Files
seaweedfs/weed/operation/lookup.go
T
Chris LuandClaude Opus 5.5 5cf053c32e volume: read a manifest chunk from another replica when one fails (#12313)
* operation: read a manifest chunk from the next replica when one fails

ChunkedFileReader read each chunk from one random location and returned
the first error, so a chunk whose cached replica went down failed the
whole file although another replica had it. Try the chunk's locations in
turn (local-data replicas first) and resume each attempt at the byte the
previous one reached. A failed write to the caller still ends the read.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* operation: drop a chunk volume's cached locations after a failed read

A chunk read that failed on a replica left that replica in the
10-minute location cache, so later reads kept trying it first. Drop the
volume's cached locations after a failure, and when every cached
replica has failed, ask the master once more and try the locations it
now lists.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* operation: don't continue a partly sent chunk on another replica

A volume server streams a large needle before it has verified the
checksum and aborts the response on a mismatch, so bytes that already
reached the reader may be corrupt. Continuing such a chunk on another
replica would hand back a mix of bad and good bytes as a complete read.
Move to the next replica only while nothing of the chunk has been
written; a mid-body failure still fails the read, and the dropped cache
entry lets the next read pick a live replica.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* operation: treat a short chunk body as a failed replica read

A replica that ends a chunk's body cleanly before the chunk's size made
the read look successful, so the next replica was never tried and the
following chunks landed at the wrong offsets. Read exactly the chunk's
remaining length and report a short body as an error.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-01 16:31:22 -08:00

179 lines
6.0 KiB
Go

package operation
import (
"context"
"errors"
"fmt"
"math/rand/v2"
"strings"
"time"
"github.com/seaweedfs/seaweedfs/weed/pb"
"google.golang.org/grpc"
"github.com/seaweedfs/seaweedfs/weed/pb/master_pb"
)
type Location struct {
Url string `json:"url,omitempty"`
PublicUrl string `json:"publicUrl,omitempty"`
DataCenter string `json:"dataCenter,omitempty"`
GrpcPort int `json:"grpcPort,omitempty"`
DataInRemote bool `json:"dataInRemote,omitempty"`
ReadOnly bool `json:"readOnly,omitempty"`
ReadOnlyCanDelete bool `json:"readOnlyCanDelete,omitempty"`
}
func (l *Location) ServerAddress() pb.ServerAddress {
return pb.NewServerAddressWithGrpcPort(l.Url, l.GrpcPort)
}
type LookupResult struct {
VolumeOrFileId string `json:"volumeOrFileId,omitempty"`
Locations []Location `json:"locations,omitempty"`
Jwt string `json:"jwt,omitempty"`
Error string `json:"error,omitempty"`
NotFound bool `json:"-"`
}
func (lr *LookupResult) String() string {
return fmt.Sprintf("VolumeOrFileId:%s, Locations:%v, Error:%s", lr.VolumeOrFileId, lr.Locations, lr.Error)
}
var (
vc VidCache // caching of volume locations, re-check if after 10 minutes
)
// LookupFileId resolves a "<vid>,<cookie>" file id to one HTTP read URL,
// preferring a volume server whose replica holds the data locally over one
// backed by remote-tier storage. If no local replica is known the function
// falls back to a random remote replica. The returned jwt is the read
// authorization the master stamped on the volume; pass it through to the
// volume server on the read request.
func LookupFileId(masterFn GetMasterFn, grpcDialOption grpc.DialOption, fileId string) (fullUrl string, jwt string, err error) {
fullUrls, jwt, err := lookupFileIdUrls(masterFn, grpcDialOption, fileId)
if err != nil {
return "", jwt, err
}
return fullUrls[0], jwt, nil
}
// lookupFileIdUrls returns the read URLs of every replica of fileId, local
// replicas first, each group in random order.
func lookupFileIdUrls(masterFn GetMasterFn, grpcDialOption grpc.DialOption, fileId string) (fullUrls []string, jwt string, err error) {
parts := strings.Split(fileId, ",")
if len(parts) != 2 {
return nil, jwt, errors.New("Invalid fileId " + fileId)
}
lookup, lookupError := LookupVolumeId(masterFn, grpcDialOption, parts[0])
if lookupError != nil {
return nil, jwt, lookupError
}
if len(lookup.Locations) == 0 {
return nil, jwt, errors.New("File Not Found")
}
var localUrls, remoteUrls []string
for _, loc := range lookup.Locations {
fullUrl := "http://" + loc.Url + "/" + fileId
if loc.DataInRemote {
remoteUrls = append(remoteUrls, fullUrl)
} else {
localUrls = append(localUrls, fullUrl)
}
}
rand.Shuffle(len(localUrls), func(i, j int) { localUrls[i], localUrls[j] = localUrls[j], localUrls[i] })
rand.Shuffle(len(remoteUrls), func(i, j int) { remoteUrls[i], remoteUrls[j] = remoteUrls[j], remoteUrls[i] })
return append(localUrls, remoteUrls...), lookup.Jwt, nil
}
func LookupVolumeId(masterFn GetMasterFn, grpcDialOption grpc.DialOption, vid string) (*LookupResult, error) {
return LookupVolumeIdWithContext(context.Background(), masterFn, grpcDialOption, vid)
}
func LookupVolumeIdWithContext(ctx context.Context, masterFn GetMasterFn, grpcDialOption grpc.DialOption, vid string) (*LookupResult, error) {
results, err := LookupVolumeIdsWithContext(ctx, masterFn, grpcDialOption, []string{vid})
return results[vid], err
}
// InvalidateVolumeIdLocationCache drops cached locations for vid so the next lookup re-queries the master.
func InvalidateVolumeIdLocationCache(vid string) {
vc.Delete(vid)
}
// LookupVolumeIds find volume locations by cache and actual lookup
func LookupVolumeIds(masterFn GetMasterFn, grpcDialOption grpc.DialOption, vids []string, useCache ...bool) (map[string]*LookupResult, error) {
return LookupVolumeIdsWithContext(context.Background(), masterFn, grpcDialOption, vids, useCache...)
}
// LookupVolumeIdsWithContext is LookupVolumeIds bound to ctx, so a caller's
// cancellation (e.g. a dropped inbound request) also stops the master query.
func LookupVolumeIdsWithContext(ctx context.Context, masterFn GetMasterFn, grpcDialOption grpc.DialOption, vids []string, useCache ...bool) (map[string]*LookupResult, error) {
ret := make(map[string]*LookupResult)
var unknown_vids []string
cacheLocations := len(useCache) == 0 || useCache[0]
//check vid cache first
for _, vid := range vids {
if cacheLocations {
locations, cacheErr := vc.Get(vid)
if cacheErr == nil {
ret[vid] = &LookupResult{VolumeOrFileId: vid, Locations: locations}
continue
}
}
unknown_vids = append(unknown_vids, vid)
}
//return success if all volume ids are known
if len(unknown_vids) == 0 {
return ret, nil
}
//only query unknown_vids
err := WithMasterServerClient(ctx, false, masterFn(ctx), grpcDialOption, func(masterClient master_pb.SeaweedClient) error {
req := &master_pb.LookupVolumeRequest{
VolumeOrFileIds: unknown_vids,
}
resp, grpcErr := masterClient.LookupVolume(ctx, req)
if grpcErr != nil {
return grpcErr
}
//set newly checked vids to cache
for _, vidLocations := range resp.VolumeIdLocations {
var locations []Location
for _, loc := range vidLocations.Locations {
locations = append(locations, Location{
Url: loc.Url,
PublicUrl: loc.PublicUrl,
DataCenter: loc.DataCenter,
GrpcPort: int(loc.GrpcPort),
DataInRemote: loc.DataInRemote,
ReadOnly: loc.ReadOnly,
ReadOnlyCanDelete: loc.ReadOnlyCanDelete,
})
}
if cacheLocations && vidLocations.Error == "" {
vc.Set(vidLocations.VolumeOrFileId, locations, 10*time.Minute)
}
ret[vidLocations.VolumeOrFileId] = &LookupResult{
VolumeOrFileId: vidLocations.VolumeOrFileId,
Locations: locations,
Jwt: vidLocations.Auth,
Error: vidLocations.Error,
}
}
return nil
})
if err != nil {
return nil, err
}
return ret, nil
}