mirror of
https://github.com/seaweedfs/seaweedfs.git
synced 2026-09-20 13:30:46 +02:00
Wire low-level fencing primitives to master/VS control plane and CSI: - Proto: replica/rebuild address fields on assignment/info/response messages - Assignment queue: retain-until-confirmed (Peek+Confirm), stale epoch pruning - VS assignment receiver: processes assignments from HeartbeatResponse - BlockService replication: ProcessAssignments, deterministic ports (FNV hash) - Registry replica tracking: SetReplica/ClearReplica/SwapPrimaryReplica - CreateBlockVolume: primary + replica, enqueues assignments, single-copy mode - Failover: lease-aware promotion, deferred timers with cancellation on reconnect - ControllerPublish: returns fresh primary iSCSI address after failover - Recovery: recoverBlockVolumes drains pendingRebuilds, enqueues Rebuilding - Real integration tests on M02: failover address switch, rebuild data consistency, full lifecycle failover+rebuild (3 tests, all PASS) Review fixes (12 findings, 5 High, 5 Medium, 2 Low): - R1-1: AllocateBlockVolume returns replication ports - R1-2: setupPrimaryReplication starts rebuild server - R1-3: VS sends periodic block heartbeat for assignment confirmation - R2-F1: LastLeaseGrant set before Register (no stale-lease race) - R2-F2: Deferred promotion timers cancelled on VS reconnect - R2-F3: SwapPrimaryReplica uses RoleToWire instead of uint32(1) - R2-F4: DeleteBlockVolume deletes replica (best-effort) - R2-F5: SwapPrimaryReplica computes epoch atomically under lock - QA: SetReplica removes old replica from byServer index (BUG-QA-CP63-1) 126 CP6-3 tests (67 dev + 48 QA + 8 integration + 3 real). Cumulative Phase 6: 352 tests. All PASS. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
146 lines
4.5 KiB
Go
146 lines
4.5 KiB
Go
package storage
|
|
|
|
import (
|
|
"fmt"
|
|
"sync"
|
|
|
|
"github.com/seaweedfs/seaweedfs/weed/glog"
|
|
"github.com/seaweedfs/seaweedfs/weed/storage/blockvol"
|
|
)
|
|
|
|
// BlockVolumeStore manages block volumes (iSCSI-backed).
|
|
// It is a standalone component held by VolumeServer, not embedded into Store,
|
|
// to keep the existing Store codebase unchanged.
|
|
type BlockVolumeStore struct {
|
|
mu sync.RWMutex
|
|
volumes map[string]*blockvol.BlockVol // keyed by volume file path
|
|
diskTypes map[string]string // path -> disk type (e.g. "ssd")
|
|
}
|
|
|
|
// NewBlockVolumeStore creates a new block volume manager.
|
|
func NewBlockVolumeStore() *BlockVolumeStore {
|
|
return &BlockVolumeStore{
|
|
volumes: make(map[string]*blockvol.BlockVol),
|
|
diskTypes: make(map[string]string),
|
|
}
|
|
}
|
|
|
|
// AddBlockVolume opens and registers a block volume.
|
|
// diskType is metadata for heartbeat reporting (e.g. "ssd", "hdd").
|
|
func (bs *BlockVolumeStore) AddBlockVolume(path, diskType string, cfgs ...blockvol.BlockVolConfig) (*blockvol.BlockVol, error) {
|
|
bs.mu.Lock()
|
|
defer bs.mu.Unlock()
|
|
|
|
if _, ok := bs.volumes[path]; ok {
|
|
return nil, fmt.Errorf("block volume already registered: %s", path)
|
|
}
|
|
|
|
vol, err := blockvol.OpenBlockVol(path, cfgs...)
|
|
if err != nil {
|
|
return nil, fmt.Errorf("open block volume %s: %w", path, err)
|
|
}
|
|
|
|
bs.volumes[path] = vol
|
|
bs.diskTypes[path] = diskType
|
|
glog.V(0).Infof("block volume registered: %s (disk=%s)", path, diskType)
|
|
return vol, nil
|
|
}
|
|
|
|
// RemoveBlockVolume closes and unregisters a block volume.
|
|
func (bs *BlockVolumeStore) RemoveBlockVolume(path string) error {
|
|
bs.mu.Lock()
|
|
defer bs.mu.Unlock()
|
|
|
|
vol, ok := bs.volumes[path]
|
|
if !ok {
|
|
return fmt.Errorf("block volume not found: %s", path)
|
|
}
|
|
|
|
if err := vol.Close(); err != nil {
|
|
glog.Warningf("error closing block volume %s: %v", path, err)
|
|
}
|
|
delete(bs.volumes, path)
|
|
delete(bs.diskTypes, path)
|
|
glog.V(0).Infof("block volume removed: %s", path)
|
|
return nil
|
|
}
|
|
|
|
// GetBlockVolume returns a registered block volume by path.
|
|
func (bs *BlockVolumeStore) GetBlockVolume(path string) (*blockvol.BlockVol, bool) {
|
|
bs.mu.RLock()
|
|
defer bs.mu.RUnlock()
|
|
vol, ok := bs.volumes[path]
|
|
return vol, ok
|
|
}
|
|
|
|
// ListBlockVolumes returns the paths of all registered block volumes.
|
|
func (bs *BlockVolumeStore) ListBlockVolumes() []string {
|
|
bs.mu.RLock()
|
|
defer bs.mu.RUnlock()
|
|
paths := make([]string, 0, len(bs.volumes))
|
|
for p := range bs.volumes {
|
|
paths = append(paths, p)
|
|
}
|
|
return paths
|
|
}
|
|
|
|
// CollectBlockVolumeHeartbeat returns status for all registered
|
|
// block volumes, suitable for inclusion in a heartbeat message.
|
|
func (bs *BlockVolumeStore) CollectBlockVolumeHeartbeat() []blockvol.BlockVolumeInfoMessage {
|
|
bs.mu.RLock()
|
|
defer bs.mu.RUnlock()
|
|
msgs := make([]blockvol.BlockVolumeInfoMessage, 0, len(bs.volumes))
|
|
for path, vol := range bs.volumes {
|
|
msgs = append(msgs, blockvol.ToBlockVolumeInfoMessage(path, bs.diskTypes[path], vol))
|
|
}
|
|
return msgs
|
|
}
|
|
|
|
// WithVolume looks up a volume by path and calls fn while holding RLock.
|
|
// This prevents RemoveBlockVolume from closing the volume while fn runs
|
|
// (BUG-CP4B3-1: TOCTOU between GetBlockVolume and HandleAssignment).
|
|
func (bs *BlockVolumeStore) WithVolume(path string, fn func(*blockvol.BlockVol) error) error {
|
|
bs.mu.RLock()
|
|
defer bs.mu.RUnlock()
|
|
vol, ok := bs.volumes[path]
|
|
if !ok {
|
|
return fmt.Errorf("block volume not found: %s", path)
|
|
}
|
|
return fn(vol)
|
|
}
|
|
|
|
// ProcessBlockVolumeAssignments applies a batch of assignments from master.
|
|
// Returns a slice of errors parallel to the input (nil = success).
|
|
// Unknown volumes and invalid transitions are logged and returned as errors,
|
|
// but do not stop processing of remaining assignments.
|
|
func (bs *BlockVolumeStore) ProcessBlockVolumeAssignments(
|
|
assignments []blockvol.BlockVolumeAssignment,
|
|
) []error {
|
|
errs := make([]error, len(assignments))
|
|
for i, a := range assignments {
|
|
role := blockvol.RoleFromWire(a.Role)
|
|
ttl := blockvol.LeaseTTLFromWire(a.LeaseTtlMs)
|
|
if err := bs.WithVolume(a.Path, func(vol *blockvol.BlockVol) error {
|
|
return vol.HandleAssignment(a.Epoch, role, ttl)
|
|
}); err != nil {
|
|
errs[i] = err
|
|
glog.Warningf("assignment: volume %s epoch=%d role=%s: %v",
|
|
a.Path, a.Epoch, role, err)
|
|
}
|
|
}
|
|
return errs
|
|
}
|
|
|
|
// Close closes all block volumes.
|
|
func (bs *BlockVolumeStore) Close() {
|
|
bs.mu.Lock()
|
|
defer bs.mu.Unlock()
|
|
for path, vol := range bs.volumes {
|
|
if err := vol.Close(); err != nil {
|
|
glog.Warningf("error closing block volume %s: %v", path, err)
|
|
}
|
|
delete(bs.volumes, path)
|
|
}
|
|
glog.V(0).Infof("all block volumes closed")
|
|
}
|