75a6a34528d6f7f3bee64298a829117e9b703835
5 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Chris Lu
|
75a6a34528 |
dlm: resilient distributed locks via consistent hashing + backup replication (#8860)
* dlm: replace modulo hashing with consistent hash ring Introduce HashRing with virtual nodes (CRC32-based consistent hashing) to replace the modulo-based hashKeyToServer. When a filer node is removed, only keys that hashed to that node are remapped to the next server on the ring, leaving all other mappings stable. This is the foundation for backup replication — the successor on the ring is always the natural takeover node. * dlm: add Generation and IsBackup fields to Lock Lock now carries IsBackup (whether this node holds the lock as a backup replica) and Generation (a monotonic fencing token that increments on each fresh acquisition, stays the same on renewal). Add helper methods: AllLocks, PromoteLock, DemoteLock, InsertBackupLock, RemoveLock, GetLock. * dlm: add ReplicateLock RPC and generation/is_backup proto fields Add generation field to LockResponse for fencing tokens. Add generation and is_backup fields to Lock message. Add ReplicateLock RPC for primary-to-backup lock replication. Add ReplicateLockRequest/ReplicateLockResponse messages. * dlm: add async backup replication to DistributedLockManager Route lock/unlock via consistent hash ring's GetPrimaryAndBackup(). After a successful lock or unlock on the primary, asynchronously replicate the operation to the backup server via ReplicateFunc callback. Single-server deployments skip replication. * dlm: add ReplicateLock handler and backup-aware topology changes Add ReplicateLock gRPC handler for primary-to-backup replication. Revise OnDlmChangeSnapshot to handle three cases on topology change: - Promote backup locks when this node becomes primary - Demote primary locks when this node becomes backup - Transfer locks when this node is neither primary nor backup Wire up SetupDlmReplication during filer server initialization. * dlm: expose generation fencing token in lock client LiveLock now captures the generation from LockResponse and exposes it via Generation() method. Consumers can use this as a fencing token to detect stale lock holders. * dlm: update empty folder cleaner to use consistent hash ring Replace local modulo-based hashKeyToServer with LockRing.GetPrimary() which uses the shared consistent hash ring for folder ownership. * dlm: add unit tests for consistent hash ring Test basic operations, consistency on server removal (only keys from removed server move), backup-is-successor property (backup becomes new primary when primary is removed), and key distribution balance. * dlm: add integration tests for lock replication failure scenarios Test cases: - Primary crash with backup promotion (backup has valid token) - Backup crash with primary continuing - Both primary and backup crash (lock lost, re-acquirable) - Rolling restart across all nodes - Generation fencing token increments on new acquisition - Replication failure (primary still works independently) - Unlock replicates deletion to backup - Lock survives server addition (topology change) - Consistent hashing minimal disruption (only removed server's keys move) * dlm: address PR review findings 1. Causal replication ordering: Add per-lock sequence number (Seq) that increments on every mutation. Backup rejects incoming mutations with seq <= current seq, preventing stale async replications from overwriting newer state. Unlock replication also carries seq and is rejected if stale. 2. Demote-after-handoff: OnDlmChangeSnapshot now transfers the lock to the new primary first and only demotes to backup after a successful TransferLocks RPC. If the transfer fails, the lock stays as primary on this node. 3. SetSnapshot candidateServers leak: Replace the candidateServers map entirely instead of appending, so removed servers don't linger. 4. TransferLocks preserves Generation and Seq: InsertLock now accepts generation and seq parameters. After accepting a transferred lock, the receiving node re-replicates to its backup. 5. Rolling restart test: Add re-replication step after promotion and assert survivedCount > 0. Add TestDLM_StaleReplicationRejected. 6. Mixed-version upgrade note: Add comment on HashRing documenting that all filer nodes must be upgraded together. * dlm: serve renewals locally during transfer window on node join When a new node joins and steals hash ranges from surviving nodes, there's a window between ring update and lock transfer where the client gets redirected to a node that doesn't have the lock yet. Fix: if the ring says primary != self but we still hold the lock locally (non-backup, matching token), serve the renewal/unlock here rather than redirecting. The lock will be transferred by OnDlmChangeSnapshot, and subsequent requests will go to the new primary once the transfer completes. Add tests: - TestDLM_NodeDropAndJoin_OwnershipDisruption: measures disruption when a node drops and a new one joins (14/100 surviving-node locks disrupted, all handled by transfer logic) - TestDLM_RenewalDuringTransferWindow: verifies renewal succeeds on old primary during the transfer window * dlm: master-managed lock ring with stabilization batching The master now owns the lock ring membership. Instead of filers independently reacting to individual ClusterNodeUpdate add/remove events, the master: 1. Tracks filer membership in LockRingManager 2. Batches rapid changes with a 1-second stabilization timer (e.g., a node drop + join within 1 second → single ring update) 3. Broadcasts the complete ring snapshot atomically via the new LockRingUpdate message in KeepConnectedResponse Filers receive the ring as a complete snapshot and apply it via SetSnapshot, ensuring all filers converge to the same ring state without intermediate churn. This eliminates the double-churn problem where a rapid drop+join would fire two separate ring mutations, each triggering lock transfers and disrupting ownership on surviving nodes. * dlm: track ring version, reject stale updates, remove dead code SetSnapshot now takes a version parameter from the master. Stale updates (version < current) are rejected, preventing reordered messages from overwriting a newer ring state. Version 0 is always accepted for bootstrap. Remove AddServer/RemoveServer from LockRing — the ring is now exclusively managed by the master via SetSnapshot. Remove the candidateServers map that was only used by those methods. * dlm: fix SelectLocks data race, advance generation on backup insert - SelectLocks: change RLock to Lock since the function deletes map entries, which is a write operation and causes a data race under RLock. - InsertBackupLock: advance nextGeneration to at least the incoming generation so that after failover promotion, new lock acquisitions get a generation strictly greater than any replicated lock. - Bump replication failure log from V(1) to Warningf for production visibility. * dlm: fix SetSnapshot race, test reliability, timer edge cases - SetSnapshot: hold LockRing lock through both version update and Ring.SetServers() so they're atomic. Prevents a concurrent caller from seeing the new version but applying stale servers. - Transfer window test: search for a key that actually moves primary when filer4 joins, instead of relying on a fixed key that may not. - renewLock redirect: pass the existing token to the new primary instead of empty string, so redirected renewals work correctly. - scheduleBroadcast: check timer.Stop() return value. If the timer already fired, the callback picks up latest state. - FlushPending: only broadcast if timer.Stop() returns true (timer was still pending). If false, the callback is already running. - Fix test comment: "idempotent" → "accepted, state-changing". * dlm: use wall-clock nanoseconds for lock ring version The lock ring version was an in-memory counter that reset to 0 on master restart. A filer that had seen version 5 would reject version 1 from the restarted master. Fix: use time.Now().UnixNano() as the version. This survives master restarts without persistence — the restarted master produces a version greater than any pre-restart value. * dlm: treat expired lock owners as missing Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * dlm: reject stale lock transfers Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * dlm: order replication by generation Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * dlm: bootstrap lock ring on reconnect Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> --------- Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> |
||
|
Chris Lu
|
0761be58d3 |
fix(s3): preserve explicit directory markers during empty folder cleanup (#8831)
* fix(s3): preserve explicit directory markers during empty folder cleanup PR #8292 switched empty-folder cleanup from per-folder implicit checks to bucket-level policy, inadvertently dropping the check that preserved explicitly created directories (e.g., PUT /bucket/folder/). This caused user-created folders to be deleted when their last file was removed. Add IsDirectoryKeyObject check in executeCleanup to skip folders that have a MIME type set, matching the canonical pattern used throughout the S3 listing and delete handlers. * fix: handle ErrNotFound in IsDirectoryKeyObject for race safety Entry may be deleted between the emptiness check and the directory marker lookup. Treat not-found as false rather than propagating the error, avoiding unnecessary error logging in the cleanup path. * refactor: consolidate directory marker tests and tidy error handling - Combine two separate test functions into a table-driven test - Nest ErrNotFound check inside the err != nil block |
||
|
Chris Lu
|
b57429ef2e |
Switch empty-folder cleanup to bucket policy (#8292)
* Fix Spark _temporary cleanup and add issue #8285 regression test * Generalize empty folder cleanup for Spark temp artifacts * Revert synchronous folder pruning and add cleanup diagnostics * Add actionable empty-folder cleanup diagnostics * Fix Spark temp marker cleanup in async folder cleaner * Fix Spark temp cleanup with implicit directory markers * Keep explicit directory markers non-implicit * logging * more logs * Switch empty-folder cleanup to bucket policy * Seaweed-X-Amz-Allow-Empty-Folders * less logs * go vet * less logs * refactoring |
||
|
promalert
|
9012069bd7 |
chore: execute goimports to format the code (#7983)
* chore: execute goimports to format the code Signed-off-by: promalert <promalert@outlook.com> * goimports -w . --------- Signed-off-by: promalert <promalert@outlook.com> Co-authored-by: Chris Lu <chris.lu@gmail.com> |
||
|
Chris Lu
|
39ba19eea6 |
filer: async empty folder cleanup via metadata events (#7614)
* filer: async empty folder cleanup via metadata events Implements asynchronous empty folder cleanup when files are deleted in S3. Key changes: 1. EmptyFolderCleaner - New component that handles folder cleanup: - Uses consistent hashing (LockRing) to determine folder ownership - Each filer owns specific folders, avoiding duplicate cleanup work - Debounces delete events (10s delay) to batch multiple deletes - Caches rough folder counts to skip unnecessary checks - Cancels pending cleanup when new files are created - Handles both file and subdirectory deletions 2. Integration with metadata events: - Listens to both local and remote filer metadata events - Processes create/delete/rename events to track folder state - Only processes folders under /buckets/<bucket>/... 3. Removed synchronous empty folder cleanup from S3 handlers: - DeleteObjectHandler no longer calls DoDeleteEmptyParentDirectories - DeleteMultipleObjectsHandler no longer tracks/cleans directories - Cleanup now happens asynchronously via metadata events Benefits: - Non-blocking: S3 delete requests return immediately - Coordinated: Only one filer (the owner) cleans each folder - Efficient: Batching and caching reduce unnecessary checks - Event-driven: Folder deletion triggers parent folder check automatically * filer: add CleanupQueue data structure for deduplicated folder cleanup CleanupQueue uses a linked list for FIFO ordering and a hashmap for O(1) deduplication. Processing is triggered when: - Queue size reaches maxSize (default 1000), OR - Oldest item exceeds maxAge (default 10 minutes) Key features: - O(1) Add, Remove, Pop, Contains operations - Duplicate folders are ignored (keeps original position/time) - Testable with injectable time function - Thread-safe with mutex protection * filer: use CleanupQueue for empty folder cleanup Replace timer-per-folder approach with queue-based processing: - Use CleanupQueue for deduplication and ordered processing - Process queue when full (1000 items) or oldest item exceeds 10 minutes - Background processor checks queue every 10 seconds - Remove from queue on create events to cancel pending cleanup Benefits: - Bounded memory: queue has max size, not unlimited timers - Efficient: O(1) add/remove/contains operations - Batch processing: handle many folders efficiently - Better for high-volume delete scenarios * filer: CleanupQueue.Add moves duplicate to back with updated time When adding a folder that already exists in the queue: - Remove it from its current position - Add it to the back of the queue - Update the queue time to current time This ensures that folders with recent delete activity are processed later, giving more time for additional deletes to occur. * filer: CleanupQueue uses event time and inserts in sorted order Changes: - Add() now takes eventTime parameter instead of using current time - Insert items in time-sorted order (oldest at front) to handle out-of-order events - When updating duplicate with newer time, reposition to maintain sort order - Ignore updates with older time (keep existing later time) This ensures proper ordering when processing events from distributed filers where event arrival order may not match event occurrence order. * filer: remove unused CleanupQueue functions (SetNowFunc, GetAll) Removed test-only functions: - SetNowFunc: tests now use real time with past event times - GetAll: tests now use Pop() to verify order Kept functions used in production: - Peek: used in filer_notify_read.go - OldestAge: used in empty_folder_cleaner.go logging * filer: initialize cache entry on first delete/create event Previously, roughCount was only updated if the cache entry already existed, but entries were only created during executeCleanup. This meant delete/create events before the first cleanup didn't track the count. Now create the cache entry on first event, so roughCount properly tracks all changes from the start. * filer: skip adding to cleanup queue if roughCount > 0 If the cached roughCount indicates there are still items in the folder, don't bother adding it to the cleanup queue. This avoids unnecessary queue entries and reduces wasted cleanup checks. * filer: don't create cache entry on create event Only update roughCount if the folder is already being tracked. New folders don't need tracking until we see a delete event. * filer: move empty folder cleanup to its own package - Created weed/filer/empty_folder_cleanup package - Defined FilerOperations interface to break circular dependency - Added CountDirectoryEntries method to Filer - Exported IsUnderPath and IsUnderBucketPath helper functions * filer: make isUnderPath and isUnderBucketPath private These helpers are only used within the empty_folder_cleanup package. |