteStorage *filer_pb.RemoteConf, remoteStorageMountLocation *filer_pb.RemoteStorageLocation) error {
+
+ dirHash := util.HashStringToLong(mountedDir)
+
+ // 1. specified by timeAgo
+ // 2. last offset timestamp for this directory
+ // 3. directory creation time
+ var lastOffsetTs time.Time
+ if *option.timeAgo == 0 {
+ mountedDirEntry, err := filer_pb.GetEntry(option, util.FullPath(mountedDir))
+ if err != nil {
+ return fmt.Errorf("lookup %s: %v", mountedDir, err)
+ }
+
+ lastOffsetTsNs, err := getOffset(option.grpcDialOption, *option.filerAddress, RemoteSyncKeyPrefix, int32(dirHash))
+ if err == nil && mountedDirEntry.Attributes.Crtime < lastOffsetTsNs/1000000 {
+ lastOffsetTs = time.Unix(0, lastOffsetTsNs)
+ glog.V(0).Infof("resume from %v", lastOffsetTs)
+ } else {
+ lastOffsetTs = time.Unix(mountedDirEntry.Attributes.Crtime, 0)
+ }
+ } else {
+ lastOffsetTs = time.Now().Add(-*option.timeAgo)
+ }
+
+ client, err := remote_storage.GetRemoteStorage(remoteStorage)
+ if err != nil {
+ return err
+ }
+
+ eachEntryFunc := func(resp *filer_pb.SubscribeMetadataResponse) error {
+ message := resp.EventNotification
+ if message.OldEntry == nil && message.NewEntry == nil {
+ return nil
+ }
+ if message.OldEntry == nil && message.NewEntry != nil {
+ if len(message.NewEntry.Chunks) == 0 {
+ return nil
+ }
+ fmt.Printf("create: %+v\n", resp)
+ dest := toRemoteStorageLocation(util.FullPath(mountedDir), util.NewFullPath(message.NewParentPath, message.NewEntry.Name), remoteStorageMountLocation)
+ reader := filer.NewChunkStreamReader(filerSource, message.NewEntry.Chunks)
+ return client.WriteFile(dest, message.NewEntry, reader)
+ }
+ if message.OldEntry != nil && message.NewEntry == nil {
+ fmt.Printf("delete: %+v\n", resp)
+ dest := toRemoteStorageLocation(util.FullPath(mountedDir), util.NewFullPath(resp.Directory, message.OldEntry.Name), remoteStorageMountLocation)
+ return client.DeleteFile(dest)
+ }
+ if message.OldEntry != nil && message.NewEntry != nil {
+ oldDest := toRemoteStorageLocation(util.FullPath(mountedDir), util.NewFullPath(resp.Directory, message.OldEntry.Name), remoteStorageMountLocation)
+ dest := toRemoteStorageLocation(util.FullPath(mountedDir), util.NewFullPath(message.NewParentPath, message.NewEntry.Name), remoteStorageMountLocation)
+ if resp.Directory == message.NewParentPath && message.OldEntry.Name == message.NewEntry.Name {
+ if isSameChunks(message.OldEntry.Chunks, message.NewEntry.Chunks) {
+ fmt.Printf("update meta: %+v\n", resp)
+ return client.UpdateFileMetadata(dest, message.NewEntry)
+ }
+ }
+ fmt.Printf("update: %+v\n", resp)
+ if err := client.DeleteFile(oldDest); err != nil {
+ return err
+ }
+ reader := filer.NewChunkStreamReader(filerSource, message.NewEntry.Chunks)
+ return client.WriteFile(dest, message.NewEntry, reader)
+ }
+
+ return nil
+ }
+
+ processEventFnWithOffset := pb.AddOffsetFunc(eachEntryFunc, 3*time.Second, func(counter int64, lastTsNs int64) error {
+ lastTime := time.Unix(0, lastTsNs)
+ glog.V(0).Infof("remote sync %s progressed to %v %0.2f/sec", *option.filerAddress, lastTime, float64(counter)/float64(3))
+ return setOffset(option.grpcDialOption, *option.filerAddress, RemoteSyncKeyPrefix, int32(dirHash), lastTsNs)
+ })
+
+ return pb.FollowMetadata(*option.filerAddress, option.grpcDialOption,
+ "filer.remote.sync", mountedDir, lastOffsetTs.UnixNano(), 0, processEventFnWithOffset, false)
+}
+
+func toRemoteStorageLocation(mountDir, sourcePath util.FullPath, remoteMountLocation *filer_pb.RemoteStorageLocation) *filer_pb.RemoteStorageLocation {
+ var dest string
+ source := string(sourcePath[len(mountDir):])
+ if strings.HasSuffix(remoteMountLocation.Path, "/") {
+ dest = remoteMountLocation.Path + source[1:]
+ } else {
+ dest = remoteMountLocation.Path + source
+ }
+ return &filer_pb.RemoteStorageLocation{
+ Name: remoteMountLocation.Name,
+ Bucket: remoteMountLocation.Bucket,
+ Path: dest,
+ }
+}
+
+func isSameChunks(a, b []*filer_pb.FileChunk) bool {
+ if len(a) != len(b) {
+ return false
+ }
+ for i := 0; i < len(a); i++ {
+ x, y := a[i], b[i]
+ if !proto.Equal(x, y) {
+ return false
+ }
+ }
+ return true
+}
diff --git a/weed/command/filer_replication.go b/weed/command/filer_replication.go
index 885c95540..bf0a3e140 100644
--- a/weed/command/filer_replication.go
+++ b/weed/command/filer_replication.go
@@ -7,12 +7,6 @@ import (
"github.com/chrislusf/seaweedfs/weed/glog"
"github.com/chrislusf/seaweedfs/weed/replication"
"github.com/chrislusf/seaweedfs/weed/replication/sink"
- _ "github.com/chrislusf/seaweedfs/weed/replication/sink/azuresink"
- _ "github.com/chrislusf/seaweedfs/weed/replication/sink/b2sink"
- _ "github.com/chrislusf/seaweedfs/weed/replication/sink/filersink"
- _ "github.com/chrislusf/seaweedfs/weed/replication/sink/gcssink"
- _ "github.com/chrislusf/seaweedfs/weed/replication/sink/localsink"
- _ "github.com/chrislusf/seaweedfs/weed/replication/sink/s3sink"
"github.com/chrislusf/seaweedfs/weed/replication/sub"
"github.com/chrislusf/seaweedfs/weed/util"
)
diff --git a/weed/command/imports.go b/weed/command/imports.go
new file mode 100644
index 000000000..d7ade1379
--- /dev/null
+++ b/weed/command/imports.go
@@ -0,0 +1,31 @@
+package command
+
+import (
+ _ "net/http/pprof"
+
+ _ "github.com/chrislusf/seaweedfs/weed/remote_storage/s3"
+
+ _ "github.com/chrislusf/seaweedfs/weed/replication/sink/azuresink"
+ _ "github.com/chrislusf/seaweedfs/weed/replication/sink/b2sink"
+ _ "github.com/chrislusf/seaweedfs/weed/replication/sink/filersink"
+ _ "github.com/chrislusf/seaweedfs/weed/replication/sink/gcssink"
+ _ "github.com/chrislusf/seaweedfs/weed/replication/sink/localsink"
+ _ "github.com/chrislusf/seaweedfs/weed/replication/sink/s3sink"
+
+ _ "github.com/chrislusf/seaweedfs/weed/filer/cassandra"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/elastic/v7"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/etcd"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/hbase"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/leveldb"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/leveldb2"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/leveldb3"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/mongodb"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/mysql"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/mysql2"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/postgres"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/postgres2"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/redis"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/redis2"
+ _ "github.com/chrislusf/seaweedfs/weed/filer/sqlite"
+
+) \ No newline at end of file
generated by cgit v1.2.3 (git 2.25.1) at 2025-12-21 20:48:24 +0000