909 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>	2026-03-30 23:29:56 -07:00
Chris Lu	ced2236cc6	Adjust rename events metadata format (#8854) ... * rename metadata events * fix subscription filter to use NewEntry.Name for rename path matching The server-side subscription filter constructed the new path using OldEntry.Name instead of NewEntry.Name when checking if a rename event's destination matches the subscriber's path prefix. This could cause events to be incorrectly filtered when a rename changes the file name. * fix bucket events to handle rename of bucket directories onBucketEvents only checked IsCreate and IsDelete. A bucket directory rename via AtomicRenameEntry now emits a single rename event (both OldEntry and NewEntry non-nil), which matched neither check. Handle IsRename by deleting the old bucket and creating the new one. * fix replicator to handle rename events across directory boundaries Two issues fixed: 1. The replicator filtered events by checking if the key (old path) was under the source directory. Rename events now use the old path as key, so renames from outside into the watched directory were silently dropped. Now both old and new paths are checked, and cross-boundary renames are converted to create or delete. 2. NewParentPath was passed to the sink without remapping to the sink's target directory structure, causing the sink to write entries at the wrong location. Now NewParentPath is remapped alongside the key. * fix filer sync to handle rename events crossing directory boundaries The early directory-prefix filter only checked resp.Directory (old parent). Rename events now carry the old parent as Directory, so renames from outside the source path into it were dropped before reaching the existing cross-boundary handling logic. Check both old and new directories against sourcePath and excludePaths so the downstream old-key/new-key logic can properly convert these to create or delete operations. * fix metadata event path matching * fix metadata event consumers for rename targets * Fix replication rename target keys Logical rename events now reach replication sinks with distinct source and target paths.\n\nHandle non-filer sinks as delete-plus-create on the translated target key, and make the rename fallback path create at the translated target key too.\n\nAdd focused tests covering non-filer renames, filer rename updates, and the fallback path.\n\nCo-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * Fix filer sync rename path scoping Use directory-boundary matching instead of raw prefix checks when classifying source and target paths during filer sync.\n\nAlso apply excludePaths per side so renames across excluded boundaries downgrade cleanly to create/delete instead of being misclassified as in-scope updates.\n\nAdd focused tests for boundary matching and rename classification.\n\nCo-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * Fix replicator directory boundary checks Use directory-boundary matching instead of raw prefix checks when deciding whether a source or target path is inside the watched tree or an excluded subtree.\n\nThis prevents sibling paths such as /foo and /foobar from being misclassified during rename handling, and preserves the earlier rename-target-key fix.\n\nAdd focused tests for boundary matching and rename classification across sibling/excluded directories.\n\nCo-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * Fix etc-remote rename-out handling Use boundary-safe source/target directory membership when classifying metadata events under DirectoryEtcRemote.\n\nThis prevents rename-out events from being processed as config updates, while still treating them as removals where appropriate for the remote sync and remote gateway command paths.\n\nAdd focused tests for update/removal classification and sibling-prefix handling.\n\nCo-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * Defer rename events until commit Queue logical rename metadata events during atomic and streaming renames and publish them only after the transaction commits successfully.\n\nThis prevents subscribers from seeing delete or logical rename events for operations that later fail during delete or commit.\n\nAlso serialize notification.Queue swaps in rename tests and add failure-path coverage.\n\nCo-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * Skip descendant rename target lookups Avoid redundant target lookups during recursive directory renames once the destination subtree is known absent.\n\nThe recursive move path now inserts known-absent descendants directly, and the test harness exercises prefixed directory listing so the optimization is covered by a directory rename regression test.\n\nCo-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * Tighten rename review tests Return filer_pb.ErrNotFound from the bucket tracking store test stub so it follows the FilerStore contract, and add a webhook filter case for same-name renames across parent directories.\n\nCo-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> * fix HardLinkId format verb in InsertEntryKnownAbsent error HardLinkId is a byte slice. %d prints each byte as a decimal number which is not useful for an identifier. Use %x to match the log line two lines above. * only skip descendant target lookup when source and dest use same store moveFolderSubEntries unconditionally passed skipTargetLookup=true for every descendant. This is safe when all paths resolve to the same underlying store, but with path-specific store configuration a child's destination may map to a different backend that already holds an entry at that path. Use FilerStoreWrapper.SameActualStore to check per-child and fall back to the full CreateEntry path when stores differ. * add nil and create edge-case tests for metadata event scope helpers * extract pathIsEqualOrUnder into util.IsEqualOrUnder Identical implementations existed in both replication/replicator.go and command/filer_sync.go. Move to util.IsEqualOrUnder (alongside the existing FullPath.IsUnder) and remove the duplicates. * use MetadataEventTargetDirectory for new-side directory in filer sync The new-side directory checks and sourceNewKey computation used message.NewParentPath directly. If NewParentPath were empty (legacy events, older filer versions during rolling upgrades), sourceNewKey would be wrong (/filename instead of /dir/filename) and the UpdateEntry parent path rewrite would panic on slice bounds. Derive targetDir once from MetadataEventTargetDirectory, which falls back to resp.Directory when NewParentPath is empty, and use it consistently for all new-side checks and the sink parent path.	2026-03-30 18:25:11 -07:00
msementsov	4c13a9ce65	Client disconnects create context cancelled errors, 500x errors and Filer lookup failures (#8845) ... * Update stream.go Client disconnects create context cancelled errors and Filer lookup failures * s3api: handle canceled stream requests cleanly * s3api: address canceled streaming review feedback --------- Co-authored-by: Chris Lu <chris.lu@gmail.com>	2026-03-30 12:11:30 -07:00
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	2026-03-29 13:46:54 -07:00
Chris Lu	c2c58419b8	filer.sync: send log file chunk fids to clients for direct volume server reads (#8792) ... * filer.sync: send log file chunk fids to clients for direct volume server reads Instead of the server reading persisted log files from volume servers, parsing entries, and streaming them over gRPC (serial bottleneck), clients that opt in via client_supports_metadata_chunks receive log file chunk references (fids) and read directly from volume servers in parallel. New proto messages: - LogFileChunkRef: chunk fids + timestamp + filer ID for one log file - SubscribeMetadataRequest.client_supports_metadata_chunks: client opt-in - SubscribeMetadataResponse.log_file_refs: server sends refs during backlog Server changes: - CollectLogFileRefs: lists log files and returns chunk refs without any volume server I/O (metadata-only operation) - SubscribeMetadata/SubscribeLocalMetadata: when client opts in, sends refs during persisted log phase, then falls back to normal streaming for in-memory events Client changes: - ReadLogFileRefs: reads log files from volume servers, parses entries, filters by path prefix, invokes processEventFn - MetadataFollowOption.LogFileReaderFn: factory for chunk readers, enables metadata chunks when non-nil - Both filer_pb_tail.go and meta_aggregator.go recv loops accumulate refs then process them at the disk→memory transition Backward compatible: old clients don't set the flag, get existing behavior. Ref: #8771 * filer.sync: merge entries across filers in timestamp order on client side ReadLogFileRefs now groups refs by filer ID and merges entries from multiple filers using a min-heap priority queue — the same algorithm the server uses in OrderedLogVisitor + LogEntryItemPriorityQueue. This ensures events are processed in correct timestamp order even when log files from different filers have interleaved timestamps. Single-filer case takes the fast path (no heap allocation). * filer.sync: integration tests for direct-read metadata chunks Three test categories: 1. Merge correctness (TestReadLogFileRefsMergeOrder): Verifies entries from 3 filers are delivered in strict timestamp order, matching the server-side OrderedLogVisitor guarantee. 2. Path filtering (TestReadLogFileRefsPathFilter): Verifies client-side path prefix filtering works correctly. 3. Throughput comparison (TestDirectReadVsServerSideThroughput): 3 filers × 7 files × 300 events = 6300 events, 2ms per file read: server-side: 6300 events 218ms 28,873 events/sec direct-read: 6300 events 51ms 123,566 events/sec (4.3x) parallel: 6300 events 17ms 378,628 events/sec (13.1x) Direct-read eliminates gRPC send overhead per event (4.3x). Parallel per-filer reading eliminates serial file I/O (13.1x). * filer.sync: parallel per-filer reads with prefetching in ReadLogFileRefs ReadLogFileRefs now has two levels of I/O overlap: 1. Cross-filer parallelism: one goroutine per filer reads its files concurrently. Entries feed into per-filer channels, merged by the main goroutine via min-heap (same ordering guarantee as the server's OrderedLogVisitor). 2. Within-filer prefetching: while the current file's entries are being consumed by the merge heap, the next file is already being read from the volume server in a background goroutine. Single-filer fast path avoids the heap and channels. Test results (3 filers × 7 files × 300 events, 2ms per file read): server-side sequential: 6300 events 212ms 29,760 events/sec parallel + prefetch: 6300 events 36ms 177,443 events/sec Speedup: 6.0x * filer.sync: address all review comments on metadata chunks PR Critical fixes: - sendLogFileRefs: bypass pipelinedSender, send directly on gRPC stream. Ref messages have TsNs=0 and were being incorrectly batched into the Events field by the adaptive batching logic, corrupting ref delivery. - readLogFileEntries: use io.ReadFull instead of reader.Read to prevent partial reads from corrupting size values or protobuf data. - Error handling: only skip chunk-not-found errors (matching server-side isChunkNotFoundError). Other I/O or decode failures are propagated so the follower can retry. High-priority fixes: - CollectLogFileRefs: remove incorrect +24h padding from stopTime. The extra day caused unnecessary log file refs to be collected. - Path filtering: ReadLogFileRefs now accepts PathFilter struct with PathPrefix, AdditionalPathPrefixes, and DirectoriesToWatch. Uses util.Join for path construction (avoids "//foo" on root). Excludes /.system/log/ internal entries. Matches server-side eachEventNotificationFn filtering logic. Medium-priority fixes: - CollectLogFileRefs: accept context.Context, propagate to ListDirectoryEntries calls for cancellation support. - NewChunkStreamReaderFromLookup: accept context.Context, propagate to doNewChunkStreamReader. Test fixes: - Check error returns from ReadLogFileRefs in all test call sites. --------- Co-authored-by: Copilot <copilot@github.com>	2026-03-27 11:01:29 -07:00
Chris Lu	d97660d0cd	filer.sync: pipelined subscription with adaptive batching for faster catch-up (#8791) ... * filer.sync: pipelined subscription with adaptive batching for faster catch-up The SubscribeMetadata pipeline was fully serial: reading a log entry from a volume server, unmarshaling, filtering, and calling stream.Send() all happened one-at-a-time. stream.Send() blocked the entire pipeline until the client acknowledged each event, limiting throughput to ~80 events/sec regardless of the -concurrency setting. Three server-side optimizations that stack: 1. Pipelined sender: decouple stream.Send() from the read loop via a buffered channel (1024 messages). A dedicated goroutine handles gRPC delivery while the reader continues processing the next events. 2. Adaptive batching: when event timestamps are >2min behind wall clock (backlog catch-up), drain multiple events from the channel and pack them into a single stream.Send() using a new `repeated events` field on SubscribeMetadataResponse. When events are recent (real-time), send one-by-one for low latency. Old clients ignore the new field (backward compatible). 3. Persisted log readahead: run the OrderedLogVisitor in a background goroutine so volume server I/O for the next log file overlaps with event processing and gRPC delivery. 4. Event-driven aggregated subscription: replace time.Sleep(1127ms) polling in SubscribeMetadata with notification-driven wake-up using the MetaLogBuffer subscriber mechanism, reducing real-time latency from ~1127ms to sub-millisecond. Combined, these create a 3-stage pipeline: [Volume I/O → readahead buffer] → [Filter → send buffer] → [gRPC Send] Test results (simulated backlog with 50µs gRPC latency per Send): direct (old): 2100 events 2100 sends 168ms 12,512 events/sec pipelined+batched: 2100 events 14 sends 40ms 52,856 events/sec Speedup: 4.2x single-stream throughput Ref: #8771 * filer.sync: require client opt-in for batch event delivery Add ClientSupportsBatching field to SubscribeMetadataRequest. The server only packs events into the Events batch field when the client explicitly sets this flag to true. Old clients (Java SDK, third-party) that don't set the flag get one-event-per-Send, preserving backward compatibility. All Go callers (FollowMetadata, MetaAggregator) set the flag to true since their recv loops already unpack batched events. * filer.sync: clear batch Events field after Send to release references Prevents the envelope message from holding references to the rest of the batch after gRPC serialization, allowing the GC to collect them sooner. * filer.sync: fix Send deadlock, add error propagation test, event-driven local subscribe - pipelinedSender.Send: add case <-s.done to unblock when sender goroutine exits (fixes deadlock when errCh was already consumed by a prior Send). - pipelinedSender.reportErr: remove for-range drain on sendCh that could block indefinitely. Send() now detects exit via s.done instead. - SubscribeLocalMetadata: replace remaining time.Sleep(1127ms) in the gap-detected-no-memory-data path with event-driven listenersCond.Wait(), consistent with the rest of the subscription paths. - Add TestPipelinedSenderErrorPropagation: verifies error surfaces via Send and Close when the underlying stream fails. - Replace goto with labeled break in test simulatePipeline. * filer.sync: check error returns in test code - direct_send: check slowStream.Send error return - pipelined_batched_send: check sender.Close error return - simulatePipeline: return error from sender.Close, propagate to callers --------- Co-authored-by: Copilot <copilot@github.com>	2026-03-26 23:55:42 -07:00
Lars Lehtonen	e8888765a1	fix(weed/filer/store_test): fix dropped errors (#8782)	2026-03-26 12:07:48 -07:00
Chris Lu	94bfa2b340	mount: stream all filer mutations over single ordered gRPC stream (#8770) ... * filer: add StreamMutateEntry bidi streaming RPC Add a bidirectional streaming RPC that carries all filer mutation types (create, update, delete, rename) over a single ordered stream. This eliminates per-request connection overhead for pipelined operations and guarantees mutation ordering within a stream. The server handler delegates each request to the existing unary handlers (CreateEntry, UpdateEntry, DeleteEntry) and uses a proxy stream adapter for rename operations to reuse StreamRenameEntry logic. The is_last field signals completion for multi-response operations (rename sends multiple events per request; create/update/delete always send exactly one response with is_last=true). * mount: add streaming mutation multiplexer (streamMutateMux) Implement a client-side multiplexer that routes all filer mutation RPCs (create, update, delete, rename) over a single bidirectional gRPC stream. Multiple goroutines submit requests through a send channel; a dedicated sendLoop serializes them on the stream; a recvLoop dispatches responses to waiting callers via per-request channels. Key features: - Lazy stream opening on first use - Automatic reconnection on stream failure - Permanent fallback to unary RPCs if filer returns Unimplemented - Monotonic request_id for response correlation - Multi-response support for rename operations (is_last signaling) The mux is initialized on WFS and closed during unmount cleanup. No call sites use it yet — wiring comes in subsequent commits. * mount: route CreateEntry and UpdateEntry through streaming mux Wire all CreateEntry call sites to use wfs.streamCreateEntry() which routes through the StreamMutateEntry stream when available, falling back to unary RPCs otherwise. Also wire Link's UpdateEntry calls through wfs.streamUpdateEntry(). Updated call sites: - flushMetadataToFiler (file flush after write) - Mkdir (directory creation) - Symlink (symbolic link creation) - createRegularFile non-deferred path (Mknod) - flushFileMetadata (periodic metadata flush) - Link (hard link: update source + create link + rollback) * mount: route UpdateEntry and DeleteEntry through streaming mux Wire remaining mutation call sites through the streaming mux: - saveEntry (Setattr/chmod/chown/utimes) → streamUpdateEntry - Unlink → streamDeleteEntry (replaces RemoveWithResponse) - Rmdir → streamDeleteEntry (replaces RemoveWithResponse) All filer mutations except Rename now go through StreamMutateEntry when the filer supports it, with automatic unary RPC fallback. * mount: route Rename through streaming mux Wire Rename to use streamMutate.Rename() when available, with fallback to the existing StreamRenameEntry unary stream. The streaming mux sends rename as a StreamRenameEntryRequest oneof variant. The server processes it through the existing rename logic and sends multiple StreamRenameEntryResponse events (one per moved entry), with is_last=true on the final response. All filer mutations now go through a single ordered stream. * mount: fix stream mux connection ownership WithGrpcClient(streamingMode=true) closes the gRPC connection when the callback returns, destroying the stream. Own the connection directly via pb.GrpcDial so it stays alive for the stream's lifetime. Close it explicitly in recvLoop on stream failure and in Close on shutdown. * mount: fix rename failure for deferred-create files Three fixes for rename operations over the streaming mux: 1. lookupEntry: fall back to local metadata store when filer returns "not found" for entries in uncached directories. Files created with deferFilerCreate=true exist only in the local leveldb store until flushed; lookupEntry skipped the local store when the parent directory had never been readdir'd, causing rename to fail with ENOENT. 2. Rename: wait for pending async flushes and force synchronous flush of dirty metadata before sending rename to the filer. Covers the writebackCache case where close() defers the flush to a background worker that may not complete before rename fires. 3. StreamMutateEntry: propagate rename errors from server to client. Add error/errno fields to StreamMutateEntryResponse so the mount can map filer errors to correct FUSE status codes instead of silently returning OK. Also fix the existing Rename error handler which could return fuse.OK on unrecognized errors. * mount: fix streaming mux error handling, sendLoop lifecycle, and fallback Address PR review comments: 1. Server: populate top-level Error/Errno on StreamMutateEntryResponse for create/update/delete errors, not just rename. Previously update errors were silently dropped and create/delete errors were only in nested response fields that the client didn't check. 2. Client: check nested error fields in CreateEntry (ErrorCode, Error) and DeleteEntry (Error) responses, matching CreateEntryWithResponse behavior. 3. Fix sendLoop lifecycle: give each stream generation a stopSend channel. recvLoop closes it on error to stop the paired sendLoop. Previously a reconnect left the old sendLoop draining sendCh, breaking ordering. 4. Transparent fallback: stream helpers and doRename fall back to unary RPCs on transport errors (ErrStreamTransport), including the first Unimplemented from ensureStream. Previously the first call failed instead of degrading. 5. Filer rotation in openStream: try all filer addresses on dial failure, matching WithFilerClient behavior. Stop early on Unimplemented. 6. Pass metadata-bearing context to StreamMutateEntry RPC call so sw-client-id header is actually sent. 7. Gate lookupEntry local-cache fallback on open dirty handle or pending async flush to avoid resurrecting deleted/renamed entries. 8. Remove dead code in flushFileMetadata (err=nil followed by if err!=nil). 9. Use string matching for rename error-to-errno mapping in the mount to stay portable across Linux/macOS (numeric errno values differ). * mount: make failAllPending idempotent with delete-before-close Change failAllPending to collect pending entries into a local slice (deleting from the sync.Map first) before closing channels. This prevents double-close panics if called concurrently. Also remove the unused err parameter. * mount: add stream generation tracking and teardownStream Introduce a generation counter on streamMutateMux that increments each time a new stream is created. Requests carry the generation they were enqueued for so sendLoop can reject stale requests after reconnect. Add teardownStream(gen) which is idempotent (only acts when gen matches current generation and stream is non-nil). Both sendLoop and recvLoop call it on error, replacing the inline cleanup in recvLoop. sendLoop now actively triggers teardown on send errors instead of silently exiting. ensureStream waits for the prior generation's recvDone before creating a new stream, ensuring all old pending waiters are failed before reconnect. recvLoop now takes the stream, generation, and recvDone channel as parameters to avoid accessing shared fields without the lock. * mount: harden Close to prevent races with teardownStream Nil out stream, cancel, and grpcConn under the lock so that any concurrent teardownStream call from recvLoop/sendLoop becomes a no-op. Call failAllPending before closing sendCh to unblock waiters promptly. Guard recvDone with a nil check for the case where Close is called before any stream was ever opened. * mount: make errCh receive ctx-aware in doUnary and Rename Replace the blocking <-sendReq.errCh with a select that also observes ctx.Done(). If sendLoop exits via stopSend without consuming a buffered request, the caller now returns ctx.Err() instead of blocking forever. The buffered errCh (capacity 1) ensures late acknowledgements from sendLoop don't block the sender. * mount: fix sendLoop/Close race and recvLoop/teardown pending channel race Three related fixes: 1. Stop closing sendCh in Close(). Closing the shared producer channel races with callers who passed ensureStream() but haven't sent yet, causing send-on-closed-channel panics. sendCh is now left open; ensureStream checks m.closed to reject new callers. 2. Drain buffered sendCh items on shutdown. sendLoop defers drainSendCh() on exit so buffered requests get an ErrStreamTransport on their errCh instead of blocking forever. Close() drains again for any stragglers enqueued between sendLoop's drain and the final shutdown. 3. Move failAllPending from teardownStream into recvLoop's defer. teardownStream (called from sendLoop on send error) was closing pending response channels while recvLoop could be between pending.Load and the channel send — a send-on-closed-channel panic. recvLoop is now the sole closer of pending channels, eliminating the race. Close() waits on recvDone (with cancel() to guarantee Recv unblocks) so pending cleanup always completes. * filer/mount: add debug logging for hardlink lifecycle Add V(0) logging at every point where a HardLinkId is created, stored, read, or deleted to trace orphaned hardlink references. Logging covers: - gRPC server: CreateEntry/UpdateEntry when request carries HardLinkId - FilerStoreWrapper: InsertEntry/UpdateEntry when entry has HardLinkId - handleUpdateToHardLinks: entry path, HardLinkId, counter, chunk count - setHardLink: KvPut with blob size - maybeReadHardLink: V(1) on read attempt and successful decode - DeleteHardLink: counter decrement/deletion events - Mount Link(): when NewHardLinkId is generated and link is created This helps diagnose how a git pack .rev file ended up with a HardLinkId during a clone (no hard links should be involved). * test: add git clone/pull integration test for FUSE mount Shell script that exercises git operations on a SeaweedFS mount: 1. Creates a bare repo on the mount 2. Clones locally, makes 3 commits, pushes to mount 3. Clones from mount bare repo into an on-mount working dir 4. Verifies clone integrity (files, content, commit hashes) 5. Pushes 2 more commits with renames and deletes 6. Checks out an older revision on the mount clone 7. Returns to branch and pulls with real changes 8. Verifies file content, renames, deletes after pull 9. Checks git log integrity and clean status 27 assertions covering file existence, content, commit hashes, file counts, renames, deletes, and git status. Run against any existing mount: bash test-git-on-mount.sh /path/to/mount * test: add git clone/pull FUSE integration test to CI suite Add TestGitOperations to the existing fuse_integration test framework. The test exercises git's full file operation surface on the mount: 1. Creates a bare repo on the mount (acts as remote) 2. Clones locally, makes 3 commits (files, bulk data, renames), pushes 3. Clones from mount bare repo into an on-mount working dir 4. Verifies clone integrity (content, commit hash, file count) 5. Pushes 2 more commits with new files, renames, and deletes 6. Checks out an older revision on the mount clone 7. Returns to branch and pulls with real fast-forward changes 8. Verifies post-pull state: content, renames, deletes, file counts 9. Checks git log integrity (5 commits) and clean status Runs automatically in the existing fuse-integration.yml CI workflow. * mount: fix permission check with uid/gid mapping The permission checks in createRegularFile() and Access() compared the caller's local uid/gid against the entry's filer-side uid/gid without applying the uid/gid mapper. With -map.uid 501:0, a directory created as uid 0 on the filer would not match the local caller uid 501, causing hasAccess() to fall through to "other" permission bits and reject write access (0755 → other has r-x, no w). Fix: map entry uid/gid from filer-space to local-space before the hasAccess() call so both sides are in the same namespace. This fixes rsync -a failing with "Permission denied" on mkstempat when using uid/gid mapping. * mount: fix Mkdir/Symlink returning filer-side uid/gid to kernel Mkdir and Symlink used `defer wfs.mapPbIdFromFilerToLocal(entry)` to restore local uid/gid, but `outputPbEntry` writes the kernel response before the function returns — so the kernel received filer-side uid/gid (e.g., 0:0). macFUSE then caches these and rejects subsequent child operations (mkdir, create) because the caller uid (501) doesn't match the directory owner (0), and "other" bits (0755 → r-x) lack write permission. Fix: replace the defer with an explicit call to mapPbIdFromFilerToLocal before outputPbEntry, so the kernel gets local uid/gid. Also add nil guards for UidGidMapper in Access and createRegularFile to prevent panics in tests that don't configure a mapper. This fixes rsync -a "Permission denied" on mkpathat for nested directories when using uid/gid mapping. * mount: fix Link outputting filer-side uid/gid to kernel, add nil guards Link had the same defer-before-outputPbEntry bug as Mkdir and Symlink: the kernel received filer-side uid/gid because the defer hadn't run yet when outputPbEntry wrote the response. Also add nil guards for UidGidMapper in Access and createRegularFile so tests without a mapper don't panic. Audit of all outputPbEntry/outputFilerEntry call sites: - Mkdir: fixed in prior commit (explicit map before output) - Symlink: fixed in prior commit (explicit map before output) - Link: fixed here (explicit map before output) - Create (existing file): entry from maybeLoadEntry (already mapped) - Create (deferred): entry has local uid/gid (never mapped to filer) - Create (non-deferred): createRegularFile defer runs before return - Mknod: createRegularFile defer runs before return - Lookup: entry from lookupEntry (already mapped) - GetAttr: entry from maybeReadEntry/maybeLoadEntry (already mapped) - readdir: entry from cache (mapIdFromFilerToLocal) or filer (mapped) - saveEntry: no kernel output - flushMetadataToFiler: no kernel output - flushFileMetadata: no kernel output * test: fix git test for same-filesystem FUSE clone When both the bare repo and working clone live on the same FUSE mount, git's local transport uses hardlinks and cross-repo stat calls that fail on FUSE. Fix: - Use --no-local on clone to disable local transport optimizations - Use reset --hard instead of checkout to stay on branch - Use fetch + reset --hard origin/<branch> instead of git pull to avoid local transport stat failures during fetch * adjust logging * test: use plain git clone/pull to exercise real FUSE behavior Remove --no-local and fetch+reset workarounds. The test should use the same git commands users run (clone, reset --hard, pull) so it reveals real FUSE issues rather than hiding them. * test: enable V(1) logging for filer/mount and collect logs on failure - Run filer and mount with -v=1 so hardlink lifecycle logs (V(0): create/delete/insert, V(1): read attempts) are captured - On test failure, automatically dump last 16KB of all process logs (master, volume, filer, mount) to test output - Copy process logs to /tmp/seaweedfs-fuse-logs/ for CI artifact upload - Update CI workflow to upload SeaweedFS process logs alongside test output * mount: clone entry for filer flush to prevent uid/gid race flushMetadataToFiler and flushFileMetadata used entry.GetEntry() which returns the file handle's live proto entry pointer, then mutated it in-place via mapPbIdFromLocalToFiler. During the gRPC call window, a concurrent Lookup (which takes entryLock.RLock but NOT fhLockTable) could observe filer-side uid/gid (e.g., 0:0) on the file handle entry and return it to the kernel. The kernel caches these attributes, so subsequent opens by the local user (uid 501) fail with EACCES. Fix: proto.Clone the entry before mapping uid/gid for the filer request. The file handle's live entry is never mutated, so concurrent Lookup always sees local uid/gid. This fixes the intermittent "Permission denied" on .git/FETCH_HEAD after the first git pull on a mount with uid/gid mapping. * mount: add debug logging for stale lock file investigation Add V(0) logging to trace the HEAD.lock recreation issue: - Create: log when O_EXCL fails (file already exists) with uid/gid/mode - completeAsyncFlush: log resolved path, saved path, dirtyMetadata, isDeleted at entry to trace whether async flush fires after rename - flushMetadataToFiler: log the dir/name/fullpath being flushed This will show whether the async flush is recreating the lock file after git renames HEAD.lock → HEAD. * mount: prevent async flush from recreating renamed .lock files When git renames HEAD.lock → HEAD, the async flush from the prior close() can run AFTER the rename and re-insert HEAD.lock into the meta cache via its CreateEntryRequest response event. The next git pull then sees HEAD.lock and fails with "File exists". Fix: add isRenamed flag on FileHandle, set by Rename before waiting for the pending async flush. The async flush checks this flag and skips the metadata flush for renamed files (same pattern as isDeleted for unlinked files). The data pages still flush normally. The Rename handler flushes deferred metadata synchronously (Case 1) before setting isRenamed, ensuring the entry exists on the filer for the rename to proceed. For already-released handles (Case 2), the entry was created by a prior flush. * mount: also mark renamed inodes via entry.Attributes.Inode fallback When GetInode fails (Forget already removed the inode mapping), the Rename handler couldn't find the pending async flush to set isRenamed. The async flush then recreated the .lock file on the filer. Fix: fall back to oldEntry.Attributes.Inode to find the pending async flush when the inode-to-path mapping is gone. Also extract MarkInodeRenamed into a method on FileHandleToInode for clarity. * mount: skip async metadata flush when saved path no longer maps to inode The isRenamed flag approach failed for refs/remotes/origin/HEAD.lock because neither GetInode nor oldEntry.Attributes.Inode could find the inode (Forget already evicted the mapping, and the entry's stored inode was 0). Add a direct check in completeAsyncFlush: before flushing metadata, verify that the saved path still maps to this inode in the inode-to-path table. If the path was renamed or removed (inode mismatch or not found), skip the metadata flush to avoid recreating a stale entry. This catches all rename cases regardless of whether the Rename handler could set the isRenamed flag. * mount: wait for pending async flush in Unlink before filer delete Unlink was deleting the filer entry first, then marking the draining async-flush handle as deleted. The async flush worker could race between these two operations and recreate the just-unlinked entry on the filer. This caused git's .lock files (e.g. refs/remotes/origin/HEAD.lock) to persist after git pull, breaking subsequent git operations. Move the isDeleted marking and add waitForPendingAsyncFlush() before the filer delete so any in-flight flush completes first. Even if the worker raced past the isDeleted check, the wait ensures it finishes before the filer delete cleans up any recreated entry. * mount: reduce async flush and metadata flush log verbosity Raise completeAsyncFlush entry log, saved-path-mismatch skip log, and flushMetadataToFiler entry log from V(0) to V(3)/V(4). These fire for every file close with writebackCache and are too noisy for normal use. * filer: reduce hardlink debug log verbosity from V(0) to V(4) HardLinkId logs in filerstore_wrapper, filerstore_hardlink, and filer_grpc_server fire on every hardlinked file operation (git pack files use hardlinks extensively) and produce excessive noise. * mount/filer: reduce noisy V(0) logs for link, rmdir, and empty folder check - weedfs_link.go: hardlink creation logs V(0) → V(4) - weedfs_dir_mkrm.go: non-empty folder rmdir error V(0) → V(1) - empty_folder_cleaner.go: "not empty" check log V(0) → V(4) * filer: handle missing hardlink KV as expected, not error A "kv: not found" on hardlink read is normal when the link blob was already cleaned up but a stale entry still references it. Log at V(1) for not-found; keep Error level for actual KV failures. * test: add waitForDir before git pull in FUSE git operations test After git reset --hard, the FUSE mount's metadata cache may need a moment to settle on slow CI. The git pull subprocess (unpack-objects) could fail to stat the working directory. Poll for up to 5s. * Update git_operations_test.go * wait * test: simplify FUSE test framework to use weed mini Replace the 4-process setup (master + volume + filer + mount) with 2 processes: "weed mini" (all-in-one) + "weed mount". This simplifies startup, reduces port allocation, and is faster on CI. * test: fix mini flag -admin → -admin.ui	2026-03-25 20:06:34 -07:00
Chris Lu	0b3867dca3	filer: add structured error codes to CreateEntryResponse (#8767) ... * filer: add FilerError enum and error_code field to CreateEntryResponse Add a machine-readable error code alongside the existing string error field. This follows the precedent set by PublishMessageResponse in the MQ broker proto. The string field is kept for human readability and backward compatibility. Defined codes: OK, ENTRY_NAME_TOO_LONG, PARENT_IS_FILE, EXISTING_IS_DIRECTORY, EXISTING_IS_FILE, ENTRY_ALREADY_EXISTS. * filer: add sentinel errors and error code mapping in filer_pb Define sentinel errors (ErrEntryNameTooLong, ErrParentIsFile, etc.) in the filer_pb package so both the filer and consumers can reference them without circular imports. Add FilerErrorToSentinel() to map proto error codes to sentinels, and update CreateEntryWithResponse() to check error_code first, falling back to the string-based path for backward compatibility with old servers. * filer: return wrapped sentinel errors and set proto error codes Replace fmt.Errorf string errors in filer.CreateEntry, UpdateEntry, and ensureParentDirectoryEntry with wrapped filer_pb sentinel errors (using %w). This preserves errors.Is() traversal on the server side. In the gRPC CreateEntry handler, map sentinel errors to the corresponding FilerError proto codes using errors.Is(), setting both resp.Error (string, for backward compat) and resp.ErrorCode (enum). * S3: use errors.Is() with filer sentinels instead of string matching Replace fragile string-based error matching in filerErrorToS3Error and other S3 API consumers with errors.Is() checks against filer_pb sentinel errors. This works because the updated CreateEntryWithResponse helper reconstructs sentinel errors from the proto FilerError code. Update iceberg stage_create and metadata_files to check resp.ErrorCode instead of parsing resp.Error strings. Update SSE-S3 to use errors.Is() for the already-exists check. String matching is retained only for non-filer errors (gRPC transport errors, checksum validation) that don't go through CreateEntryResponse. * filer: remove backward-compat string fallbacks for error codes Clients and servers are always deployed together, so there is no need for backward-compatibility fallback paths that parse resp.Error strings when resp.ErrorCode is unset. Simplify all consumers to rely solely on the structured error code. * iceberg: ensure unknown non-OK error codes are not silently ignored When FilerErrorToSentinel returns nil for an unrecognized error code, return an error including the code and message rather than falling through to return nil. * filer: fix redundant error message and restore error wrapping in helper Use request path instead of resp.Error in the sentinel error format string to avoid duplicating the sentinel message (e.g. "entry already exists: entry already exists"). Restore %w wrapping with errors.New() in the fallback paths so callers can use errors.Is()/errors.As(). * filer: promote file to directory on path conflict instead of erroring S3 allows both "foo/bar" (object) and "foo/bar/xyzzy" (another object) to coexist because S3 has a flat key space. When ensureParentDirectoryEntry finds a parent path that is a file instead of a directory, promote it to a directory by setting ModeDir while preserving the original content and chunks. Use Store.UpdateEntry directly to bypass the Filer.UpdateEntry type-change guard. This fixes the S3 compatibility test failures where creating overlapping keys (e.g. "foo/bar" then "foo/bar/xyzzy") returned ExistingObjectIsFile.	2026-03-24 17:08:22 -07:00
Chris Lu	2877febd73	S3: fix silent PutObject failure and enforce 1024-byte key limit (#8764) ... * S3: add KeyTooLongError error code Add ErrKeyTooLongError (HTTP 400, code "KeyTooLongError") to match the standard AWS S3 error for object keys that exceed length limits. * S3: fix silent PutObject failure when entry name exceeds max_file_name_length putToFiler called client.CreateEntry() directly and discarded the gRPC response. The filer embeds application errors like "entry name too long" in resp.Error (not as gRPC transport errors), so the error was silently swallowed and clients received HTTP 200 with an ETag for objects that were never stored. Switch to the filer_pb.CreateEntry() helper which properly checks resp.Error, and map "entry name too long" to KeyTooLongError (HTTP 400). To avoid fragile string parsing across the gRPC boundary, define shared error message constants in weed/util/constants and use them in both the filer (producing errors) and S3 API (matching errors). Switch filerErrorToS3Error to use strings.Contains/HasSuffix with these constants so matches work regardless of any wrapper prefix. Apply filerErrorToS3Error to the mkdir path for directory markers. Fixes #8759 * S3: enforce 1024-byte maximum object key length AWS S3 limits object keys to 1024 bytes. Add early validation on write paths (PutObject, CopyObject, CreateMultipartUpload) to reject keys exceeding the limit with the standard KeyTooLongError (HTTP 400). The key length check runs before bucket auto-creation to prevent overlong keys from triggering unnecessary side effects. Also use filerErrorToS3Error for CopyObject's mkFile error paths so name-too-long errors from the filer return KeyTooLongError instead of InternalError. Ref #8758 * S3: add handler-level tests for key length validation and error mapping Add tests for filerErrorToS3Error mapping "entry name too long" to KeyTooLongError, including a regression test for the CreateEntry-prefixed "existing ... is a directory" form. Add handler-level integration tests that exercise PutObjectHandler, CopyObjectHandler, and NewMultipartUploadHandler via httptest, verifying HTTP 400 and KeyTooLongError XML response for overlong keys and acceptance of keys at the 1024-byte limit.	2026-03-24 13:35:28 -07:00
Chris Lu	c31e6b4684	Use filer-side copy for mounted whole-file copy_file_range (#8747) ... * Optimize mounted whole-file copy_file_range * Address mounted copy review feedback * Harden mounted copy fast path --------- Co-authored-by: Copilot <copilot@github.com>	2026-03-23 18:35:15 -07:00
Chris Lu	6bf654c25c	fix: keep metadata subscriptions progressing (#8730) (#8746) ... * fix: keep metadata subscriptions progressing (#8730) * test: cancel slow metadata writers with parent context * filer: ignore missing persisted log chunks	2026-03-23 15:26:54 -07:00
Chris Lu	d6a872c4b9	Preserve explicit directory markers with octet-stream MIME (#8726) ... * Preserve octet-stream MIME on explicit directory markers * Run empty directory marker regression in CI * Run S3 Spark workflow for filer changes	2026-03-21 19:31:56 -07:00
Chris Lu	81369b8a83	improve: large file sync throughput for remote.cache and filer.sync (#8676) ... * improve large file sync throughput for remote.cache and filer.sync Three main throughput improvements: 1. Adaptive chunk sizing for remote.cache: targets ~32 chunks per file instead of always starting at 5MB. A 500MB file now uses ~16MB chunks (32 chunks) instead of 5MB chunks (100 chunks), reducing per-chunk overhead (volume assign, gRPC call, needle write) by 3x. 2. Configurable concurrency at every layer: - remote.cache chunk concurrency: -chunkConcurrency flag (default 8) - remote.cache S3 download concurrency: -downloadConcurrency flag (default raised from 1 to 5 per chunk) - filer.sync chunk concurrency: -chunkConcurrency flag (default 32) 3. S3 multipart download concurrency raised from 1 to 5: the S3 manager downloader was using Concurrency=1, serializing all part downloads within each chunk. This alone can 5x per-chunk download speed. The concurrency values flow through the gRPC request chain: shell command → CacheRemoteObjectToLocalClusterRequest → FetchAndWriteNeedleRequest → S3 downloader Zero values in the request mean "use server defaults", maintaining full backward compatibility with existing callers. Ref #8481 * fix: use full maxMB for chunk size cap and remove loop guard Address review feedback: - Use full maxMB instead of maxMB/2 for maxChunkSize to avoid unnecessarily limiting chunk size for very large files. - Remove chunkSize < maxChunkSize guard from the safety loop so it can always grow past maxChunkSize when needed to stay under 1000 chunks (e.g., extremely large files with small maxMB). * address review feedback: help text, validation, naming, docs - Fix help text for -chunkConcurrency and -downloadConcurrency flags to say "0 = server default" instead of advertising specific numeric defaults that could drift from the server implementation. - Validate chunkConcurrency and downloadConcurrency are within int32 range before narrowing, returning a user-facing error if out of range. - Rename ReadRemoteErr to readRemoteErr to follow Go naming conventions. - Add doc comment to SetChunkConcurrency noting it must be called during initialization before replication goroutines start. - Replace doubling loop in chunk size safety check with direct ceil(remoteSize/1000) computation to guarantee the 1000-chunk cap. * address Copilot review: clamp concurrency, fix chunk count, clarify proto docs - Use ceiling division for chunk count check to avoid overcounting when file size is an exact multiple of chunk size. - Clamp chunkConcurrency (max 1024) and downloadConcurrency (max 1024 at filer, max 64 at volume server) to prevent excessive goroutines. - Always use ReadFileWithConcurrency when the client supports it, falling back to the implementation's default when value is 0. - Clarify proto comments that download_concurrency only applies when the remote storage client supports it (currently S3). - Include specific server defaults in help text (e.g., "0 = server default 8") so users see the actual values in -h output. * fix data race on executionErr and use %w for error wrapping - Protect concurrent writes to executionErr in remote.cache worker goroutines with a sync.Mutex to eliminate the data race. - Use %w instead of %v in volume_grpc_remote.go error formatting to preserve the error chain for errors.Is/errors.As callers.	2026-03-17 16:49:56 -07:00
Chris Lu	8cde3d4486	Add data file compaction to iceberg maintenance (Phase 2) (#8503) ... * Add iceberg_maintenance plugin worker handler (Phase 1) Implement automated Iceberg table maintenance as a new plugin worker job type. The handler scans S3 table buckets for tables needing maintenance and executes operations in the correct Iceberg order: expire snapshots, remove orphan files, and rewrite manifests. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * Add data file compaction to iceberg maintenance handler (Phase 2) Implement bin-packing compaction for small Parquet data files: - Enumerate data files from manifests, group by partition - Merge small files using parquet-go (read rows, write merged output) - Create new manifest with ADDED/DELETED/EXISTING entries - Commit new snapshot with compaction metadata Add 'compact' operation to maintenance order (runs before expire_snapshots), configurable via target_file_size_bytes and min_input_files thresholds. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * Fix memory exhaustion in mergeParquetFiles by processing files sequentially Previously all source Parquet files were loaded into memory simultaneously, risking OOM when a compaction bin contained many small files. Now each file is loaded, its rows are streamed into the output writer, and its data is released before the next file is loaded — keeping peak memory proportional to one input file plus the output buffer. * Validate bucket/namespace/table names against path traversal Reject names containing '..', '/', or '\' in Execute to prevent directory traversal via crafted job parameters. * Add filer address failover in iceberg maintenance handler Try each filer address from cluster context in order instead of only using the first one. This improves resilience when the primary filer is temporarily unreachable. * Add separate MinManifestsToRewrite config for manifest rewrite threshold The rewrite_manifests operation was reusing MinInputFiles (meant for compaction bin file counts) as its manifest count threshold. Add a dedicated MinManifestsToRewrite field with its own config UI section and default value (5) so the two thresholds can be tuned independently. * Fix risky mtime fallback in orphan removal that could delete new files When entry.Attributes is nil, mtime defaulted to Unix epoch (1970), which would always be older than the safety threshold, causing the file to be treated as eligible for deletion. Skip entries with nil Attributes instead, matching the safer logic in operations.go. * Fix undefined function references in iceberg_maintenance_handler.go Use the exported function names (ShouldSkipDetectionByInterval, BuildDetectorActivity, BuildExecutorActivity) matching their definitions in vacuum_handler.go. * Remove duplicated iceberg maintenance handler in favor of iceberg/ subpackage The IcebergMaintenanceHandler and its compaction code in the parent pluginworker package duplicated the logic already present in the iceberg/ subpackage (which self-registers via init()). The old code lacked stale-plan guards, proper path normalization, CAS-based xattr updates, and error-returning parseOperations. Since the registry pattern (default "all") makes the old handler unreachable, remove it entirely. All functionality is provided by iceberg.Handler with the reviewed improvements. * Fix MinManifestsToRewrite clamping to match UI minimum of 2 The clamp reset values below 2 to the default of 5, contradicting the UI's advertised MinValue of 2. Clamp to 2 instead. * Sort entries by size descending in splitOversizedBin for better packing Entries were processed in insertion order which is non-deterministic from map iteration. Sorting largest-first before the splitting loop improves bin packing efficiency by filling bins more evenly. * Add context cancellation check to drainReader loop The row-streaming loop in drainReader did not check ctx between iterations, making long compaction merges uncancellable. Check ctx.Done() at the top of each iteration. * Fix splitOversizedBin to always respect targetSize limit The minFiles check in the split condition allowed bins to grow past targetSize when they had fewer than minFiles entries, defeating the OOM protection. Now bins always split at targetSize, and a trailing runt with fewer than minFiles entries is merged into the previous bin. * Add integration tests for iceberg table maintenance plugin worker Tests start a real weed mini cluster, create S3 buckets and Iceberg table metadata via filer gRPC, then exercise the iceberg.Handler operations (ExpireSnapshots, RemoveOrphans, RewriteManifests) against the live filer. A full maintenance cycle test runs all operations in sequence and verifies metadata consistency. Also adds exported method wrappers (testing_api.go) so the integration test package can call the unexported handler methods. * Fix splitOversizedBin dropping files and add source path to drainReader errors The runt-merge step could leave leading bins with fewer than minFiles entries (e.g. [80,80,10,10] with targetSize=100, minFiles=2 would drop the first 80-byte file). Replace the filter-based approach with an iterative merge that folds any sub-minFiles bin into its smallest neighbor, preserving all eligible files. Also add the source file path to drainReader error messages so callers can identify which Parquet file caused a read/write failure. * Harden integration test error handling - s3put: fail immediately on HTTP 4xx/5xx instead of logging and continuing - lookupEntry: distinguish NotFound (return nil) from unexpected RPC errors (fail the test) - writeOrphan and orphan creation in FullMaintenanceCycle: check CreateEntryResponse.Error in addition to the RPC error * go fmt --------- Co-authored-by: Copilot <copilot@github.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>	2026-03-15 11:27:42 -07:00
Chris Lu	f3c5ba3cd6	feat(filer): add lazy directory listing for remote mounts (#8615) ... * feat(filer): add lazy directory listing for remote mounts Directory listings on remote mounts previously only queried the local filer store. With lazy mounts the listing was empty; with eager mounts it went stale over time. Add on-demand directory listing that fetches from remote and caches results with a 5-minute TTL: - Add `ListDirectory` to `RemoteStorageClient` interface (delimiter-based, single-level listing, separate from recursive `Traverse`) - Implement in S3, GCS, and Azure backends using each platform's hierarchical listing API - Add `maybeLazyListFromRemote` to filer: before each directory listing, check if the directory is under a remote mount with an expired cache, fetch from remote, persist entries to the local store, then let existing listing logic run on the populated store - Use singleflight to deduplicate concurrent requests for the same directory - Skip local-only entries (no RemoteEntry) to avoid overwriting unsynced uploads - Errors are logged and swallowed (availability over consistency) * refactor: extract xattr key to constant xattrRemoteListingSyncedAt * feat: make listing cache TTL configurable per mount via listing_cache_ttl_seconds Add listing_cache_ttl_seconds field to RemoteStorageLocation protobuf. When 0 (default), lazy directory listing is disabled for that mount. When >0, enables on-demand directory listing with the specified TTL. Expose as -listingCacheTTL flag on remote.mount command. * refactor: address review feedback for lazy directory listing - Add context.Context to ListDirectory interface and all implementations - Capture startTime before remote call for accurate TTL tracking - Simplify S3 ListDirectory using ListObjectsV2PagesWithContext - Make maybeLazyListFromRemote return void (errors always swallowed) - Remove redundant trailing-slash path manipulation in caller - Update tests to match new signatures * When an existing entry has Remote != nil, we should merge remote metadata into it rather than replacing it. * fix(gcs): wrap ListDirectory iterator error with context The raw iterator error was returned without bucket/path context, making it harder to debug. Wrap it consistently with the S3 pattern. * fix(s3): guard against nil pointer dereference in Traverse and ListDirectory Some S3-compatible backends may return nil for LastModified, Size, or ETag fields. Check for nil before dereferencing to prevent panics. * fix(filer): remove blanket 2-minute timeout from lazy listing context Individual SDK operations (S3, GCS, Azure) already have per-request timeouts and retry policies. The blanket timeout could cut off large directory listings mid-operation even though individual pages were succeeding. * fix(filer): preserve trace context in lazy listing with WithoutCancel Use context.WithoutCancel(ctx) instead of context.Background() so trace/span values from the incoming request are retained for distributed tracing, while still decoupling cancellation. * fix(filer): use Store.FindEntry for internal lookups, add Uid/Gid to files, fix updateDirectoryListingSyncedAt - Use f.Store.FindEntry instead of f.FindEntry for staleness check and child lookups to avoid unnecessary lazy-fetch overhead - Set OS_UID/OS_GID on new file entries for consistency with directories - In updateDirectoryListingSyncedAt, use Store.UpdateEntry for existing directories instead of CreateEntry to avoid deleteChunksIfNotNew and NotifyUpdateEvent side effects * fix(filer): distinguish not-found from store errors in lazy listing Previously, any error from Store.FindEntry was treated as "not found," which could cause entry recreation/overwrite on transient DB failures. Now check for filer_pb.ErrNotFound explicitly and skip entries or bail out on real store errors. * refactor(filer): use errors.Is for ErrNotFound comparisons	2026-03-13 09:36:54 -07:00
Peter Dodd	146a090754	filer: propagate lazy metadata deletes to remote mounts (#8522) ... * filer: propagate lazy metadata deletes to remote mounts Delete operations now call the remote backend for mounted remote-only entries before removing filer metadata, keeping remote state aligned and preserving retry semantics on remote failures. Made-with: Cursor * filer: harden remote delete metadata recovery Persist remote-delete metadata pendings so local entry removal can be retried after failures, and return explicit errors when remote client resolution fails to prevent silent local-only deletes. Made-with: Cursor * filer: streamline remote delete client lookup and logging Avoid a redundant mount trie traversal by resolving the remote client directly from the matched mount location, and add parity logging for successful remote directory deletions. Made-with: Cursor * filer: harden pending remote metadata deletion flow Retry pending-marker writes before local delete, fail closed when marking cannot be persisted, and start remote pending reconciliation only after the filer store is initialised to avoid nil store access. Made-with: Cursor * filer: avoid lazy fetch in pending metadata reconciliation Use a local-only entry lookup during pending remote metadata reconciliation so cache misses do not trigger remote lazy fetches. Made-with: Cursor * filer: serialise concurrent index read-modify-write in pending metadata deletion Add remoteMetadataDeletionIndexMu to Filer and acquire it for the full read→mutate→commit sequence in markRemoteMetadataDeletionPending and clearRemoteMetadataDeletionPending, preventing concurrent goroutines from overwriting each other's index updates. Made-with: Cursor * filer: start remote deletion reconciliation loop in NewFiler Move the background goroutine for pending remote metadata deletion reconciliation from SetStore (where it was gated by sync.Once) to NewFiler alongside the existing loopProcessingDeletion goroutine. The sync.Once approach was problematic: it buried a goroutine launch as a side effect of a setter, was unrecoverable if the goroutine panicked, could race with store initialisation, and coupled its lifecycle to unrelated shutdown machinery. The existing nil-store guard in reconcilePendingRemoteMetadataDeletions handles the window before SetStore is called. * filer: skip remote delete for replicated deletes from other filers When isFromOtherCluster is true the delete was already propagated to the remote backend by the originating filer. Repeating the remote delete on every replica doubles API calls, and a transient remote failure on the replica would block local metadata cleanup — leaving filers inconsistent. * filer: skip pending marking for directory remote deletes Directory remote deletes are idempotent and do not need the pending/reconcile machinery that was designed for file deletes where the local metadata delete might fail after the remote object is already removed. * filer: propagate remote deletes for children in recursive folder deletion doBatchDeleteFolderMetaAndData iterated child files but only called NotifyUpdateEvent and collected chunks — it never called maybeDeleteFromRemote for individual children. This left orphaned objects in the remote backend when a directory containing remote-only files was recursively deleted. Also fix isFromOtherCluster being hardcoded to false in the recursive call to doBatchDeleteFolderMetaAndData for subdirectories. * filer: simplify pending remote deletion tracking to single index key Replace the double-bookkeeping scheme (individual KV entry per path + newline-delimited index key) with a single index key that stores paths directly. This removes the per-path KV writes/deletes, the base64 encoding round-trip, and the transaction overhead that was only needed to keep the two representations in sync. * filer: address review feedback on remote deletion flow - Distinguish missing remote config from client initialization failure in maybeDeleteFromRemote error messages. - Use a detached context (30s timeout) for pending-mark and pending-clear KV writes so they survive request cancellation after the remote object has already been deleted. - Emit NotifyUpdateEvent in reconcilePendingRemoteMetadataDeletions after a successful retry deletion so downstream watchers and replicas learn about the eventual metadata removal. * filer: remove background reconciliation for pending remote deletions The pending-mark/reconciliation machinery (KV index, mutex, background loop, detached contexts) handled the narrow case where the remote object was deleted but the subsequent local metadata delete failed. The client already receives the error and can retry — on retry the remote not-found is treated as success and the local delete proceeds normally. The added complexity (and new edge cases around NotifyUpdateEvent, multi-filer consistency during reconciliation, and context lifetime) is not justified for a transient store failure the caller already handles. Remove: loopProcessingRemoteMetadataDeletionPending, reconcilePendingRemoteMetadataDeletions, markRemoteMetadataDeletionPending, clearRemoteMetadataDeletionPending, listPendingRemoteMetadataDeletionPaths, encodePendingRemoteMetadataDeletionIndex, FindEntryLocal, and all associated constants, fields, and test infrastructure. * filer: fix test stubs and add early exit on child remote delete error - Refactor stubFilerStore to release lock before invoking callbacks and propagate callback errors, preventing potential deadlocks in tests - Implement ListDirectoryPrefixedEntries with proper prefix filtering instead of delegating to the unfiltered ListDirectoryEntries - Add continue after setting err on child remote delete failure in doBatchDeleteFolderMetaAndData to skip further processing of the failed entry * filer: propagate child remote delete error instead of silently continuing Replace `continue` with early `break` when maybeDeleteFromRemote fails for a child entry during recursive folder deletion. The previous `continue` skipped the error check at the end of the loop body, so a subsequent successful entry would overwrite err and the remote delete error was silently lost. Now the loop breaks, the existing error check returns the error, and NotifyUpdateEvent / chunk collection are correctly skipped for the failed entry. * filer: delete remote file when entry has Remote pointer, not only when remote-only Replace IsInRemoteOnly() guard with entry.Remote == nil check in maybeDeleteFromRemote. IsInRemoteOnly() requires zero local chunks and RemoteSize > 0, which incorrectly skips remote deletion for cached files (local chunks exist) and zero-byte remote objects (RemoteSize 0). The correct condition is whether the entry has a remote backing object at all. --------- Co-authored-by: Chris Lu <chris.lu@gmail.com>	2026-03-12 13:16:28 -07:00
Chris Lu	3f946fc0c0	mount: make metadata cache rebuilds snapshot-consistent (#8531) ... * filer: expose metadata events and list snapshots * mount: invalidate hot directory caches * mount: read hot directories directly from filer * mount: add sequenced metadata cache applier * mount: apply metadata responses through cache applier * mount: replay snapshot-consistent directory builds * mount: dedupe self metadata events * mount: factor directory build cleanup * mount: replace proto marshal dedup with composite key and ring buffer The dedup logic was doing a full deterministic proto.Marshal on every metadata event just to produce a dedup key. Replace with a cheap composite string key (TsNs|Directory|OldName|NewName). Also replace the sliding-window slice (which leaked the backing array unboundedly) with a fixed-size ring buffer that reuses the same array. * filer: remove mutex and proto.Clone from request-scoped MetadataEventSink MetadataEventSink is created per-request and only accessed by the goroutine handling the gRPC call. The mutex and double proto.Clone (once in Record, once in Last) were unnecessary overhead on every filer write operation. Store the pointer directly instead. * mount: skip proto.Clone for caller-owned metadata events Add ApplyMetadataResponseOwned that takes ownership of the response without cloning. Local metadata events (mkdir, create, flush, etc.) are freshly constructed and never shared, so the clone is unnecessary. * filer: only populate MetadataEvent on successful DeleteEntry Avoid calling eventSink.Last() on error paths where the sink may contain a partial event from an intermediate child deletion during recursive deletes. * mount: avoid map allocation in collectDirectoryNotifications Replace the map with a fixed-size array and linear dedup. There are at most 3 directories to notify (old parent, new parent, new child if directory), so a 3-element array avoids the heap allocation on every metadata event. * mount: fix potential deadlock in enqueueApplyRequest Release applyStateMu before the blocking channel send. Previously, if the channel was full (cap 128), the send would block while holding the mutex, preventing Shutdown from acquiring it to set applyClosed. * mount: restore signature-based self-event filtering as fast path Re-add the signature check that was removed when content-based dedup was introduced. Checking signatures is O(1) on a small slice and avoids enqueuing and processing events that originated from this mount instance. The content-based dedup remains as a fallback. * filer: send snapshotTsNs only in first ListEntries response The snapshot timestamp is identical for every entry in a single ListEntries stream. Sending it in every response message wastes wire bandwidth for large directories. The client already reads it only from the first response. * mount: exit read-through mode after successful full directory listing MarkDirectoryRefreshed was defined but never called, so directories that entered read-through mode (hot invalidation threshold) stayed there permanently, hitting the filer on every readdir even when cold. Call it after a complete read-through listing finishes. * mount: include event shape and full paths in dedup key The previous dedup key only used Names, which could collapse distinct rename targets. Include the event shape (C/D/U/R), source directory, new parent path, and both entry names so structurally different events are never treated as duplicates. * mount: drain pending requests on shutdown in runApplyLoop After receiving the shutdown sentinel, drain any remaining requests from applyCh non-blockingly and signal each with errMetaCacheClosed so callers waiting on req.done are released. * mount: include IsDirectory in synthetic delete events metadataDeleteEvent now accepts an isDirectory parameter so the applier can distinguish directory deletes from file deletes. Rmdir passes true, Unlink passes false. * mount: fall back to synthetic event when MetadataEvent is nil In mknod and mkdir, if the filer response omits MetadataEvent (e.g. older filer without the field), synthesize an equivalent local metadata event so the cache is always updated. * mount: make Flush metadata apply best-effort after successful commit After filer_pb.CreateEntryWithResponse succeeds, the entry is persisted. Don't fail the Flush syscall if the local metadata cache apply fails — log and invalidate the directory cache instead. Also fall back to a synthetic event when MetadataEvent is nil. * mount: make Rename metadata apply best-effort The rename has already succeeded on the filer by the time we apply the local metadata event. Log failures instead of returning errors that would be dropped by the caller anyway. * mount: make saveEntry metadata apply best-effort with fallback After UpdateEntryWithResponse succeeds, treat local metadata apply as non-fatal. Log and invalidate the directory cache on failure. Also fall back to a synthetic event when MetadataEvent is nil. * filer_pb: preserve snapshotTsNs on error in ReadDirAllEntriesWithSnapshot Return the snapshot timestamp even when the first page fails, so callers receive the snapshot boundary when partial data was received. * filer: send snapshot token for empty directory listings When no entries are streamed, send a final ListEntriesResponse with only SnapshotTsNs so clients always receive the snapshot boundary. * mount: distinguish not-found vs transient errors in lookupEntry Return fuse.EIO for non-not-found filer errors instead of unconditionally returning ENOENT, so transient failures don't masquerade as missing entries. * mount: make CacheRemoteObject metadata apply best-effort The file content has already been cached successfully. Don't fail the read if the local metadata cache update fails. * mount: use consistent snapshot for readdir in direct mode Capture the SnapshotTsNs from the first loadDirectoryEntriesDirect call and store it on the DirectoryHandle. Subsequent batch loads pass this stored timestamp so all batches use the same snapshot. Also export DoSeaweedListWithSnapshot so mount can use it directly with snapshot passthrough. * filer_pb: fix test fake to send SnapshotTsNs only on first response Match the server behavior: only the first ListEntriesResponse in a page carries the snapshot timestamp, subsequent entries leave it zero. * Fix nil pointer dereference in ListEntries stream consumers Remove the empty-directory snapshot-only response from ListEntries that sent a ListEntriesResponse with Entry==nil, which crashed every raw stream consumer that assumed resp.Entry is always non-nil. Also add defensive nil checks for resp.Entry in all raw ListEntries stream consumers across: S3 listing, broker topic lookup, broker topic config, admin dashboard, topic retention, hybrid message scanner, Kafka integration, and consumer offset storage. * Add nil guards for resp.Entry in remaining ListEntries stream consumers Covers: S3 object lock check, MQ management dashboard (version/ partition/offset loops), and topic retention version loop. * Make applyLocalMetadataEvent best-effort in Link and Symlink The filer operations already succeeded; failing the syscall because the local cache apply failed is wrong. Log a warning and invalidate the parent directory cache instead. * Make applyLocalMetadataEvent best-effort in Mkdir/Rmdir/Mknod/Unlink The filer RPC already committed; don't fail the syscall when the local metadata cache apply fails. Log a warning and invalidate the parent directory cache to force a re-fetch on next access. * flushFileMetadata: add nil-fallback for metadata event and best-effort apply Synthesize a metadata event when resp.GetMetadataEvent() is nil (matching doFlush), and make the apply best-effort with cache invalidation on failure. * Prevent double-invocation of cleanupBuild in doEnsureVisited Add a cleanupDone guard so the deferred cleanup and inline error-path cleanup don't both call DeleteFolderChildren/AbortDirectoryBuild. * Fix comment: signature check is O(n) not O(1) * Prevent deferred cleanup after successful CompleteDirectoryBuild Set cleanupDone before returning from the success path so the deferred context-cancellation check cannot undo a published build. * Invalidate parent directory caches on rename metadata apply failure When applyLocalMetadataEvent fails during rename, invalidate the source and destination parent directory caches so subsequent accesses trigger a re-fetch from the filer. * Add event nil-fallback and cache invalidation to Link and Symlink Synthesize metadata events when the server doesn't return one, and invalidate parent directory caches on apply failure. * Match requested partition when scanning partition directories Parse the partition range format (NNNN-NNNN) and match against the requested partition parameter instead of using the first directory. * Preserve snapshot timestamp across empty directory listings Initialize actualSnapshotTsNs from the caller-requested value so it isn't lost when the server returns no entries. Re-add the server-side snapshot-only response for empty directories (all raw stream consumers now have nil guards for Entry). * Fix CreateEntry error wrapping to support errors.Is/errors.As Use errors.New + %w instead of %v for resp.Error so callers can unwrap the underlying error. * Fix object lock pagination: only advance on non-nil entries Move entriesReceived inside the nil check so nil entries don't cause repeated ListEntries calls with the same lastFileName. * Guard Attributes nil check before accessing Mtime in MQ management * Do not send nil-Entry response for empty directory listings The snapshot-only ListEntriesResponse (with Entry == nil) for empty directories breaks consumers that treat any received response as an entry (Java FilerClient, S3 listing). The Go client-side DoSeaweedListWithSnapshot already preserves the caller-requested snapshot via actualSnapshotTsNs initialization, so the server-side send is unnecessary. * Fix review findings: subscriber dedup, invalidation normalization, nil guards, shutdown race - Remove self-signature early-return in processEventFn so all events flow through the applier (directory-build buffering sees self-originated events that arrive after a snapshot) - Normalize NewParentPath in collectEntryInvalidations to avoid duplicate invalidations when NewParentPath is empty (same-directory update) - Guard resp.Entry.Attributes for nil in admin_server.go and topic_retention.go to prevent panics on entries without attributes - Fix enqueueApplyRequest race with shutdown by using select on both applyCh and applyDone, preventing sends after the apply loop exits - Add cleanupDone check to deferred cleanup in meta_cache_init.go for clarity alongside the existing guard in cleanupBuild - Add empty directory test case for snapshot consistency * Propagate authoritative metadata event from CacheRemoteObjectToLocalCluster and generate client-side snapshot for empty directories - Add metadata_event field to CacheRemoteObjectToLocalClusterResponse proto so the filer-emitted event is available to callers - Use WithMetadataEventSink in the server handler to capture the event from NotifyUpdateEvent and return it on the response - Update filehandle_read.go to prefer the RPC's metadata event over a locally fabricated one, falling back to metadataUpdateEvent when the server doesn't provide one (e.g., older filers) - Generate a client-side snapshot cutoff in DoSeaweedListWithSnapshot when the server sends no snapshot (empty directory), so callers like CompleteDirectoryBuild get a meaningful boundary for filtering buffered events * Skip directory notifications for dirs being built to prevent mid-build cache wipe When a metadata event is buffered during a directory build, applyMetadataSideEffects was still firing noteDirectoryUpdate for the building directory. If the directory accumulated enough updates to become "hot", markDirectoryReadThrough would call DeleteFolderChildren, wiping entries that EnsureVisited had already inserted. The build would then complete and mark the directory cached with incomplete data. Fix by using applyMetadataSideEffectsSkippingBuildingDirs for buffered events, which suppresses directory notifications for dirs currently in buildingDirs while still applying entry invalidations. * Add test for directory notification suppression during active build TestDirectoryNotificationsSuppressedDuringBuild verifies that metadata events targeting a directory under active EnsureVisited build do NOT fire onDirectoryUpdate for that directory. In production, this prevents markDirectoryReadThrough from calling DeleteFolderChildren mid-build, which would wipe entries already inserted by the listing. The test inserts an entry during a build, sends multiple metadata events for the building directory, asserts no notifications fired for it, verifies the entry survives, and confirms buffered events are replayed after CompleteDirectoryBuild. * Fix create invalidations, build guard, event shape, context, and snapshot error path - collectEntryInvalidations: invalidate FUSE kernel cache on pure create events (OldEntry==nil && NewEntry!=nil), not just updates and deletes - completeDirectoryBuildNow: only call markCachedFn when an active build existed (state != nil), preventing an unpopulated directory from being marked as cached - Add metadataCreateEvent helper that produces a create-shaped event (NewEntry only, no OldEntry) and use it in mkdir, mknod, symlink, and hardlink create fallback paths instead of metadataUpdateEvent which incorrectly set both OldEntry and NewEntry - applyMetadataResponseEnqueue: use context.Background() for the queued mutation so a cancelled caller context cannot abort the apply loop mid-write - DoSeaweedListWithSnapshot: move snapshot initialization before ListEntries call so the error path returns the preserved snapshot instead of 0 * Fix review findings: test loop, cache race, context safety, snapshot consistency - Fix build test loop starting at i=1 instead of i=0, missing new-0.txt verification - Re-check IsDirectoryCached after cache miss to avoid ENOENT race with markDirectoryReadThrough - Use context.Background() in enqueueAndWait so caller cancellation can't abort build/complete mid-way - Pass dh.snapshotTsNs in skip-batch loadDirectoryEntriesDirect for snapshot consistency - Prefer resp.MetadataEvent over fallback in Unlink event derivation - Add comment on MetadataEventSink.Record single-event assumption * Fix empty-directory snapshot clock skew and build cancellation race Empty-directory snapshot: Remove client-side time.Now() synthesis when the server returns no entries. Instead return snapshotTsNs=0, and in completeDirectoryBuildNow replay ALL buffered events when snapshot is 0. This eliminates the clock-skew bug where a client ahead of the filer would filter out legitimate post-list events. Build cancellation: Use context.Background() for BeginDirectoryBuild and CompleteDirectoryBuild calls in doEnsureVisited, so errgroup cancellation doesn't cause enqueueAndWait to return early and trigger cleanupBuild while the operation is still queued. * Add tests for empty-directory build replay and cancellation resilience TestEmptyDirectoryBuildReplaysAllBufferedEvents: verifies that when CompleteDirectoryBuild receives snapshotTsNs=0 (empty directory, no server snapshot), ALL buffered events are replayed regardless of their TsNs values — no clock-skew-sensitive filtering occurs. TestBuildCompletionSurvivesCallerCancellation: verifies that once CompleteDirectoryBuild is enqueued, a cancelled caller context does not prevent the build from completing. The apply loop runs with context.Background(), so the directory becomes cached and buffered events are replayed even when the caller gives up waiting. * Fix directory subtree cleanup, Link rollback, test robustness - applyMetadataResponseLocked: when a directory entry is deleted or moved, call DeleteFolderChildren on the old path so cached descendants don't leak as stale entries. - Link: save original HardLinkId/Counter before mutation. If CreateEntryWithResponse fails after the source was already updated, rollback the source entry to its original state via UpdateEntry. - TestBuildCompletionSurvivesCallerCancellation: replace fixed time.Sleep(50ms) with a deadline-based poll that checks IsDirectoryCached in a loop, failing only after 2s timeout. - TestReadDirAllEntriesWithSnapshotEmptyDirectory: assert that ListEntries was actually invoked on the mock client so the test exercises the RPC path. - newMetadataEvent: add early return when both oldEntry and newEntry are nil to avoid emitting events with empty Directory. --------- Co-authored-by: Copilot <copilot@github.com>	2026-03-07 09:19:40 -08:00
Chris Lu	f9311a3422	s3api: fix static IAM policy enforcement after reload (#8532) ... * s3api: honor attached IAM policies over legacy actions * s3api: hydrate IAM policy docs during config reload * s3api: use policy-aware auth when listing buckets * credential: propagate context through filer_etc policy reads * credential: make legacy policy deletes durable * s3api: exercise managed policy runtime loader * s3api: allow static IAM users without session tokens * iam: deny unmatched attached policies under default allow * iam: load embedded policy files from filer store * s3api: require session tokens for IAM presigning * s3api: sync runtime policies into zero-config IAM * credential: respect context in policy file loads * credential: serialize legacy policy deletes * iam: align filer policy store naming * s3api: use authenticated principals for presigning * iam: deep copy policy conditions * s3api: require request creation in policy tests * filer: keep ReadInsideFiler as the context-aware API * iam: harden filer policy store writes * credential: strengthen legacy policy serialization test * credential: forward runtime policy loaders through wrapper * s3api: harden runtime policy merging * iam: require typed already-exists errors	2026-03-06 12:35:08 -08:00
Peter Dodd	16f2269a33	feat(filer): lazy metadata pulling (#8454) ... * Add remote storage index for lazy metadata pull Introduces remoteStorageIndex, which maintains a map of filer directory to remote storage client/location, refreshed periodically from the filer's mount mappings. Provides lazyFetchFromRemote, ensureRemoteEntryInFiler, and isRemoteBacked on S3ApiServer as integration points for handler-level work in a follow-up PR. Nothing is wired into the server yet. Made-with: Cursor * Add unit tests for remote storage index and wire field into S3ApiServer Adds tests covering isEmpty, findForPath (including longest-prefix resolution), and isRemoteBacked. Also removes a stray PR review annotation from the index file and adds the remoteStorageIdx field to S3ApiServer so the package compiles ahead of the wiring PR. Made-with: Cursor * Address review comments on remote storage index - Use filer_pb.CreateEntry helper so resp.Error is checked, not just the RPC error - Extract keepPrev closure to remove duplicated error-handling in refresh loop - Add comment explaining availability-over-consistency trade-off on filer save failure Made-with: Cursor * Move lazy metadata pull from S3 API to filer - Add maybeLazyFetchFromRemote in filer: on FindEntry miss, stat remote and CreateEntry when path is under a remote mount - Use singleflight for dedup; context guard prevents CreateEntry recursion - Availability-over-consistency: return in-memory entry if CreateEntry fails - Add longest-prefix test for nested mounts in remote_storage_test.go - Remove remoteStorageIndex, lazyFetchFromRemote, ensureRemoteEntryInFiler, doLazyFetch from s3api; filer now owns metadata operations - Add filer_lazy_remote_test.go with tests for hit, miss, not-found, CreateEntry failure, longest-prefix, and FindEntry integration Made-with: Cursor * Address review: fix context guard test, add FindMountDirectory comment, remove dead code Made-with: Cursor * Nitpicks: restore prev maker in registerStubMaker, instance-scope lazyFetchGroup, nil-check remoteEntry Made-with: Cursor * Fix remotePath when mountDir is root: ensure relPath has leading slash Made-with: Cursor * filer: decouple lazy-fetch persistence from caller context Use context.Background() inside the singleflight closure for CreateEntry so persistence is not cancelled when the winning request's context is cancelled. Fixes CreateEntry failing for all waiters when the first caller times out. Made-with: Cursor * filer: remove redundant Mode bitwise OR with zero Made-with: Cursor * filer: use bounded context for lazy-fetch persistence Replace context.Background() with context.WithTimeout(30s) and defer cancel() to prevent indefinite blocking and release resources. Made-with: Cursor * filer: use checked type assertion for singleflight result Made-with: Cursor * filer: rename persist context vars to avoid shadowing function parameter Made-with: Cursor	2026-03-03 13:01:10 -08:00
Chris Lu	e4b70c2521	go fix	2026-02-20 18:42:00 -08:00
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	2026-02-10 18:38:38 -08:00
Chris Lu	403592bb9f	Add Spark Iceberg catalog integration tests and CI support (#8242) ... * Add Spark Iceberg catalog integration tests and CI support Implement comprehensive integration tests for Spark with SeaweedFS Iceberg REST catalog: - Basic CRUD operations (Create, Read, Update, Delete) on Iceberg tables - Namespace (database) management - Data insertion, querying, and deletion - Time travel capabilities via snapshot versioning - Compatible with SeaweedFS S3 and Iceberg REST endpoints Tests mirror the structure of existing Trino integration tests but use Spark's Python SQL API and PySpark for testing. Add GitHub Actions CI job for spark-iceberg-catalog-tests in s3-tables-tests.yml to automatically run Spark integration tests on pull requests. * fmt * Fix Spark integration tests - code review feedback * go mod tidy * Add go mod tidy step to integration test jobs Add 'go mod tidy' step before test runs for all integration test jobs: - s3-tables-tests - iceberg-catalog-tests - trino-iceberg-catalog-tests - spark-iceberg-catalog-tests This ensures dependencies are clean before running tests. * Fix remaining Spark operations test issues Address final code review comments: Setup & Initialization: - Add waitForSparkReady() helper function that polls Spark readiness with backoff instead of hardcoded 10-second sleep - Extract setupSparkTestEnv() helper to reduce boilerplate duplication between TestSparkCatalogBasicOperations and TestSparkTimeTravel - Both tests now use helpers for consistent, reliable setup Assertions & Validation: - Make setup-critical operations (namespace, table creation, initial insert) use t.Fatalf instead of t.Errorf to fail fast - Validate setupSQL output in TestSparkTimeTravel and fail if not 'Setup complete' - Add validation after second INSERT in TestSparkTimeTravel: verify row count increased to 2 before time travel test - Add context to error messages with namespace and tableName params Code Quality: - Remove code duplication between test functions - All critical paths now properly validated - Consistent error handling throughout * Fix go vet errors in S3 Tables tests Fixes: 1. setup_test.go (Spark): - Add missing import: github.com/testcontainers/testcontainers-go/wait - Use wait.ForLog instead of undefined testcontainers.NewLogStrategy - Remove unused strings import 2. trino_catalog_test.go: - Use net.JoinHostPort instead of fmt.Sprintf for address formatting - Properly handles IPv6 addresses by wrapping them in brackets * Use weed mini for simpler SeaweedFS startup Replace complex multi-process startup (master, volume, filer, s3) with single 'weed mini' command that starts all services together. Benefits: - Simpler, more reliable startup - Single weed mini process vs 4 separate processes - Automatic coordination between components - Better port management with no manual coordination Changes: - Remove separate master, volume, filer process startup - Use weed mini with -master.port, -filer.port, -s3.port flags - Keep Iceberg REST as separate service (still needed) - Increase timeout to 15s for port readiness (weed mini startup) - Remove volumePort and filerProcess fields from TestEnvironment - Simplify cleanup to only handle two processes (mini, iceberg rest) * Clean up dead code and temp directory leaks Fixes: 1. Remove dead s3Process field and cleanup: - weed mini bundles S3 gateway, no separate process needed - Removed s3Process field from TestEnvironment - Removed unnecessary s3Process cleanup code 2. Fix temp config directory leak: - Add sparkConfigDir field to TestEnvironment - Store returned configDir in writeSparkConfig - Clean up sparkConfigDir in Cleanup() with os.RemoveAll - Prevents accumulation of temp directories in test runs 3. Simplify Cleanup: - Now handles only necessary processes (weed mini, iceberg rest) - Removes both seaweedfsDataDir and sparkConfigDir - Cleaner shutdown sequence * Use weed mini's built-in Iceberg REST and fix python binary Changes: - Add -s3.port.iceberg flag to weed mini for built-in Iceberg REST Catalog - Remove separate 'weed server' process for Iceberg REST - Remove icebergRestProcess field from TestEnvironment - Simplify Cleanup() to only manage weed mini + Spark - Add port readiness check for iceberg REST from weed mini - Set Spark container Cmd to '/bin/sh -c sleep 3600' to keep it running - Change python to python3 in container.Exec calls This simplifies to truly one all-in-one weed mini process (master, filer, s3, iceberg-rest) plus just the Spark container. * go fmt * clean up * bind on a non-loopback IP for container access, aligned Iceberg metadata saves/locations with table locations, and reworked Spark time travel to use TIMESTAMP AS OF with safe timestamp extraction. * shared mini start * Fixed internal directory creation under /buckets so .objects paths can auto-create without failing bucket-name validation, which restores table bucket object writes * fix path Updated table bucket objects to write under `/buckets/<bucket>` and saved Iceberg metadata there, adjusting Spark time-travel timestamp to committed_at +1s. Rebuilt the weed binary (`go install ./weed`) and confirmed passing tests for Spark and Trino with focused test commands. * Updated table bucket creation to stop creating /buckets/.objects and switched Trino REST warehouse to s3://<bucket> to match Iceberg layout. * Stabilize S3Tables integration tests * Fix timestamp extraction and remove dead code in bucketDir * Use table bucket as warehouse in s3tables tests * Update trino_blog_operations_test.go * adds the CASCADE option to handle any remaining table metadata/files in the schema directory * skip namespace not empty	2026-02-08 10:03:53 -08:00
Chris Lu	c284e51d20	fix: multipart upload ETag calculation (#8238) ... * fix multipart etag * address comments * clean up * clean up * optimization * address comments * unquoted etag * dedup * upgrade * clean * etag * return quoted tag * quoted etag * debug * s3api: unify ETag retrieval and quoting across handlers Refactor newListEntry to take *S3ApiServer and use getObjectETag, and update setResponseHeaders to use the same logic. This ensures consistent ETags are returned for both listing and direct access. * s3api: implement ListObjects deduplication for versioned buckets Handle duplicate entries between the main path and the .versions directory by prioritizing the latest version when bucket versioning is enabled. * s3api: cleanup stale main file entries during versioned uploads Add explicit deletion of pre-existing "main" files when creating new versions in versioned buckets. This prevents stale entries from appearing in bucket listings and ensures consistency. * s3api: fix cleanup code placement in versioned uploads Correct the placement of rm calls in completeMultipartUpload and putVersionedObject to ensure stale main files are properly deleted during versioned uploads. * s3api: improve getObjectETag fallback for empty ExtETagKey Ensure that when ExtETagKey exists but contains an empty value, the function falls through to MD5/chunk-based calculation instead of returning an empty string. * s3api: fix test files for new newListEntry signature Update test files to use the new newListEntry signature where the first parameter is *S3ApiServer. Created mockS3ApiServer to properly test owner display name lookup functionality. * s3api: use filer.ETag for consistent Md5 handling in getEtagFromEntry Change getEtagFromEntry fallback to use filer.ETag(entry) instead of filer.ETagChunks to ensure legacy entries with Attributes.Md5 are handled consistently with the rest of the codebase. * s3api: optimize list logic and fix conditional header logging - Hoist bucket versioning check out of per-entry callback to avoid repeated getVersioningState calls - Extract appendOrDedup helper function to eliminate duplicate dedup/append logic across multiple code paths - Change If-Match mismatch logging from glog.Errorf to glog.V(3).Infof and remove DEBUG prefix for consistency * s3api: fix test mock to properly initialize IAM accounts Fixed nil pointer dereference in TestNewListEntryOwnerDisplayName by directly initializing the IdentityAccessManagement.accounts map in the test setup. This ensures newListEntry can properly look up account display names without panicking. * cleanup * s3api: remove premature main file cleanup in versioned uploads Removed incorrect cleanup logic that was deleting main files during versioned uploads. This was causing test failures because it deleted objects that should have been preserved as null versions when versioning was first enabled. The deduplication logic in listing is sufficient to handle duplicate entries without deleting files during upload. * s3api: add empty-value guard to getEtagFromEntry Added the same empty-value guard used in getObjectETag to prevent returning quoted empty strings. When ExtETagKey exists but is empty, the function now falls through to filer.ETag calculation instead of returning "". * s3api: fix listing of directory key objects with matching prefix Revert prefix handling logic to use strings.TrimPrefix instead of checking HasPrefix with empty string result. This ensures that when a directory key object exactly matches the prefix (e.g. prefix="dir/", object="dir/"), it is correctly handled as a regular entry instead of being skipped or incorrectly processed as a common prefix. Also fixed missing variable definition. * s3api: refactor list inline dedup to use appendOrDedup helper Refactored the inline deduplication logic in listFilerEntries to use the shared appendOrDedup helper function. This ensures consistent behavior and reduces code duplication. * test: fix port allocation race in s3tables integration test Updated startMiniCluster to find all required ports simultaneously using findAvailablePorts instead of sequentially. This prevents race conditions where the OS reallocates a port that was just released, causing multiple services (e.g. Filer and Volume) to be assigned the same port and fail to start.	2026-02-06 21:54:43 -08:00
Chris Lu	066410dbd0	Fix S3 Gateway Read Failover #8076 (#8087) ... * fix s3 read failover #8076 - Implement cache invalidation in vidMapClient - Add retry logic in shared PrepareStreamContentWithThrottler - Update S3 Gateway to use FilerClient directly for invalidation support - Remove obsolete simpleMasterClient struct * improve observability for chunk re-lookup failures Added a warning log when volume location re-lookup fails after cache invalidation in PrepareStreamContentWithThrottler. * address code review feedback - Prevent infinite retry loops by comparing old/new URLs before retry - Update fileId2Url map after successful re-lookup for subsequent references - Add comprehensive test coverage for failover logic - Add tests for InvalidateCache method * Fix: prevent data duplication in stream retry and improve VidMap robustness * Cleanup: remove redundant check in InvalidateCache	2026-01-22 14:07:24 -08:00
Chris Lu	8880f9932f	filer: auto clean empty implicit s3 folders (#8051) ... * filer: auto clean empty s3 implicit folders Explicitly tag implicitly created S3 folders (parent directories from object uploads) with 'Seaweed-X-Amz-Implicit-Dir'. Update EmptyFolderCleaner to check for this attribute and cache the result efficiently. * filer: correctly handle nil attributes in empty folder cleaner cache * filer: refine implicit tagging logic Prevent tagging buckets as implicit directories. Reduce code duplication. * filer: safeguard GetEntryAttributes against nil entry and not found error * filer: move ErrNotFound handling to EmptyFolderCleaner * filer: add comment to explain level > 3 check for implicit directories	2026-01-17 22:10:15 -08:00
Chris Lu	796a911cb3	Prevent bucket renaming in filer, fuse mount, and S3 (#8048) ... * prevent bucket renaming in filer, fuse mount, s3 * refactor CanRename to support context propagation * harden bucket rename validation to fail closed on find error	2026-01-16 19:48:09 -08:00
Walnuts	691aea84c3	feat: add TLS configuration options for Cassandra2 store (#7998) ... * feat: add TLS configuration options for Cassandra2 store Signed-off-by: walnuts1018 <r.juglans.1018@gmail.com> * fix: use 9142 port in tls connection Signed-off-by: walnuts1018 <r.juglans.1018@gmail.com> * Align the setting field names with gocql's SSLOpts. Signed-off-by: walnuts1018 <r.juglans.1018@gmail.com> * Removed: store.cluster.Port = 9142 * chore: update gocql dependency to v2 * refactor: improve Cassandra TLS configuration and port logic * docs: update filer.toml scaffold with ssl_enable_host_verification --------- Signed-off-by: walnuts1018 <r.juglans.1018@gmail.com> Co-authored-by: Chris Lu <chris.lu@gmail.com>	2026-01-14 17:59:59 -08:00
Chris Lu	379c032868	Fix chown Input/output error on large file sets (#7996) ... * Fix chown Input/output error on large file sets (Fixes #7911) Implemented retry logic for MySQL/MariaDB backend to handle transient errors like deadlocks and timeouts. * Fix syntax error: missing closing brace * Refactor: Use %w for error wrapping and errors.As for extraction * Fix: Disable retry logic inside transactions	2026-01-09 18:02:59 -08:00
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>	2026-01-07 13:06:08 -08:00
Alasdair Macmillan	9778b9589e	Fix unaligned 64-bit atomic operation on ARM32 (#7958) (#7959)	2026-01-04 11:34:17 -08:00
cduk	568f1fe5b1	fix: include DiskType in metadata log volume assignment (#7918) ... When writing metadata logs to /topics/.system/log, the filer was not respecting the disk type configuration from path-specific rules (fs.configure). This caused volume assignment failures when volume servers used a specific disk type (e.g., "ssd") because the assign request defaulted to empty disk type. The fix adds DiskType to the VolumeAssignRequest in the filer's metadata log write path, ensuring that path-specific disk type configurations are properly honored for internal system writes. Fixes errors like: "metadata log write failed /topics/.system/log/...: AssignVolume: failed to find writable volumes for collection" Signed-off-by: Charles Darke <s.cduk@toodevious.com> Co-authored-by: Charles Darke <s.cduk@toodevious.com>	2025-12-30 17:32:33 -08:00
Chris Lu	288ba5fec8	mount: let filer handle chunk deletion decision (#7900) ... * mount: let filer handle chunk deletion decision Remove chunk deletion decision from FUSE mount's Unlink operation. Previously, the mount decided whether to delete chunks based on its locally cached entry's HardLinkCounter, which could be stale. Now always pass isDeleteData=true and let the filer make the authoritative decision based on its own data. This prevents potential inconsistencies when: - The FUSE mount's cached entry is stale - Race conditions occur between multiple mounts - Direct filer operations change hard link counts * filer: check hard link counter before deleting chunks When deleting an entry, only delete the underlying chunks if: 1. It is not a hard link 2. OR it is the last hard link (counter <= 1) This protects against data loss when a client (like FUSE mount) requests chunk deletion for a file that has multiple hard links.	2025-12-28 23:22:13 -08:00
steve.wei	e439e33888	fix(filer): check error from FindEntry (#7878) ... * fix(filer): check error from FindEntry * remove --------- Co-authored-by: Chris Lu <chris.lu@gmail.com>	2025-12-25 11:28:31 -08:00
Chris Lu	1261e93ef2	fix: comprehensive go vet error fixes and add CI enforcement (#7861) ... * fix: use keyed fields in struct literals - Replace unsafe reflect.StringHeader/SliceHeader with safe unsafe.String/Slice (weed/query/sqltypes/unsafe.go) - Add field names to Type_ScalarType struct literals (weed/mq/schema/schema_builder.go) - Add Duration field name to FlexibleDuration struct literals across test files - Add field names to bson.D struct literals (weed/filer/mongodb/mongodb_store_kv.go) Fixes go vet warnings about unkeyed struct literals. * fix: remove unreachable code - Remove unreachable return statements after infinite for loops - Remove unreachable code after if/else blocks where all paths return - Simplify recursive logic by removing unnecessary for loop (inode_to_path.go) - Fix Type_ScalarType literal to use enum value directly (schema_builder.go) - Call onCompletionFn on stream error (subscribe_session.go) Files fixed: - weed/query/sqltypes/unsafe.go - weed/mq/schema/schema_builder.go - weed/mq/client/sub_client/connect_to_sub_coordinator.go - weed/filer/redis3/ItemList.go - weed/mq/client/agent_client/subscribe_session.go - weed/mq/broker/broker_grpc_pub_balancer.go - weed/mount/inode_to_path.go - weed/util/skiplist/name_list.go * fix: avoid copying lock values in protobuf messages - Use proto.Merge() instead of direct assignment to avoid copying sync.Mutex in S3ApiConfiguration (iamapi_server.go) - Add explicit comments noting that channel-received values are already copies before taking addresses (volume_grpc_client_to_master.go) The protobuf messages contain sync.Mutex fields from the message state, which should not be copied. Using proto.Merge() properly merges messages without copying the embedded mutex. * fix: correct byte array size for uint32 bit shift operations The generateAccountId() function only needs 4 bytes to create a uint32 value. Changed from allocating 8 bytes to 4 bytes to match the actual usage. This fixes go vet warning about shifting 8-bit values (bytes) by more than 8 bits. * fix: ensure context cancellation on all error paths In broker_client_subscribe.go, ensure subscriberCancel() is called on all error return paths: - When stream creation fails - When partition assignment fails - When sending initialization message fails This prevents context leaks when an error occurs during subscriber creation. * fix: ensure subscriberCancel called for CreateFreshSubscriber stream.Send error Ensure subscriberCancel() is called when stream.Send fails in CreateFreshSubscriber. * ci: add go vet step to prevent future lint regressions - Add go vet step to GitHub Actions workflow - Filter known protobuf lock warnings (MessageState sync.Mutex) These are expected in generated protobuf code and are safe - Prevents accumulation of go vet errors in future PRs - Step runs before build to catch issues early * fix: resolve remaining syntax and logic errors in vet fixes - Fixed syntax errors in filer_sync.go caused by missing closing braces - Added missing closing brace for if block and function - Synchronized fixes to match previous commits on branch * fix: add missing return statements to daemon functions - Add 'return false' after infinite loops in filer_backup.go and filer_meta_backup.go - Satisfies declared bool return type signatures - Maintains consistency with other daemon functions (runMaster, runFilerSynchronize, runWorker) - While unreachable, explicitly declares the return satisfies function signature contract * fix: add nil check for onCompletionFn in SubscribeMessageRecord - Check if onCompletionFn is not nil before calling it - Prevents potential panic if nil function is passed - Matches pattern used in other callback functions * docs: clarify unreachable return statements in daemon functions - Add comments documenting that return statements satisfy function signature - Explains that these returns follow infinite loops and are unreachable - Improves code clarity for future maintainers	2025-12-23 14:48:50 -08:00
chrislu	4a764dbb37	fmt	2025-12-19 15:33:16 -08:00
G-OD	504b258258	s3: fix remote object not caching (#7790) ... * s3: fix remote object not caching * s3: address review comments for remote object caching - Fix leading slash in object name by using strings.TrimPrefix - Return cached entry from CacheRemoteObjectToLocalCluster to get updated local chunk locations - Reuse existing helper function instead of inline gRPC call * s3/filer: add singleflight deduplication for remote object caching - Add singleflight.Group to FilerServer to deduplicate concurrent cache operations - Wrap CacheRemoteObjectToLocalCluster with singleflight to ensure only one caching operation runs per object when multiple clients request the same file - Add early-return check for already-cached objects - S3 API calls filer gRPC with timeout and graceful fallback on error - Clear negative bucket cache when bucket is created via weed shell - Add integration tests for remote cache with singleflight deduplication This benefits all clients (S3, HTTP, Hadoop) accessing remote-mounted objects by preventing redundant cache operations and improving concurrent access performance. Fixes: https://github.com/seaweedfs/seaweedfs/discussions/7599 * fix: data race in concurrent remote object caching - Add mutex to protect chunks slice from concurrent append - Add mutex to protect fetchAndWriteErr from concurrent read/write - Fix incorrect error check (was checking assignResult.Error instead of parseErr) - Rename inner variable to avoid shadowing fetchAndWriteErr * fix: address code review comments - Remove duplicate remote caching block in GetObjectHandler, keep only singleflight version - Add mutex protection for concurrent chunk slice and error access (data race fix) - Use lazy initialization for S3 client in tests to avoid panic during package load - Fix markdown linting: add language specifier to code fence, blank lines around tables - Add 'all' target to Makefile as alias for test-with-server - Remove unused 'util' import * style: remove emojis from test files * fix: add defensive checks and sort chunks by offset - Add nil check and type assertion check for singleflight result - Sort chunks by offset after concurrent fetching to maintain file order * fix: improve test diagnostics and path normalization - runWeedShell now returns error for better test diagnostics - Add all targets to .PHONY in Makefile (logs-primary, logs-remote, health) - Strip leading slash from normalizedObject to avoid double slashes in path --------- Co-authored-by: chrislu <chris.lu@gmail.com> Co-authored-by: Chris Lu <chrislusf@users.noreply.github.com>	2025-12-16 12:41:04 -08:00
Chris Lu	5a03b5538f	filer: improve FoundationDB performance by disabling batch by default (#7770) ... * filer: improve FoundationDB performance by disabling batch by default This PR addresses a performance issue where FoundationDB filer was achieving only ~757 ops/sec with 12 concurrent S3 clients, despite FDB being capable of 17,000+ ops/sec. Root cause: The write batcher was waiting up to 5ms for each operation to batch, even though S3 semantics require waiting for durability confirmation. This added artificial latency that defeated the purpose of batching. Changes: - Disable write batching by default (batch_enabled = false) - Each write now commits immediately in its own transaction - Reduce batch interval from 5ms to 1ms when batching is enabled - Add batch_enabled config option to toggle behavior - Improve batcher to collect available ops without blocking - Add benchmarks comparing batch vs no-batch performance Benchmark results (16 concurrent goroutines): - With batch: 2,924 ops/sec (342,032 ns/op) - Without batch: 4,625 ops/sec (216,219 ns/op) - Improvement: +58% faster Configuration: - Default: batch_enabled = false (optimal for S3 PUT latency) - For bulk ingestion: set batch_enabled = true Also fixes ARM64 Docker test setup (shell compatibility, fdbserver path). * fix: address review comments - use atomic counter and remove duplicate batcher - Use sync/atomic.Uint64 for unique filenames in concurrent benchmarks - Remove duplicate batcher creation in createBenchmarkStoreWithBatching (initialize() already creates batcher when batchEnabled=true) * fix: add realistic default values to benchmark store helper Set directoryPrefix, timeout, and maxRetryDelay to reasonable defaults for more realistic benchmark conditions.	2025-12-15 13:03:34 -08:00
Chris Lu	59a7c40043	Add keyPrefix support for TiKV store (#7756) ... * Add keyPrefix support for TiKV store Similar to the Redis keyPrefix feature (#7299), this adds keyPrefix support for TiKV stores to enable sharing a single TiKV cluster as metadata store for multitenant SeaweedFS clusters. Changes: - Add keyPrefix field to TikvStore struct - Update Initialize function to read keyPrefix from config - Add getKey method to prepend prefix to all keys - Update generateKey, getNameFromKey, and genDirectoryKeyPrefix methods to be store receiver methods and handle key prefixing - Update filer.toml scaffold with keyPrefix configuration option Fixes #7752 * Fix potential slice corruption in getKey method Use a new slice with proper capacity to avoid modifying the underlying array of store.keyPrefix when appending. * Add keyPrefix validation and defensive bounds check - Add validation in Initialize to reject keyPrefix longer than 256 bytes - Add bounds check in getNameFromKey to prevent panic on malformed keys * Update weed/filer/tikv/tikv_store.go Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> * Update weed/command/scaffold/filer.toml Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> --------- Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>	2025-12-14 23:40:01 -08:00
Chris Lu	e8b7347031	Reduce memory allocations in hot paths (#7725) ... * filer: reduce allocations in MatchStorageRule Optimize MatchStorageRule to avoid allocations in common cases: - Return singleton emptyPathConf when no rules match (zero allocations) - Return existing rule directly when only one rule matches (zero allocations) - Only allocate and merge when multiple rules match (rare case) Based on heap profile analysis showing 111MB allocated from 1.64M calls to this function during 180 seconds of operation. * filer: add fast path for getActualStore when no path-specific stores Add hasPathSpecificStore flag to FilerStoreWrapper to skip the MatchPrefix() call and []byte(path) conversion when no path-specific stores are configured (the common case). Based on heap profile analysis showing 1.39M calls to this function during 180 seconds of operation, each requiring a string-to-byte slice conversion for the MatchPrefix call. * filer/foundationdb: use sync.Pool for tuple allocation in genKey Use sync.Pool to reuse tuple.Tuple slices in genKey(), reducing allocation overhead for every FoundationDB operation. Based on heap profile analysis showing 102MB allocated from 1.79M calls to genKey() during 180 seconds of operation. The Pack() call still allocates internally, but this reduces the tuple slice allocation overhead by ~50%. * filer: use sync.Pool for protobuf Entry and FuseAttributes Add pooling for filer_pb.Entry and filer_pb.FuseAttributes in EncodeAttributesAndChunks and DecodeAttributesAndChunks to reduce allocations during filer store operations. Changes: - Add pbEntryPool with pre-allocated FuseAttributes - Add EntryAttributeToExistingPb for in-place attribute conversion - Update ToExistingProtoEntry to reuse existing Attributes when available Based on heap profile showing: - EncodeAttributesAndChunks: 69.5MB cumulative - DecodeAttributesAndChunks: 46.5MB cumulative - EntryAttributeToPb: 47.5MB flat allocations * log_buffer: use sync.Pool for LogEntry in readTs Add logEntryPool to reuse filer_pb.LogEntry objects in readTs(), which is called frequently during binary search in ReadFromBuffer. This function only needs the TsNs field from the unmarshaled entry, so pooling the LogEntry avoids repeated allocations. Based on heap profile showing readTs with 188MB cumulative allocations from timestamp lookups during log buffer reads. * pb: reduce gRPC metadata allocations in interceptor Optimize requestIDUnaryInterceptor and WithGrpcClient to reduce metadata allocations on every gRPC request: - Use AppendToOutgoingContext instead of NewOutgoingContext + New() This avoids creating a new map[string]string for single key-value pairs - Check FromIncomingContext return value before using metadata Based on heap profile showing metadata operations contributing 0.45GB (10.5%) of allocations, with requestIDUnaryInterceptor being the main source at 0.44GB cumulative. Expected reduction: ~0.2GB from avoiding map allocations per request. * filer/log_buffer: address code review feedback - Use proto.Reset() instead of manual field clearing in resetLogEntry for more idiomatic and comprehensive state clearing - Add resetPbEntry() call before pool return in error path for consistency with success path in DecodeAttributesAndChunks * log_buffer: reduce PreviousBufferCount from 32 to 4 Reduce the number of retained previous buffers from 32 to 4. Each buffer is 8MB, so this reduces the maximum retained memory from 256MB to 32MB for previous buffers. Most subscribers catch up quickly, so 4 buffers (32MB) should be sufficient while significantly reducing memory footprint. * filer/foundationdb: use defer for tuple pool cleanup in genKey Refactor genKey to use defer for returning the pooled tuple. This ensures the pooled object is always returned even if store.seaweedfsDir.Pack panics, making the code more robust. Also simplifies the code by removing the temporary variable. * filer: early-stop MatchStorageRule prescan after 2 matches Stop the prescan callback after finding 2 matches since we only need to know if there are 0, 1, or multiple matches. This avoids unnecessarily scanning the rest of the trie when many rules exist. * fix: address critical code review issues filer_conf.go: - Remove mutable singleton emptyPathConf that could corrupt shared state - Return fresh copy for no-match case and cloned copy for single-match case - Add clonePathConf helper to create shallow copies safely grpc_client_server.go: - Remove incorrect AppendToOutgoingContext call in server interceptor (that API is for outbound client calls, not server-side handlers) - Rely on request_id.Set and SetTrailer for request ID propagation * fix: treat FilerConf_PathConf as immutable Fix callers that were incorrectly mutating the returned PathConf: - filer_server_handlers_write.go: Use local variable for MaxFileNameLength instead of mutating the shared rule - command_s3_bucket_quota_check.go: Create new PathConf explicitly when modifying config instead of mutating the returned one This allows MatchStorageRule to safely return the singleton or direct references without copying, restoring the memory optimization. Callers must NOT mutate the returned *FilerConf_PathConf. * filer: add ClonePathConf helper for creating mutable copies Add reusable ClonePathConf function that creates a mutable copy of a PathConf. This is useful when callers need to modify config before calling SetLocationConf. Update command_s3_bucket_quota_check.go to use the new helper. Also fix redundant return statement in DeleteLocationConf. * fmt * filer: fix protobuf pool reset to clear internal fields Address code review feedback: 1. resetPbEntry/resetFuseAttributes: Use struct assignment (*e = T{}) instead of field-by-field reset to clear protobuf internal fields (unknownFields, sizeCache) that would otherwise accumulate across pool reuses, causing data corruption or memory bloat. 2. EntryAttributeToExistingPb: Add nil guard for attr parameter to prevent panic if caller passes nil. * log_buffer: reset logEntry before pool return in error path For consistency with success path, reset the logEntry before putting it back in the pool in the error path. This prevents the pooled object from holding references to partially unmarshaled data. * filer: optimize MatchStorageRule and document ClonePathConf 1. Avoid double []byte(path) conversion in multi-match case by converting once and reusing pathBytes. 2. Add IMPORTANT comment to ClonePathConf documenting that it must be kept in sync with filer_pb.FilerConf_PathConf fields when the protobuf evolves. * filer/log_buffer: fix data race and use defer for pool cleanup 1. entry_codec.go EncodeAttributesAndChunks: Fix critical data race - proto.Marshal may return a slice sharing memory with the message. Copy the data before returning message to pool to prevent corruption. 2. entry_codec.go DecodeAttributesAndChunks: Use defer for cleaner pool management, ensuring message is always returned to pool. 3. log_buffer.go readTs: Use defer for pool cleanup, removing duplicated resetLogEntry/Put calls in success and error paths. * filer: fix ClonePathConf field order and add comprehensive test 1. Fix field order in ClonePathConf to match protobuf struct definition (WormGracePeriodSeconds before WormRetentionTimeSeconds). 2. Add TestClonePathConf that constructs a fully-populated PathConf, calls ClonePathConf, and asserts equality of all exported fields. This will catch future schema drift when new fields are added. 3. Add TestClonePathConfNil to verify nil handling. * filer: use reflection in ClonePathConf test to detect schema drift Replace hardcoded field comparisons with reflection-based comparison. This automatically catches: 1. New fields added to the protobuf but not copied in ClonePathConf 2. Missing non-zero test values for any exported field The test iterates over all exported fields using reflect and compares src vs clone values, failing if any field differs. * filer: update EntryAttributeToExistingPb comment to reflect nil handling The function safely handles nil attr by returning early, but the comment incorrectly stated 'attr must not be nil'. Update comment to accurately describe the defensive behavior. * Fix review feedback: restore request ID propagation and remove redundant resets 1. grpc_client_server.go: Restore AppendToOutgoingContext for request ID so handlers making downstream gRPC calls will automatically propagate the request ID to downstream services. 2. entry_codec.go: Remove redundant resetPbEntry calls after Get. The defer block ensures reset before Put, so next Get receives clean object. 3. log_buffer.go: Remove redundant resetLogEntry call after Get for same reason - defer already handles reset before Put.	2025-12-12 12:51:48 -08:00
Chris Lu	c153420022	filer: add write batching for FoundationDB store to improve throughput (#7708) ... This addresses issue #7699 where FoundationDB filer store had low throughput (~400-500 obj/s) due to each write operation creating a separate transaction. Changes: - Add writeBatcher that collects multiple writes into batched transactions - New config options: batch_size (default: 100), batch_interval (default: 5ms) - Batching provides ~5.7x throughput improvement (from ~456 to ~2600 obj/s) Benchmark results with different batch sizes: - batch_size=1: ~456 obj/s (baseline, no batching) - batch_size=10: ~2621 obj/s (5.7x improvement) - batch_size=16: ~2514 obj/s (5.5x improvement) - batch_size=100: ~2617 obj/s (5.7x improvement) - batch_size=1000: ~2593 obj/s (5.7x improvement) The batch_interval timer (5ms) ensures writes are flushed promptly even when batch is not full, providing good latency characteristics. Addressed review feedback: - Changed wait=false to wait=true in UpdateEntry/DeleteEntry to properly propagate errors to callers - Fixed timer reset race condition by stopping and draining before reset Fixes #7699	2025-12-10 12:42:09 -08:00
Chris Lu	0cd9f34177	mount: improve EnsureVisited performance with dedup, parallelism, and batching (#7697) ... * mount: add singleflight to deduplicate concurrent EnsureVisited calls When multiple goroutines access the same uncached directory simultaneously, they would all make redundant network requests to the filer. This change uses singleflight.Group to ensure only one goroutine fetches the directory entries while others wait for the result. This fixes a race condition where concurrent lookups or readdir operations on the same uncached directory would: 1. Make duplicate network requests to the filer 2. Insert duplicate entries into LevelDB cache 3. Waste CPU and network bandwidth * mount: fetch parent directories in parallel during EnsureVisited Previously, when accessing a deep path like /a/b/c/d, the parent directories were fetched serially from target to root. This change: 1. Collects all uncached directories from target to root first 2. Fetches them all in parallel using errgroup 3. Relies on singleflight (from previous commit) for deduplication This reduces latency when accessing deep uncached paths, especially in high-latency network environments where parallel requests can significantly improve performance. * mount: add batch inserts for LevelDB meta cache When populating the meta cache from filer, entries were inserted one-by-one into LevelDB. This change: 1. Adds BatchInsertEntries method to LevelDBStore that uses LevelDB's native batch write API 2. Updates MetaCache to keep a direct reference to the LevelDB store for batch operations 3. Modifies doEnsureVisited to collect entries and insert them in batches of 100 entries Batch writes are more efficient because: - Reduces number of individual write operations - Reduces disk syncs - Improves throughput for large directories * mount: fix potential nil dereference in MarkChildrenCached Add missing check for inode existence in inode2path map before accessing the InodeEntry. This prevents a potential nil pointer dereference if the inode exists in path2inode but not in inode2path (which could happen due to race conditions or bugs). This follows the same pattern used in IsChildrenCached which properly checks for existence before accessing the entry. * mount: fix batch flush when last entry is hidden The previous batch insert implementation relied on the isLast flag to flush remaining entries. However, if the last entry is a hidden system entry (like 'topics' or 'etc' in root), the callback returns early and the remaining entries in the batch are never flushed. Fix by: 1. Only flush when batch reaches threshold inside the callback 2. Flush any remaining entries after ReadDirAllEntries completes 3. Use error wrapping instead of logging+returning to avoid duplicate logs 4. Create new slice after flush to allow GC of flushed entries 5. Add documentation for batchInsertSize constant This ensures all entries are properly inserted regardless of whether the last entry is hidden, and prevents memory retention issues. * mount: add context support for cancellation in EnsureVisited Thread context.Context through the batch insert call chain to enable proper cancellation and timeout support: 1. Use errgroup.WithContext() so if one fetch fails, others are cancelled 2. Add context parameter to BatchInsertEntries for consistency with InsertEntry 3. Pass context to ReadDirAllEntries for cancellation during network calls 4. Check context cancellation before starting work in doEnsureVisited 5. Use %w for error wrapping to preserve error types for inspection This prevents unnecessary work when one directory fetch fails and makes the batch operations consistent with the existing context-aware APIs.	2025-12-09 23:44:15 -08:00
Chris Lu	1b13324fb7	fix: skip log files with deleted volumes in filer backup (#7692) ... fix: skip log files with deleted volumes in filer backup (#3720) When filer.backup or filer.meta.backup resumes after being stopped, it may encounter persisted log files stored on volumes that have since been deleted (via volume.deleteEmpty -force). Previously, this caused the backup to get stuck in an infinite retry loop with 'volume X not found' errors. This fix catches 'volume not found' errors when reading log files and skips the problematic file instead of failing. The backup will now: - Log a warning about the missing volume - Skip the problematic log file - Continue with the next log file, allowing progress The VolumeNotFoundPattern regex was already defined but never used - this change puts it to use. Fixes #3720	2025-12-09 19:03:15 -08:00
Chris Lu	5c1de633cb	mount: improve read throughput with parallel chunk fetching (#7627) ... * filer: remove lock contention during chunk download This addresses issue #7504 where a single weed mount FUSE instance does not fully utilize node network bandwidth when reading large files. The SingleChunkCacher was holding a mutex during the entire HTTP download, causing readers to block until the download completed. This serialized chunk reads even when multiple goroutines were downloading in parallel. Changes: - Add sync.Cond to SingleChunkCacher for efficient waiting - Move HTTP download outside the critical section in startCaching() - Use condition variable in readChunkAt() to wait for download completion - Add isComplete flag to track download state Now multiple chunk downloads can proceed truly in parallel, and readers wait efficiently using the condition variable instead of blocking on a mutex held during I/O operations. Ref: #7504 * filer: parallel chunk fetching within doReadAt This addresses issue #7504 by enabling parallel chunk downloads within a single read operation. Previously, doReadAt() processed chunks sequentially in a loop, meaning each chunk had to be fully downloaded before the next one started. This left significant network bandwidth unused when chunks resided on different volume servers. Changes: - Collect all chunk read tasks upfront - Use errgroup to fetch multiple chunks in parallel - Each chunk reads directly into its correct buffer position - Limit concurrency to prefetchCount (min 4) to avoid overwhelming the system - Handle gaps and zero-filling before parallel fetch - Trigger prefetch after parallel reads complete For a read spanning N chunks on different volume servers, this can now utilize up to N times the bandwidth of a single connection. Ref: #7504 * http: direct buffer read to reduce memory copies This addresses issue #7504 by reducing memory copy overhead during chunk downloads. Previously, RetriedFetchChunkData used ReadUrlAsStream which: 1. Allocated a 64KB intermediate buffer 2. Read data in 64KB chunks 3. Called a callback to copy each chunk to the destination For a 16MB chunk, this meant 256 copy operations plus the callback overhead. Profiling showed significant time spent in memmove. Changes: - Add readUrlDirectToBuffer() that reads directly into the destination - Add retriedFetchChunkDataDirect() for unencrypted, non-gzipped chunks - Automatically use direct read path when possible (cipher=nil, gzip=false) - Use http.NewRequestWithContext for proper cancellation For unencrypted chunks (the common case), this eliminates the intermediate buffer entirely, reading HTTP response bytes directly into the final destination buffer. Ref: #7504 * address review comments - Use channel (done) instead of sync.Cond for download completion signaling This integrates better with context cancellation patterns - Remove redundant groupErr check in reader_at.go (errors are already captured in task.err) - Remove buggy URL encoding logic from retriedFetchChunkDataDirect (The existing url.PathEscape on full URL is a pre-existing bug that should be fixed separately) * address review comments (round 2) - Return io.ErrUnexpectedEOF when HTTP response is truncated This prevents silent data corruption from incomplete reads - Simplify errgroup error handling by using g.Wait() error directly Remove redundant task.err field and manual error aggregation loop - Define minReadConcurrency constant instead of magic number 4 Improves code readability and maintainability Note: Context propagation to startCaching() is intentionally NOT changed. The downloaded chunk is a shared resource that may be used by multiple readers. Using context.Background() ensures the download completes even if one reader cancels, preventing data loss for other waiting readers. * http: inject request ID for observability in direct read path Add request_id.InjectToRequest() call to readUrlDirectToBuffer() for consistency with ReadUrlAsStream path. This ensures full-chunk reads carry the same tracing/correlation headers for server logs and metrics. * filer: consistent timestamp handling in sequential read path Use max(ts, task.chunk.ModifiedTsNs) in sequential path to match parallel path behavior. Also update ts before error check so that on failure, the returned timestamp reflects the max of all chunks processed so far. * filer: document why context.Background() is used in startCaching Add comment explaining the intentional design decision: the downloaded chunk is a shared resource that may be used by multiple concurrent readers. Using context.Background() ensures the download completes even if one reader cancels, preventing errors for other waiting readers. * filer: propagate context for reader cancellation Address review comment: pass context through ReadChunkAt call chain so that a reader can cancel its wait for a download. The key distinction is: - Download uses context.Background() - shared resource, always completes - Reader wait uses request context - can be cancelled individually If a reader cancels, it stops waiting and returns ctx.Err(), but the download continues to completion for other readers waiting on the same chunk. This properly handles the shared resource semantics while still allowing individual reader cancellation. * filer: use defer for close(done) to guarantee signal on panic Move close(s.done) to a defer statement at the start of startCaching() to ensure the completion signal is always sent, even if an unexpected panic occurs. This prevents readers from blocking indefinitely. * filer: remove unnecessary code - Remove close(s.cacheStartedCh) in destroy() - the channel is only used for one-time synchronization, closing it provides no benefit - Remove task := task loop variable capture - Go 1.22+ fixed loop variable semantics, this capture is no longer necessary (go.mod specifies Go 1.24.0) * filer: restore fallback to chunkCache when cacher returns no data Fix critical issue where ReadChunkAt would return 0,nil immediately if SingleChunkCacher couldn't provide data for the requested offset, without trying the chunkCache fallback. Now if cacher.readChunkAt returns n=0 and err=nil, we fall through to try chunkCache. * filer: add comprehensive tests for ReaderCache Tests cover: - Context cancellation while waiting for download - Fallback to chunkCache when cacher returns n=0, err=nil - Multiple concurrent readers waiting for same chunk - Partial reads at different offsets - Downloader cleanup when exceeding cache limit - Done channel signaling (no hangs on completion) * filer: prioritize done channel over context cancellation If data is already available (done channel closed), return it even if the reader's context is also cancelled. This avoids unnecessary errors when the download has already completed. * filer: add lookup error test and document test limitations Add TestSingleChunkCacherLookupError to test error handling when lookup fails. Document that full HTTP integration tests for SingleChunkCacher require global HTTP client initialization which is complex in unit tests. The download path is tested via FUSE integration tests. * filer: add tests that exercise SingleChunkCacher concurrency logic Add tests that use blocking lookupFileIdFn to exercise the actual SingleChunkCacher wait/cancellation logic: - TestSingleChunkCacherContextCancellationDuringLookup: tests reader cancellation while lookup is blocked - TestSingleChunkCacherMultipleReadersWaitForDownload: tests multiple readers waiting on the same download - TestSingleChunkCacherOneReaderCancelsOthersContinue: tests that when one reader cancels, other readers continue waiting These tests properly exercise the done channel wait/cancel logic without requiring HTTP calls - the blocking lookup simulates a slow download.	2025-12-04 23:40:56 -08:00
chrislu	8d110b29dd	fmt	2025-12-04 10:40:01 -08:00
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.	2025-12-03 21:12:19 -08:00
Chris Lu	61c0514a1c	filer: add username and keyPrefix support for Redis stores (#7591) ... * filer: add username and keyPrefix support for Redis stores Addresses https://github.com/seaweedfs/seaweedfs/issues/7299 - Add username config option to redis2, redis_cluster2, redis_lua, and redis_lua_cluster stores (sentinel stores already had it) - Add keyPrefix config option to all Redis stores to prefix all keys, useful for Envoy Redis Proxy or multi-tenant Redis setups * refactor: reduce duplication in redis.NewClient creation Address code review feedback by defining redis.Options once and conditionally setting TLSConfig instead of duplicating the entire NewClient call. * filer.toml: add username and keyPrefix to redis2.tmp example	2025-12-01 13:31:35 -08:00
Chris Lu	d48e1e1659	mount: improve read throughput with parallel chunk fetching (#7569) ... * mount: improve read throughput with parallel chunk fetching This addresses issue #7504 where a single weed mount FUSE instance does not fully utilize node network bandwidth when reading large files. Changes: - Add -concurrentReaders mount option (default: 16) to control the maximum number of parallel chunk fetches during read operations - Implement parallel section reading in ChunkGroup.ReadDataAt() using errgroup for better throughput when reading across multiple sections - Enhance ReaderCache with MaybeCacheMany() to prefetch multiple chunks ahead in parallel during sequential reads (now prefetches 4 chunks) - Increase ReaderCache limit dynamically based on concurrentReaders to support higher read parallelism The bottleneck was that chunks were being read sequentially even when they reside on different volume servers. By introducing parallel chunk fetching, a single mount instance can now better saturate available network bandwidth. Fixes: #7504 * fmt * Address review comments: make prefetch configurable, improve error handling Changes: 1. Add DefaultPrefetchCount constant (4) to reader_at.go 2. Add GetPrefetchCount() method to ChunkGroup that derives prefetch count from concurrentReaders (1/4 ratio, min 1, max 8) 3. Pass prefetch count through NewChunkReaderAtFromClient 4. Fix error handling in readDataAtParallel to prioritize errgroup error 5. Update all callers to use DefaultPrefetchCount constant For mount operations, prefetch scales with -concurrentReaders: - concurrentReaders=16 (default) -> prefetch=4 - concurrentReaders=32 -> prefetch=8 (capped) - concurrentReaders=4 -> prefetch=1 For non-mount paths (WebDAV, query engine, MQ), uses DefaultPrefetchCount. * fmt * Refactor: use variadic parameter instead of new function name Use NewChunkGroup with optional concurrentReaders parameter instead of creating a separate NewChunkGroupWithConcurrency function. This maintains backward compatibility - existing callers without the parameter get the default of 16 concurrent readers. * Use explicit concurrentReaders parameter instead of variadic * Refactor: use MaybeCache with count parameter instead of new MaybeCacheMany function * Address nitpick review comments - Add upper bound (128) on concurrentReaders to prevent excessive goroutine fan-out - Cap readerCacheLimit at 256 accordingly - Fix SetChunks: use Lock() instead of RLock() since we are writing to group.sections	2025-11-29 10:06:11 -08:00
Chris Lu	4106fc0436	fix(tikv): improve context propagation and refactor batch delete logic (#7558) ... * fix(tikv): improve context propagation and refactor batch delete logic Address review comments from PR #7557: 1. Replace context.TODO() with ctx in txn.Get calls - Fixes timeout/cancellation propagation in FindEntry - Fixes timeout/cancellation propagation in KvGet 2. Refactor DeleteFolderChildren to use flush helper - Eliminates code duplication - Cleaner and more maintainable These changes ensure proper context propagation throughout all TiKV operations and improve code maintainability. * error formatting	2025-11-26 15:54:26 -08:00
Chris Lu	5287d9f3e3	fix(tikv): replace DeleteRange with transaction-based batch deletes (#7557) ... * fix(tikv): replace DeleteRange with transaction-based batch deletes Fixes #7187 Problem: TiKV's DeleteRange API is a RawKV operation that bypasses transaction isolation. When SeaweedFS filer uses TiKV with txn client and another service uses RawKV client on the same cluster, DeleteFolderChildren can accidentally delete KV pairs from the RawKV client because DeleteRange operates at the raw key level without respecting transaction boundaries. Reproduction: 1. SeaweedFS filer using TiKV txn client for metadata 2. Another service using rawkv client on same TiKV cluster 3. Filer performs batch file deletion via DeleteFolderChildren 4. Result: ~50% of rawkv client's KV pairs get deleted Solution: Replace client.DeleteRange() (RawKV API) with transactional batch deletes using txn.Delete() within transactions. This ensures: - Transaction isolation - operations respect TiKV's MVCC boundaries - Keyspace separation - txn client and RawKV client stay isolated - Proper key handling - keys are copied to avoid iterator reuse issues - Batch processing - deletes batched (10K default) to manage memory Changes: 1. Core data structure: - Removed deleteRangeConcurrency field - Added batchCommitSize field (configurable, default 10000) 2. DeleteFolderChildren rewrite: - Replaced DeleteRange with iterative batch deletes - Added proper transaction lifecycle management - Implemented key copying to avoid iterator buffer reuse - Added batching to prevent memory exhaustion 3. New deleteBatch helper: - Handles transaction creation and lifecycle - Batches deletes within single transaction - Properly commits/rolls back based on context 4. Context propagation: - Updated RunInTxn to accept context parameter - All RunInTxn call sites now pass context - Enables proper timeout/cancellation handling 5. Configuration: - Removed deleterange_concurrency setting - Added batchdelete_count setting (default 10000) All critical review comments from PR #7188 have been addressed: - Proper key copying with append([]byte(nil), key...) - Conditional transaction rollback based on inContext flag - Context propagation for commits - Proper transaction lifecycle management - Configurable batch size Co-authored-by: giftz <giftz@users.noreply.github.com> * fix: remove extra closing brace causing syntax error in tikv_store.go --------- Co-authored-by: giftz <giftz@users.noreply.github.com>	2025-11-26 14:45:56 -08:00