mirror of
https://github.com/seaweedfs/seaweedfs.git
synced 2026-09-11 00:50:43 +02:00
* ec: add EC bitrot checksum protobuf EcBitrotProtection/EcShardChecksums/ChecksumAlgorithm sidecar messages, copy_ecsum_file and unsafe_ignore_sidecar fields, and a CHECKSUM scrub mode. * ec: bitrot checksum sidecar format, validation, and per-volume load Per-shard CRC32C block checksums in an optional <base>.ecsum sidecar with a self-integrity header; validation, rolling builder, backfill primitive, and EcVolume load on mount + removal on destroy. * ec: capture per-shard checksums at encode; verify-and-exclude on rebuild WriteEcFilesWithContext returns the protection computed inline during encoding. generateMissingEcFiles verifies present inputs against the sidecar, excludes corrupt ones, regenerates in place, and re-verifies; fail-closed unless unsafe_ignore_sidecar, removing all generated outputs on failure. * ec: read-only checksum scrub with Reed-Solomon arbiter ChecksumScrub verifies each local shard against the sidecar and reconstructs flagged shards from the clean shards so stale-sidecar false positives are not reported. Wired to the gRPC CHECKSUM mode and ec.scrub -mode checksum. * ec: server-side bitrot sidecar write, copy, cleanup, and opportunistic backfill Write .ecsum at fresh encode; propagate it with copy_ecsum_file (tolerant); remove it on full delete and decode; rebuild honors unsafe_ignore_sidecar and opportunistically backfills a sidecar when all shards are reachable. * ec: volume server bitrot config flags -ec.bitrotChecksum (default on) and -ec.bitrotBlockSizeMB (default 16). * fix(ec_bitrot): bound -ec.bitrotBlockSizeMB before the int64 multiply Validate the MiB value is in [1, 1024] before multiplying by 1 MiB, so a huge flag value cannot overflow int64 and slip past the power-of-two check, and a block size cannot collapse a sidecar to a few oversized blocks. * fix(ec_bitrot): distribute the .ecsum sidecar from the worker encode path The worker EC encode wrote the generation-0 sidecar locally but never added it to shardFiles, so DistributeEcShards never shipped it and the distributed holders came up unprotected. Append it to shardFiles and map the ecsum shard type to its extension in the sender so it travels with the shards. * fix(ec_bitrot): remove orphaned sidecars when the generation is gone Gate sidecar removal on existingShardCount==0 alone rather than also requiring a stray .ecx. A sidecar whose shards have all been deleted is orphaned and must be removed even when no .ecx remains, or it leaks. .ecx/.ecj/.vif removal stays gated on hasEcxFile as before. * fix(ec_bitrot): do not fold checksum blocks scanned into TotalFiles ChecksumScrub's first return is blocks scanned, not files. Discard it so the scrub response's TotalFiles (a needle/file count) is not inflated by the block count for CHECKSUM mode. * test(ec_bitrot): clean up generated .ecsum sidecars in removeGeneratedFiles * fix(ec_bitrot): reject an oversized sidecar payload before the uint32 cast The header stores payload_len as a uint32; bound the payload before the conversion so a pathological manifest cannot truncate the length field and corrupt the sidecar. A real manifest is a few KB, so this never trips. * fix(ec_bitrot): cap -ec.bitrotBlockSizeMB at 64 MiB The block size becomes the per-shard scratch buffer the scrub/backfill path allocates, so an over-large value (e.g. 1 GiB) is a memory hazard per concurrent scrub worker. Lower the upper bound from 1024 to 64 MiB. * fix(ec_bitrot): add -ecUnsafeIgnoreSidecar to weed tool fix -ecx The -ecx recovery path reconstructs missing shards via RebuildEcFilesWithContext, which fails closed on a malformed/stale .ecsum. Without an override flag an operator could not complete the rebuild without manually deleting the sidecar. Expose -ecUnsafeIgnoreSidecar (default false) and thread it through. * fix(ec_bitrot): bound sidecar payload with a direct int constant; drop readFull Guard len(payload) against a plain int constant (1 GiB) before the allocation instead of a uint64 MaxUint32 compare, so the allocation-size value is provably bounded (clears the CodeQL overflow alert) and the math import is no longer needed. Inline os.File.ReadAt with io.EOF handling in verifyShardFileBlocks and remove the now-redundant readFull helper (os.File.ReadAt fills the slice or errors). * test(ec_bitrot): use slices.Contains instead of a hand-rolled containsU32 * refactor(ec): fold the EcFiles WithContext variants into the base functions RebuildEcFiles now takes the *ECContext directly (nil => derive from .vif as before) and WriteEcFiles takes it too (nil => default), removing the parallel RebuildEcFilesWithContext / WriteEcFilesWithContext names. Callers that had an explicit context drop the WithContext suffix; the default-context callers pass nil. No behavior change. * refactor(ec): pass BackgroundECContext instead of nil to Write/RebuildEcFiles Add a non-nil BackgroundECContext placeholder (analogous to context.Background()) and have callers with no specific layout pass it instead of a nil *ECContext. WriteEcFiles resolves a zero/background context to the default ratio and RebuildEcFiles resolves it from the .vif, so behavior is unchanged. * fix(ec_bitrot): make BackgroundECContext a func; RebuildEcFiles fails closed on bad .vif - BackgroundECContext is now a function returning a fresh *ECContext, so callers cannot mutate a shared singleton or race on it (and it mirrors context.Background, which is also a function). - RebuildEcFiles now propagates the MaybeLoadVolumeInfo error: a present-but- unreadable .vif fails closed instead of silently rebuilding with the default ratio (which would corrupt a custom-ratio volume). Pass an explicit ctx to override.
394 lines
11 KiB
Go
394 lines
11 KiB
Go
package erasure_coding
|
|
|
|
import (
|
|
"context"
|
|
"errors"
|
|
"fmt"
|
|
"io"
|
|
"os"
|
|
"reflect"
|
|
"strings"
|
|
|
|
"github.com/seaweedfs/seaweedfs/weed/glog"
|
|
"github.com/seaweedfs/seaweedfs/weed/operation"
|
|
"github.com/seaweedfs/seaweedfs/weed/pb"
|
|
"github.com/seaweedfs/seaweedfs/weed/pb/volume_server_pb"
|
|
"github.com/seaweedfs/seaweedfs/weed/pb/worker_pb"
|
|
"google.golang.org/grpc"
|
|
)
|
|
|
|
func ensureLogger(logger logger) logger {
|
|
if logger == nil {
|
|
return &glogFallbackLogger{}
|
|
}
|
|
return logger
|
|
}
|
|
|
|
type logger interface {
|
|
Info(string, ...interface{})
|
|
Warning(string, ...interface{})
|
|
Error(string, ...interface{})
|
|
}
|
|
|
|
type withFieldLogger interface {
|
|
WithFields(map[string]interface{}) logger
|
|
}
|
|
|
|
func withFields(log logger, fields map[string]interface{}) logger {
|
|
if fields == nil {
|
|
return log
|
|
}
|
|
if wf, ok := log.(withFieldLogger); ok {
|
|
return wf.WithFields(fields)
|
|
}
|
|
|
|
val := reflect.ValueOf(log)
|
|
method := val.MethodByName("WithFields")
|
|
if !method.IsValid() {
|
|
return log
|
|
}
|
|
methodType := method.Type()
|
|
if methodType.NumIn() != 1 || methodType.NumOut() != 1 {
|
|
return log
|
|
}
|
|
argType := methodType.In(0)
|
|
mapType := reflect.TypeOf(map[string]interface{}{})
|
|
var arg reflect.Value
|
|
if mapType.AssignableTo(argType) {
|
|
arg = reflect.ValueOf(fields)
|
|
} else if mapType.ConvertibleTo(argType) {
|
|
arg = reflect.ValueOf(fields).Convert(argType)
|
|
} else {
|
|
return log
|
|
}
|
|
result := method.Call([]reflect.Value{arg})[0]
|
|
if result.IsValid() && result.CanInterface() {
|
|
if enhanced, ok := result.Interface().(logger); ok {
|
|
return enhanced
|
|
}
|
|
}
|
|
return log
|
|
}
|
|
|
|
type glogFallbackLogger struct{}
|
|
|
|
func (g *glogFallbackLogger) Info(msg string, args ...interface{}) {
|
|
if len(args) > 0 {
|
|
glog.Infof(msg, args...)
|
|
return
|
|
}
|
|
glog.Info(msg)
|
|
}
|
|
|
|
func (g *glogFallbackLogger) Warning(msg string, args ...interface{}) {
|
|
if len(args) > 0 {
|
|
glog.Warningf(msg, args...)
|
|
return
|
|
}
|
|
glog.Warning(msg)
|
|
}
|
|
|
|
func (g *glogFallbackLogger) Error(msg string, args ...interface{}) {
|
|
if len(args) > 0 {
|
|
glog.Errorf(msg, args...)
|
|
return
|
|
}
|
|
glog.Error(msg)
|
|
}
|
|
|
|
// DistributeEcShards distributes locally generated EC shards to destination servers.
|
|
// Returns the shard assignment map used for mounting.
|
|
func DistributeEcShards(volumeID uint32, collection string, targets []*worker_pb.TaskTarget, shardFiles map[string]string, dialOption grpc.DialOption, logger logger) (map[string][]string, error) {
|
|
if len(targets) == 0 {
|
|
return nil, fmt.Errorf("no targets specified for EC shard distribution")
|
|
}
|
|
|
|
if len(shardFiles) == 0 {
|
|
return nil, fmt.Errorf("no shard files available for distribution")
|
|
}
|
|
|
|
log := ensureLogger(logger)
|
|
|
|
shardAssignment := make(map[string][]string)
|
|
// node → shardType → planner-assigned diskID. Metadata files (ecx/ecj/vif)
|
|
// inherit the first data shard's disk so they land next to the shards.
|
|
shardDisks := make(map[string]map[string]uint32)
|
|
|
|
for _, target := range targets {
|
|
if len(target.ShardIds) == 0 {
|
|
continue
|
|
}
|
|
|
|
var assignedShards []string
|
|
for _, shardId := range target.ShardIds {
|
|
shardType := fmt.Sprintf("ec%02d", shardId)
|
|
assignedShards = append(assignedShards, shardType)
|
|
}
|
|
|
|
if len(assignedShards) > 0 {
|
|
if _, hasEcx := shardFiles["ecx"]; hasEcx {
|
|
assignedShards = append(assignedShards, "ecx")
|
|
}
|
|
if _, hasEcj := shardFiles["ecj"]; hasEcj {
|
|
assignedShards = append(assignedShards, "ecj")
|
|
}
|
|
if _, hasVif := shardFiles["vif"]; hasVif {
|
|
assignedShards = append(assignedShards, "vif")
|
|
}
|
|
if _, hasEcsum := shardFiles["ecsum"]; hasEcsum {
|
|
assignedShards = append(assignedShards, "ecsum")
|
|
}
|
|
}
|
|
|
|
if shardDisks[target.Node] == nil {
|
|
shardDisks[target.Node] = make(map[string]uint32)
|
|
}
|
|
for _, shardType := range assignedShards {
|
|
if _, already := shardDisks[target.Node][shardType]; !already {
|
|
shardDisks[target.Node][shardType] = target.DiskId
|
|
}
|
|
}
|
|
|
|
existing := shardAssignment[target.Node]
|
|
if len(existing) == 0 {
|
|
shardAssignment[target.Node] = assignedShards
|
|
continue
|
|
}
|
|
|
|
seen := make(map[string]struct{}, len(existing))
|
|
for _, shard := range existing {
|
|
seen[shard] = struct{}{}
|
|
}
|
|
for _, shard := range assignedShards {
|
|
if _, ok := seen[shard]; ok {
|
|
continue
|
|
}
|
|
seen[shard] = struct{}{}
|
|
existing = append(existing, shard)
|
|
}
|
|
shardAssignment[target.Node] = existing
|
|
}
|
|
|
|
if len(shardAssignment) == 0 {
|
|
return nil, fmt.Errorf("no shard assignments found from planning phase")
|
|
}
|
|
|
|
for destNode, assignedShards := range shardAssignment {
|
|
withFields(log, map[string]interface{}{
|
|
"destination": destNode,
|
|
"assigned_shards": len(assignedShards),
|
|
"shard_types": assignedShards,
|
|
}).Info("Starting shard distribution to destination server")
|
|
|
|
var transferredBytes int64
|
|
for _, shardType := range assignedShards {
|
|
filePath, exists := shardFiles[shardType]
|
|
if !exists {
|
|
return nil, fmt.Errorf("shard file %s not found for destination %s", shardType, destNode)
|
|
}
|
|
|
|
if info, err := os.Stat(filePath); err == nil {
|
|
transferredBytes += info.Size()
|
|
withFields(log, map[string]interface{}{
|
|
"destination": destNode,
|
|
"shard_type": shardType,
|
|
"file_path": filePath,
|
|
"size_bytes": info.Size(),
|
|
"size_kb": float64(info.Size()) / 1024,
|
|
}).Info("Starting shard file transfer")
|
|
}
|
|
|
|
diskID := uint32(0)
|
|
if byShard, ok := shardDisks[destNode]; ok {
|
|
diskID = byShard[shardType]
|
|
}
|
|
if err := sendShardFileToDestination(volumeID, collection, dialOption, destNode, diskID, filePath, shardType); err != nil {
|
|
return nil, fmt.Errorf("failed to send %s to %s: %w", shardType, destNode, err)
|
|
}
|
|
|
|
withFields(log, map[string]interface{}{
|
|
"destination": destNode,
|
|
"shard_type": shardType,
|
|
}).Info("Shard file transfer completed")
|
|
}
|
|
|
|
withFields(log, map[string]interface{}{
|
|
"destination": destNode,
|
|
"shards_transferred": len(assignedShards),
|
|
"total_bytes": transferredBytes,
|
|
"total_mb": float64(transferredBytes) / (1024 * 1024),
|
|
}).Info("All shards distributed to destination server")
|
|
}
|
|
|
|
glog.V(1).Infof("Successfully distributed EC shards to %d destinations", len(shardAssignment))
|
|
return shardAssignment, nil
|
|
}
|
|
|
|
// MountEcShards mounts EC shards on destination servers using an assignment map.
|
|
// sourceDiskType is forwarded to VolumeEcShardsMount so the resulting EC volume
|
|
// reports under the source's disk type rather than the destination's (#9423).
|
|
func MountEcShards(volumeID uint32, collection string, shardAssignment map[string][]string, sourceDiskType string, dialOption grpc.DialOption, logger logger) error {
|
|
if shardAssignment == nil {
|
|
return fmt.Errorf("shard assignment not available for mounting")
|
|
}
|
|
|
|
log := ensureLogger(logger)
|
|
|
|
var mountErrors []error
|
|
for destNode, assignedShards := range shardAssignment {
|
|
var shardIds []uint32
|
|
var metadataFiles []string
|
|
|
|
for _, shardType := range assignedShards {
|
|
if strings.HasPrefix(shardType, "ec") && len(shardType) == 4 {
|
|
var shardId uint32
|
|
if _, err := fmt.Sscanf(shardType[2:], "%d", &shardId); err == nil {
|
|
shardIds = append(shardIds, shardId)
|
|
}
|
|
} else {
|
|
metadataFiles = append(metadataFiles, shardType)
|
|
}
|
|
}
|
|
|
|
withFields(log, map[string]interface{}{
|
|
"destination": destNode,
|
|
"shard_ids": shardIds,
|
|
"shard_count": len(shardIds),
|
|
"metadata_files": metadataFiles,
|
|
}).Info("Starting EC shard mount operation")
|
|
|
|
if len(shardIds) == 0 {
|
|
withFields(log, map[string]interface{}{
|
|
"destination": destNode,
|
|
"metadata_files": metadataFiles,
|
|
}).Info("No EC shards to mount (only metadata files)")
|
|
continue
|
|
}
|
|
|
|
err := operation.WithVolumeServerClient(false, pb.ServerAddress(destNode), dialOption,
|
|
func(client volume_server_pb.VolumeServerClient) error {
|
|
_, mountErr := client.VolumeEcShardsMount(context.Background(), &volume_server_pb.VolumeEcShardsMountRequest{
|
|
VolumeId: volumeID,
|
|
Collection: collection,
|
|
ShardIds: shardIds,
|
|
SourceDiskType: sourceDiskType,
|
|
})
|
|
return mountErr
|
|
})
|
|
|
|
if err != nil {
|
|
mountErrors = append(mountErrors, fmt.Errorf("mount %s shards %v: %w", destNode, shardIds, err))
|
|
withFields(log, map[string]interface{}{
|
|
"destination": destNode,
|
|
"shard_ids": shardIds,
|
|
"error": err.Error(),
|
|
}).Error("Failed to mount EC shards")
|
|
} else {
|
|
withFields(log, map[string]interface{}{
|
|
"destination": destNode,
|
|
"shard_ids": shardIds,
|
|
"volume_id": volumeID,
|
|
"collection": collection,
|
|
}).Info("Successfully mounted EC shards")
|
|
}
|
|
}
|
|
|
|
if len(mountErrors) > 0 {
|
|
return errors.Join(mountErrors...)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// diskID=0 leaves disk placement to the server's auto-select.
|
|
func sendShardFileToDestination(volumeID uint32, collection string, dialOption grpc.DialOption, destServer string, diskID uint32, filePath, shardType string) error {
|
|
return operation.WithVolumeServerClient(false, pb.ServerAddress(destServer), dialOption,
|
|
func(client volume_server_pb.VolumeServerClient) error {
|
|
file, err := os.Open(filePath)
|
|
if err != nil {
|
|
return fmt.Errorf("failed to open shard file %s: %v", filePath, err)
|
|
}
|
|
defer file.Close()
|
|
|
|
fileInfo, err := file.Stat()
|
|
if err != nil {
|
|
return fmt.Errorf("failed to get file info for %s: %v", filePath, err)
|
|
}
|
|
|
|
var ext string
|
|
var shardId uint32
|
|
if shardType == "ecx" {
|
|
ext = ".ecx"
|
|
shardId = 0
|
|
} else if shardType == "ecj" {
|
|
ext = ".ecj"
|
|
shardId = 0
|
|
} else if shardType == "vif" {
|
|
ext = ".vif"
|
|
shardId = 0
|
|
} else if shardType == "ecsum" {
|
|
ext = BitrotSidecarExt
|
|
shardId = 0
|
|
} else if strings.HasPrefix(shardType, "ec") && len(shardType) == 4 {
|
|
ext = "." + shardType
|
|
fmt.Sscanf(shardType[2:], "%d", &shardId)
|
|
} else {
|
|
return fmt.Errorf("unknown shard type: %s", shardType)
|
|
}
|
|
|
|
stream, err := client.ReceiveFile(context.Background())
|
|
if err != nil {
|
|
return fmt.Errorf("failed to create receive stream: %v", err)
|
|
}
|
|
|
|
err = stream.Send(&volume_server_pb.ReceiveFileRequest{
|
|
Data: &volume_server_pb.ReceiveFileRequest_Info{
|
|
Info: &volume_server_pb.ReceiveFileInfo{
|
|
VolumeId: volumeID,
|
|
Ext: ext,
|
|
Collection: collection,
|
|
IsEcVolume: true,
|
|
ShardId: shardId,
|
|
FileSize: uint64(fileInfo.Size()),
|
|
DiskId: diskID,
|
|
},
|
|
},
|
|
})
|
|
if err != nil {
|
|
return fmt.Errorf("failed to send file info: %v", err)
|
|
}
|
|
|
|
buffer := make([]byte, 64*1024)
|
|
for {
|
|
n, readErr := file.Read(buffer)
|
|
if n > 0 {
|
|
err = stream.Send(&volume_server_pb.ReceiveFileRequest{
|
|
Data: &volume_server_pb.ReceiveFileRequest_FileContent{
|
|
FileContent: buffer[:n],
|
|
},
|
|
})
|
|
if err != nil {
|
|
return fmt.Errorf("failed to send file content: %v", err)
|
|
}
|
|
}
|
|
if readErr == io.EOF {
|
|
break
|
|
}
|
|
if readErr != nil {
|
|
return fmt.Errorf("failed to read file: %v", readErr)
|
|
}
|
|
}
|
|
|
|
resp, err := stream.CloseAndRecv()
|
|
if err != nil {
|
|
return fmt.Errorf("failed to close stream: %v", err)
|
|
}
|
|
|
|
if resp.Error != "" {
|
|
return fmt.Errorf("server error: %s", resp.Error)
|
|
}
|
|
|
|
glog.V(2).Infof("Successfully sent %s (%d bytes) to %s", shardType, resp.BytesWritten, destServer)
|
|
return nil
|
|
})
|
|
}
|