gian/seaweedfs
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

quiet noisy 'shard X not found' log when EC shard lives on another server (#9316)

Browse Source 
quiet "shard X not found" log when EC shard lives on another server

When EC shards are spread across multiple volume servers, every read
that targets a shard not present locally was logging at V(0) with the
same wording as a real read failure. Under rclone-style traffic this
floods the logs and makes a healthy EC cluster look broken (issue #9310).

Distinguish the two cases with an errShardNotLocal sentinel: the
expected fall-through-to-remote path now logs at V(4); genuine local
read failures still log at V(0).
This commit is contained in:
Chris Lu 2026-05-04 15:30:22 -07:00 committed by GitHub
parent 6aa353716a
commit 57bb7b2f39
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 14 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
weed/storage/store_ec.go
View File

@ -2,6 +2,7 @@ package storage
import (
"context"
"errors"
"fmt"
"io"
"os"
@ -22,6 +23,12 @@ import (
"github.com/seaweedfs/seaweedfs/weed/storage/types"
)
// errShardNotLocal indicates that the requested EC shard is simply not
// stored on this volume server. It is expected during normal reads when
// shards are spread across multiple servers, so callers should not log
// it as an error.
var errShardNotLocal = errors.New("ec shard not on this server")
// FindEcShardTargetLocation returns the disk that should receive a new
// shard / index file for (collection, vid). The selection order is:
//
@ -317,7 +324,12 @@ func (s *Store) readOneEcShardInterval(needleId types.NeedleId, ecVolume *erasur
if err == nil {
return
}
glog.V(0).Infof("read local ec shard %d.%d offset %d: %v", ecVolume.VolumeId, shardId, actualOffset, err)
if errors.Is(err, errShardNotLocal) {
// expected when shards are spread across servers; fall through to remote read
glog.V(4).Infof("ec shard %d.%d not local, will try remote", ecVolume.VolumeId, shardId)
} else {
glog.V(0).Infof("read local ec shard %d.%d offset %d: %v", ecVolume.VolumeId, shardId, actualOffset, err)
}
ecVolume.ShardLocationsLock.RLock()
sourceDataNodes, hasShardIdLocation := ecVolume.ShardLocations[shardId]
@ -395,7 +407,7 @@ func (s *Store) cachedLookupEcShardLocations(ecVolume *erasure_coding.EcVolume)
func (s *Store) readLocalEcShardInterval(ecVolume *erasure_coding.EcVolume, shardId erasure_coding.ShardId, buf []byte, offset int64) error {
shard, found := ecVolume.FindEcVolumeShard(shardId)
if !found {
return fmt.Errorf("shard %d not found for volume %d", shardId, ecVolume.VolumeId)
return fmt.Errorf("shard %d for volume %d: %w", shardId, ecVolume.VolumeId, errShardNotLocal)
}
readBytes, err := shard.ReadAt(buf, offset)