995dfc4d5d
* chore: remove unreachable dead code across the codebase Remove ~50,000 lines of unreachable code identified by static analysis. Major removals: - weed/filer/redis_lua: entire unused Redis Lua filer store implementation - weed/wdclient/net2, resource_pool: unused connection/resource pool packages - weed/plugin/worker/lifecycle: unused lifecycle plugin worker - weed/s3api: unused S3 policy templates, presigned URL IAM, streaming copy, multipart IAM, key rotation, and various SSE helper functions - weed/mq/kafka: unused partition mapping, compression, schema, and protocol functions - weed/mq/offset: unused SQL storage and migration code - weed/worker: unused registry, task, and monitoring functions - weed/query: unused SQL engine, parquet scanner, and type functions - weed/shell: unused EC proportional rebalance functions - weed/storage/erasure_coding/distribution: unused distribution analysis functions - Individual unreachable functions removed from 150+ files across admin, credential, filer, iam, kms, mount, mq, operation, pb, s3api, server, shell, storage, topology, and util packages * fix(s3): reset shared memory store in IAM test to prevent flaky failure TestLoadIAMManagerFromConfig_EmptyConfigWithFallbackKey was flaky because the MemoryStore credential backend is a singleton registered via init(). Earlier tests that create anonymous identities pollute the shared store, causing LookupAnonymous() to unexpectedly return true. Fix by calling Reset() on the memory store before the test runs. * style: run gofmt on changed files * fix: restore KMS functions used by integration tests * fix(plugin): prevent panic on send to closed worker session channel The Plugin.sendToWorker method could panic with "send on closed channel" when a worker disconnected while a message was being sent. The race was between streamSession.close() closing the outgoing channel and sendToWorker writing to it concurrently. Add a done channel to streamSession that is closed before the outgoing channel, and check it in sendToWorker's select to safely detect closed sessions without panicking.
96 lines
3.6 KiB
Go
96 lines
3.6 KiB
Go
package offset
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"time"
|
|
|
|
"github.com/seaweedfs/seaweedfs/weed/filer"
|
|
"github.com/seaweedfs/seaweedfs/weed/filer_client"
|
|
"github.com/seaweedfs/seaweedfs/weed/pb/filer_pb"
|
|
"github.com/seaweedfs/seaweedfs/weed/pb/schema_pb"
|
|
"github.com/seaweedfs/seaweedfs/weed/util"
|
|
)
|
|
|
|
// FilerOffsetStorage implements OffsetStorage using SeaweedFS filer
|
|
// Stores offset data as files in the same directory structure as SMQ
|
|
// Path: /topics/{namespace}/{topic}/{version}/{partition}/checkpoint.offset
|
|
// The namespace and topic are derived from the actual partition information
|
|
type FilerOffsetStorage struct {
|
|
filerClientAccessor *filer_client.FilerClientAccessor
|
|
}
|
|
|
|
// NewFilerOffsetStorageWithAccessor creates a new filer-based offset storage using existing filer client accessor
|
|
func NewFilerOffsetStorageWithAccessor(filerClientAccessor *filer_client.FilerClientAccessor) *FilerOffsetStorage {
|
|
return &FilerOffsetStorage{
|
|
filerClientAccessor: filerClientAccessor,
|
|
}
|
|
}
|
|
|
|
// SaveCheckpoint saves the checkpoint for a partition
|
|
// Stores as: /topics/{namespace}/{topic}/{version}/{partition}/checkpoint.offset
|
|
func (f *FilerOffsetStorage) SaveCheckpoint(namespace, topicName string, partition *schema_pb.Partition, offset int64) error {
|
|
partitionDir := f.getPartitionDir(namespace, topicName, partition)
|
|
fileName := "checkpoint.offset"
|
|
|
|
// Use SMQ's 8-byte offset format
|
|
offsetBytes := make([]byte, 8)
|
|
util.Uint64toBytes(offsetBytes, uint64(offset))
|
|
|
|
return f.filerClientAccessor.WithFilerClient(false, func(client filer_pb.SeaweedFilerClient) error {
|
|
return filer.SaveInsideFiler(client, partitionDir, fileName, offsetBytes)
|
|
})
|
|
}
|
|
|
|
// LoadCheckpoint loads the checkpoint for a partition
|
|
func (f *FilerOffsetStorage) LoadCheckpoint(namespace, topicName string, partition *schema_pb.Partition) (int64, error) {
|
|
partitionDir := f.getPartitionDir(namespace, topicName, partition)
|
|
fileName := "checkpoint.offset"
|
|
|
|
var offset int64 = -1
|
|
err := f.filerClientAccessor.WithFilerClient(false, func(client filer_pb.SeaweedFilerClient) error {
|
|
data, err := filer.ReadInsideFiler(context.Background(), client, partitionDir, fileName)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
if len(data) != 8 {
|
|
return fmt.Errorf("invalid checkpoint file format: expected 8 bytes, got %d", len(data))
|
|
}
|
|
offset = int64(util.BytesToUint64(data))
|
|
return nil
|
|
})
|
|
|
|
if err != nil {
|
|
return -1, err
|
|
}
|
|
|
|
return offset, nil
|
|
}
|
|
|
|
// GetHighestOffset returns the highest offset stored for a partition
|
|
// For filer storage, this is the same as the checkpoint since we don't store individual records
|
|
func (f *FilerOffsetStorage) GetHighestOffset(namespace, topicName string, partition *schema_pb.Partition) (int64, error) {
|
|
return f.LoadCheckpoint(namespace, topicName, partition)
|
|
}
|
|
|
|
// Reset clears all data for testing
|
|
func (f *FilerOffsetStorage) Reset() error {
|
|
// For testing, we could delete all offset files, but this is dangerous
|
|
// Instead, just return success - individual tests should clean up their own data
|
|
return nil
|
|
}
|
|
|
|
// Helper methods
|
|
|
|
// getPartitionDir returns the directory path for a partition following SMQ convention
|
|
// Format: /topics/{namespace}/{topic}/{version}/{partition}
|
|
func (f *FilerOffsetStorage) getPartitionDir(namespace, topicName string, partition *schema_pb.Partition) string {
|
|
// Generate version from UnixTimeNs
|
|
version := time.Unix(0, partition.UnixTimeNs).UTC().Format("v2006-01-02-15-04-05")
|
|
|
|
// Generate partition range string
|
|
partitionRange := fmt.Sprintf("%04d-%04d", partition.RangeStart, partition.RangeStop)
|
|
|
|
return fmt.Sprintf("%s/%s/%s/%s/%s", filer.TopicsDir, namespace, topicName, version, partitionRange)
|
|
}
|