mirror of
https://github.com/GreptimeTeam/greptimedb.git
synced 2026-10-25 13:22:30 +00:00
bb3f07b4825e074c207ae29db28de1e3354c11ae
6442
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Lei, HUANG
|
bb3f07b482 |
fix(mito2): pick TWCS L1 merges by closed units so compaction terminates (#9741)
* refactor(mito2): pick TWCS L1 merges by closed units TWCS could keep rewriting L1 files without new writes. Overlap-only picks could split their inputs into one more file than they read, and L1 file reduction then rewrote up to 16 files to remove that single file. The outputs, split by key, created new overlaps and the cycle repeated. Replace the three global pick phases with one pick per window, scheduled by class across windows. Every pick now decreases a termination measure when the window receives no writes. - Active windows only take merges that remove files. L1 files are grouped by conflicts and merged in key order, so size-split tails are merged together. A reduction rewrites at most 4 output files' worth of bytes per removed file. - Inactive windows absorb L0 files, then take closed merges: no other file intrudes on the inputs' bounding box, so outputs cannot add overlaps. Each merge resolves conflicts or removes files, decreasing (conflicts, files). Files that do not take part in an overlap are not rewritten. - Overlaps that only share a boundary key no longer count as conflicts, so a size-split sorted run does not look overlapping. - Closed merges may take 4x the input limit to resolve a whole window once. - Cache primary key ranges per pick instead of resolving them per comparison. - Log the pick branch of each compaction output. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * docs(mito2): state the compactor assumptions behind TWCS termination Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * feat(mito2): resolve active window overlaps on manual compaction A window that stops receiving writes keeps being the active window, whose picks only remove files, so its overlaps were never resolved. Manual compactions now pick every window as inactive, so a closed merge can resolve them. Automatic compactions are unchanged. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): defer overlap resolution in the window before the active one Late data keeps landing in the window right before the active window. A closed merge that resolves its overlaps early rewrites the whole window, then the late data brings the overlaps back, and the balanced run check keeps them there. Overlap-only merges now skip that window until it is older. File-reducing merges and manual compactions are unchanged. On the 12h/1GB out-of-order bench this reaches overlap depth 1.0 instead of 2.08 at the same write amplification (3.01 vs 3.02). Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * refactor(mito2): only check earlier-starting files when closing units Closing units only checks files that start before a unit ends, and the overlap checks compare key bounds without cloning them. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): keep settling windows from rewriting L1 files The window before the active one still receives late data. Skipping only its overlap-only merges was not enough: file-reducing merges of conflicting files and mixed L0/L1 merges still rewrote its L1 files, and a mixed merge could rewrite a nearly full output for each small late file. A settling window now only merges L0 files among themselves and conflict-free files; the window resolves its overlaps once it is older. On the out-of-order bench, write amplification drops from 3.88 to 3.30 at 2GB/6h, 3.05 to 2.94 at 1GB/6h and 3.01 to 2.96 at 1GB/12h, with overlap depth 1.0 in all three. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): resolve late overlaps once a window is quiet Late data that lands in a window older than the settling one is merged into a small file covering the whole key range. Resolving it rewrites the window's dominant sorted run, which the balanced run check refuses, so the overlap stayed. Removing the check instead rewrites the window for every late flush. A window is quiet once the region has written at least as many rows as the window holds since its newest file. Overlap merges in a quiet window may rewrite a dominant run, so each quiet period pays for at most one such rewrite. Sequences are per row and kept by compaction, so this needs no state. On the bench with late data up to 2 hours, mean overlap depth drops from 1.67 to 1.17 at 2GB/6h and from 2.33 to 1.83 at 1GB/6h, at write amplification 3.70 to 4.25 and 3.26 to 3.74; all inactive windows of 2GB/6h realtime reach depth 1 (write amplification 2.80 to 3.30). The remaining depth is in the final active window. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * test(mito2): check closed merges of quiet and settling windows terminate Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): split TWCS closures over the input limit A file spanning the keys of a large sorted run joins every run file into one closure. Once a closure held more files than one compaction accepts, closed merges skipped it, so the window never merged those files again and late files kept piling up. Split such closures into single files, like active-window picks do, and replace the closed check with a conflict bound: outputs only conflict with files intruding on the inputs' box, so a merge is taken if `outputs · I` does not exceed the conflicts its inputs had with other files. Intervals of whole closures have no such conflicts and still need to be closed, so other windows pick as before, and `(conflicts, files)` still strictly decreases. Small run files and late files over a large run can be merged again. A late file over a run that cannot shrink stays: resolving it would rewrite the whole run. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): address TWCS picker review findings - Manual compactions may rewrite a dominant sorted run to resolve overlaps. The active window is never quiet, so manual compactions could not resolve its late-data overlaps before. - Single-key intervals predict one output: the writer never splits a series, so large files of one series were never merged. - The active window falls back to single files when no group of conflicting files qualifies, so a large file no longer keeps its small peers unmerged. - Compute the settling window distance with checked_sub, which overflowed for windows at the timestamp limits. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): count inner keys as TWCS conflicts and check balance per group - A single key strictly inside another file's key range now conflicts, so a late file of one series inside a wide file is resolved like other overlaps. Files sharing only a boundary key still do not conflict. Intrusion covers the new pairs, which a randomized test checks. - Check the dominant-run guard per group of conflicting files. Runs of separate groups never overlap, so pairing one group's run with another group's peer made an interval of two dominant runs look balanced. - A window with unknown row counts, written before they were recorded, is not quiet: its threshold would drop to a single row. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * perf(mito2): index TWCS files by key to avoid quadratic grouping Grouping a window's files compared every file with all files sharing its time range, and closing groups and checking intrusions scanned the whole window. A sorted run split by key shares one time range, so a window of N such files took O(N²) checks without any conflict. The 1024-file fallback only covers the sorted-run path, not these groups. Index the files by lower primary key in a segment tree that keeps each subtree's largest upper key and its time bounds. Grouping, closing, and the conflict bound only check the files whose key and time ranges meet, so a file of a key-sliced run only meets its neighbors. Conflicts among split files are precomputed per file and counted per interval with a binary search instead of pairwise. On a run of 20,000 key-sliced files with 5 late files spanning all keys, a closed merge pick takes ~62 ms instead of ~3.9 s and an active pick ~19 ms instead of ~3.5 s. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * test(object-store): retry transient S3 transport failures Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): bound TWCS closed-merge conflict graph work Budget the candidate files visited while building conflict units and split-unit adjacency. When the budget is exceeded, treat every file as a split unit and pick a bounded file reduction instead of materializing a quadratic graph. The fallback candidate is admitted only through adds_no_conflicts, so outputs do not add conflict pairs and the (P, N) termination invariant still holds. Dense picks are logged with their own branch. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): charge TWCS index nodes to the conflict-graph budget The conflict-graph work budget only counted files that reached a callback, so a traversal with poor pruning could still visit a quadratic number of internal nodes without triggering the dense fallback. Charge every ExtentIndex node, including pruned nodes, and stop the traversal as soon as the shared budget is exhausted. Add a regression test where the root straddles the query time but neither leaf matches, so a callback-only budget would never fire. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): retry smaller TWCS dense compaction candidates If a dense fallback candidate fails admission, try one single-output subinterval within it before giving up. Reuse the reduction guards and check both candidates against the full window's external conflicts, preserving bounded work and termination.\n\nCover the chain regression through both fallback unit constructions and reject one-output retries that introduce conflicts across time gaps. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): compact shared primary-key boundaries Count shared primary-key endpoints as conflicts and apply the same overlap rule to bounding-box intrusions. Existing compaction writers keep series intact, so fresh output ranges remain disjoint. Align simulated output splits with the writer and cover boundary-only cleanup in inactive and manual windows, plus stable real output files after a subsequent TWCS request. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * fix(mito2): compact small L1 files in inactive windows Allow bounded small-L1 reductions in settling and older inactive windows when normal picks cannot proceed. Preserve full-window conflict checks, large-file protection, and LastNonNull closure semantics.\n\nCover selection limits, termination, and real SST compaction with tombstones and reopen checks. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * docs(mito2): clarify TWCS unit grouping guarantees Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> --------- Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> |
||
 fys
|
f35057d9c2 |
feat(json2): support inverted index (#9324)
feat: build inverted indexes for JSON2 type hints Signed-off-by: fys <fengys1996@gmail.com> |
||
 Weny Xu
|
4aa4aa63fe |
fix(test): avoid MinIO connection reuse after conditional writes (#9877)
Signed-off-by: WenyXu <wenymedia@gmail.com> |
||
 Ning Sun
|
af52dfafd8 |
chore(deps): upgrade sqlx to 0.9 to pick up webpki-roots 1.0.10 (#9862)
sqlx 0.8 pins the unmaintained webpki-roots 0.26 line for its rustls trust store, which still contains the recently compromised e-Tugra CA root. Upgrading sqlx to 0.9 switches its dependency to webpki-roots 1.x, and the lockfile now resolves to 1.0.10 where the distrusted root has been removed from the compiled-in Mozilla trust store. sqlx 0.9 requires dynamic SQL strings to be explicitly audited via the new SqlSafeStr trait: only &'static str literals are accepted as-is. All affected call sites (MySQL KV backend, MySQL election, integration tests and fuzz targets) now wrap their internally generated SQL templates in AssertSqlSafe; the wrapped SQL is built from operator config or test-local identifiers only, and dynamic data keeps flowing through bind parameters. The mysql-rsa feature is enabled alongside mysql to preserve 0.8 authentication behavior: sqlx 0.9 gates non-TLS RSA password auth (caching_sha2_password / sha256_password) behind it. Signed-off-by: Ning Sun <sunning@greptime.com> |
||
 Yingwen
|
8b1d68369b |
fix(ci): align nightly Windows and macOS artifact test stack sizes (#9860)
* fix(ci): increase Windows nightly test stack size Signed-off-by: evenyag <realevenyag@gmail.com> * fix(ci): align optional macOS artifact test stack size Signed-off-by: evenyag <realevenyag@gmail.com> --------- Signed-off-by: evenyag <realevenyag@gmail.com> |
||
|
Weny Xu
|
b4aae2f73b |
fix(meta): release all PostgreSQL election advisory locks on step-down (#9864)
Signed-off-by: WenyXu <wenymedia@gmail.com> |
||
|
Weny Xu
|
f595e9bbf0 |
ci: allow independent Linux release architecture selection (#9865)
* ci: allow independent Linux release architecture selection Signed-off-by: WenyXu <wenymedia@gmail.com> * ci: use Linux architecture checkboxes and preserve release defaults Signed-off-by: WenyXu <wenymedia@gmail.com> --------- Signed-off-by: WenyXu <wenymedia@gmail.com> |
||
|
Weny Xu
|
34210f5022 |
feat(mito): add an environment switch to disable compaction (#9863)
Signed-off-by: WenyXu <wenymedia@gmail.com> |
||
 LFC
|
67bd279586 |
fix(query): preserve Arrow types when decoding dictionary output (#9858)
Signed-off-by: luofucong <luofc@foxmail.com> |
||
|
Weny Xu
|
9e9346d2ff |
fix: preserve datanode write batches when retrying (#9720)
* fix: preserve datanode write batches when retrying Signed-off-by: WenyXu <wenymedia@gmail.com> * fix: isolate invalid entries in metric write batches Signed-off-by: WenyXu <wenymedia@gmail.com> * refactor: adapt batch writes to the unified metric API Signed-off-by: WenyXu <wenymedia@gmail.com> * fix: explicitly accept partial results for batched retries Signed-off-by: WenyXu <wenymedia@gmail.com> * test: assert terminal retry hint instead of masked error text Signed-off-by: WenyXu <wenymedia@gmail.com> * test: preserve application status in region failure fixtures Signed-off-by: WenyXu <wenymedia@gmail.com> * fix: refresh routes after region close Signed-off-by: WenyXu <wenymedia@gmail.com> --------- Signed-off-by: WenyXu <wenymedia@gmail.com> |
||
 dennis zhuang
|
436d07bf2d |
fix(mito): return from manual index builds only after the index is published (#9643)
* fix(mito): return from manual index builds only after the index is published A manual build_index could report success while the index was missing: the request was rejected or coalesced behind another build, the manifest commit failed, the indexer dropped an error, or the region was staging. Manual builds now wait for active builds of the same SST, keep their waiters across coalescing and schema-change retries, are notified after the worker applies the edit, and verify the files against the manifest before returning. The consistency check also honors inverted_index.ignore_column_ids and a zero segment row count, so such SSTs are no longer rebuilt on every call. Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * test(mito): cover manual index build waiting for a schema-change retry Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(mito): select manual index build files by the applied version Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * chore(mito): log why index builds are skipped or aborted Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * refactor(mito): split IndexBuildTask::index_build into phases Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(mito): stop scheduling index builds when the region state changes Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * chore(mito): document region file id and silent aborts in index builds Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(mito): log why a manual index build is incomplete Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(mito): expose incomplete index builds and treat compacted files as done Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * test(mito): pin index build concurrency in the compaction test Signed-off-by: Dennis Zhuang <killme2008@gmail.com> --------- Signed-off-by: Dennis Zhuang <killme2008@gmail.com>v1.4.0-nightly-20261022 |
||
|
Weny Xu
|
27c531336a |
fix(mito2): restore source splitting for cross-range series merges (#9818)
* fix: restore batched series merges across ranges Signed-off-by: WenyXu <wenymedia@gmail.com> * fix(mito2): split range sources before last-row selection Signed-off-by: WenyXu <wenymedia@gmail.com> * fix(mito2): avoid tiny splits for single-source cross-range merges Signed-off-by: WenyXu <wenymedia@gmail.com> --------- Signed-off-by: WenyXu <wenymedia@gmail.com> |
||
|
Ning Sun
|
ec4ef45f40 |
feat: use pgwire auth mechanisms API to simplify startuphandler (#9634)
* feat: use pgwire auth mechanisms API to simplify startuphandler * chore: update pgwire * chore: update pgwire revision to merged one * refactor: model based on upstream api * refactor: simplify implementation by introducing authsource * refactor: update to latest pgwire apis * refactor: adopt pgwire api changes * chore: switch to released pgwire * test: harden pg scram mixed-auth integration test and reject channel-binding clients - Fix MixedAuthProvider to return a SCRAM verifier for the denied user, so the test exercises rejection by authorization after successful SCRAM verification; record the selected mechanism from the returned variant and assert authorize ran exactly once per verified user. - Assert the denied connection fails with 28P01 before any terminal message, protecting the completion-ordering introduced by pgwire's authentication gate. - Add a raw-protocol test proving a client-first with a p= channel binding flag is rejected (RFC 5802 section 6), while n-flag clients (tokio_postgres) continue normally. Bump pgwire to 0.42.1, which enforces RFC 5802 channel binding negotiation and rejects short/overlong SCRAM proofs on the verifier path (sunng87/pgwire#476). Signed-off-by: Ning Sun <sunning@greptime.com> --------- Signed-off-by: Ning Sun <sunning@greptime.com> |
||
|
Weny Xu
|
0dc649e028 |
fix(datanode): release region map guard before awaiting role changes (#9824)
Signed-off-by: WenyXu <wenymedia@gmail.com> |
||
|
LFC
|
5af1e3301e |
fix(mysql): cap timestamp output at microsecond precision (#9831)
* fix(mysql): cap timestamp output at microsecond precision Truncate MySQL text timestamp results to microseconds while preserving the existing second and millisecond formatting. Keep stored nanoseconds and explicit integer/string representations unchanged. Fixes #9828 Signed-off-by: luofucong <luofc@foxmail.com> * chore: ignore local output directory Signed-off-by: luofucong <luofc@foxmail.com> --------- Signed-off-by: luofucong <luofc@foxmail.com> |
||
 Chenglong Zheng
|
9f9b35c270 |
feat: support wait and timeout options for TRUNCATE TABLE (#9753)
* feat: support wait and timeout options for TRUNCATE TABLE * fix: address review feedback for truncate DDL options * fix: address review feedback for truncate DDL options |
||
|
Yingwen
|
e6fa1a0b21 |
fix(ci): increase release integration test stack size (#9817)
Signed-off-by: evenyag <realevenyag@gmail.com> |
||
|
Weny Xu
|
f866ef2bd9 |
feat: return per-entry region write failures (#9728)
* feat: return per-entry region write failures Signed-off-by: WenyXu <wenymedia@gmail.com> * fix: preserve retry hints for single-entry region writes Signed-off-by: WenyXu <wenymedia@gmail.com> * fix: preserve region write error compatibility Signed-off-by: WenyXu <wenymedia@gmail.com> * refactor: simplify region partial result handling Signed-off-by: WenyXu <wenymedia@gmail.com> * fix: align partial failure summaries with input order Signed-off-by: WenyXu <wenymedia@gmail.com> * refactor: separate region write response handling Signed-off-by: WenyXu <wenymedia@gmail.com> * test: fix partial result expectations and region map guard Signed-off-by: WenyXu <wenymedia@gmail.com> * chore: limit lockfile changes to greptime proto Signed-off-by: WenyXu <wenymedia@gmail.com> * fix: make partial region results an explicit client opt-in Signed-off-by: WenyXu <wenymedia@gmail.com> * chore: pin greptime-proto to merged partial-result changes Signed-off-by: WenyXu <wenymedia@gmail.com> --------- Signed-off-by: WenyXu <wenymedia@gmail.com> |
||
|
Lei, HUANG
|
a1b2362e58 |
fix: avoid duplicate keys in update flow metadata txn (#9816)
CREATE OR REPLACE FLOW deleted and then put the same flow route, flownode flow and table flow keys in a single txn when the replacement kept the partitions, flownodes or source tables, which etcd rejects with "duplicate key given in txn request". Only delete the old keys that are not put again. Closes #9810. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> |
||
|
LFC
|
9a2ac7f9de |
feat: support dot access for JSON columns (#9775)
Signed-off-by: luofucong <luofc@foxmail.com> |
||
 Aum
|
6f53d57d15 |
fix: avoid exact scan row counts for SSTs with unknown counts (#9759)
fix: avoid exact scan row counts for legacy SSTs Treat selected SSTs with unknown row counts as non-exact before checking region partition visibility. Preserve exact statistics for known counts and cover legacy metadata mixed with known-count files. Fixes GreptimeTeam/greptimedb#9718 Signed-off-by: Aumnertic <hznuzc@gmail.com> |
||
|
Ning Sun
|
a7230f7a92 |
fix(partition): coerce partition exprs against the batch schema before routing (#9752)
* fix(partition): coerce partition exprs against the batch schema before routing Batch partition routing (MultiDimPartitionRule::split_record_batch) builds physical expressions via PartitionExpr::try_as_physical_expr, which did no type coercion. After widening the time index unit (e.g. ms -> ns via ALTER TABLE ... MODIFY COLUMN ts TIMESTAMP_NS), the persisted partition bounds keep their original unit, so evaluating the bounds against the batch failed with: Invalid comparison operation: Timestamp(ns) < Timestamp(ms) Row inserts were unaffected because find_region compares Values and Timestamp::cmp converts units; scans were unaffected because PredicateGroup::new coerces the region partition expr against the region schema first. Coerce the logical partition expression against the evaluation schema in try_as_physical_expr before creating the physical expression, the same way PredicateGroup does. This also covers the region partition filter built for SST pruning (src/mito2/src/sst/parquet/reader.rs), which shares this conversion. Fixes #9747 Signed-off-by: Ning Sun <sunning@greptime.com> * test: regenerate change_col_type result after comment update The ts_widen_part comment in change_col_type.sql was reworded after the result file was last generated; sqlness echoes SQL comments into the result, so the stale comment made the case fail. Regenerate it. Signed-off-by: Ning Sun <sunning@greptime.com> --------- Signed-off-by: Ning Sun <sunning@greptime.com> |
||
|
dennis zhuang
|
b9aa627180 |
fix(query): forward scalar subquery filters to region scans (#9738)
* fix(query): forward scalar subquery filters to region scans Since DataFusion 55, uncorrelated scalar subqueries run in ScalarSubqueryExec instead of being rewritten into joins, so filters such as ts = (SELECT ts ... ORDER BY ts DESC LIMIT 1) no longer reach region scans through a join dynamic filter. MergeScanExec now captures such filters as remote dynamic filters, keeps them in the parent FilterExec, and replaces the subquery with its result before querying regions. Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(query): check the remote dyn filter budget across pushdown phases Filters captured in the Pre and Post phases are sent in one request, but each phase validated the initial registration budget alone. When their sum exceeded the budget, serialization dropped every registration, including join and TopK filters. Validate the combined set and keep filters accepted by the other phase. Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(query): keep IN-bearing scalar subquery filters at the parent Oversized remote dyn filter payloads replace IN lists with true. That only widens a conjunction of bounds, but inside an arbitrary predicate such as (host IN (...)) = (SELECT flag ...) it can drop rows the parent filter would keep. Do not forward scalar subquery filters that contain IN lists. Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * chore(query): note the IN fallback restriction for scalar subquery filters Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * chore(query): document the scalar subquery dyn filter bridge removal condition Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * test(query): keep only dyn_filters of region scans in the scalar subquery plan case #9565 added scan diagnostics to EXPLAIN ANALYZE VERBOSE and partition filter rendering changed; the case only checks the forwarded dyn filter. Signed-off-by: Dennis Zhuang <killme2008@gmail.com> --------- Signed-off-by: Dennis Zhuang <killme2008@gmail.com> |
||
|
dennis zhuang
|
977e76d0a4 |
fix(mito2): apply SST internal column encodings to bulk-encoded parts (#9735)
* fix(mito2): apply SST internal column encodings to bulk-encoded parts Bulk memtable parts are encoded to parquet and flushed as SSTs as-is, but the bulk encoder did not apply the time index / __sequence DELTA_BINARY_PACKED encoding (with dictionary disabled) and the uncompressed op_type setting that the regular SST writer uses. Share these settings through one helper. Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * chore(mito2): drop the answered bulk encoder compression TODO Signed-off-by: Dennis Zhuang <killme2008@gmail.com> --------- Signed-off-by: Dennis Zhuang <killme2008@gmail.com> |
||
 shuiyisong
|
41ec5e0763 |
fix: decouple slow-query logging from the global log level (#9765)
Signed-off-by: shuiyisong <xixing.sys@gmail.com> |
||
|
dennis zhuang
|
c51293836e |
fix(query): push LIKE filters down into MergeScan remote plans (#9648)
Signed-off-by: Dennis Zhuang <killme2008@gmail.com> |
||
|
Yingwen
|
3e2751d71d |
feat: improve SQL analyze scan diagnostics (#9565)
* feat: improve SQL analyze scan diagnostics Signed-off-by: evenyag <realevenyag@gmail.com> * test: redact dynamic filters in sqlness results Signed-off-by: evenyag <realevenyag@gmail.com> * test: stabilize analyze snapshots for merge queue Signed-off-by: evenyag <realevenyag@gmail.com> * test: remove obsolete scan_details assertion Signed-off-by: evenyag <realevenyag@gmail.com> * test: refresh explain snapshots after partition filter changes Signed-off-by: evenyag <realevenyag@gmail.com> --------- Signed-off-by: evenyag <realevenyag@gmail.com> |
||
|
dennis zhuang
|
4645ab9ed7 |
perf(mito2): gather last rows with one take per input batch (#9737)
* perf(mito2): gather last rows with one take per input batch FlatLastTimestampSelector buffered a single-row slice per series and concatenated up to 8192 of them. Concatenating slices with dictionary columns merges the dictionaries once per slice, which made high-cardinality instant PromQL queries using the LastRow scan path (#9034) slower. Select the rows of series that end inside a batch with one take instead. Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * test(mito2): cover duplicate last timestamps in the take path Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(mito2): group last rows by primary key bytes Flat batches may assign different dictionary keys to the same primary key, for example when binary keys are wrapped with identity keys. Merge adjacent ranges with equal key bytes before selecting last rows. Signed-off-by: Dennis Zhuang <killme2008@gmail.com> --------- Signed-off-by: Dennis Zhuang <killme2008@gmail.com> |
||
|
shuiyisong
|
9c7e1e1605 |
fix: record slow INSERT SELECT queries (#9733)
Signed-off-by: shuiyisong <xixing.sys@gmail.com> |
||
 jeremyhi
|
a436213bf3 |
ci: compare several images in one object store WAL S3 benchmark run (#9691)
* ci: compare several images in one object store WAL S3 benchmark run
A new `images` input takes a JSON list of 2..4 {label, image}. Every
cell then runs once per image, back to back, and the image order
reverses with the configuration order every other round. The cell id
gains the label. The single-image inputs keep their meaning and cell ids.
The validate job pulls every image before an instance is allocated. It
pins each image to its digest and records the version and commit that
the image reports. Each cell starts the image of its label. An optional
`warmup` runs the first load cell once per image as round 0, and no
statistic uses it.
The plan step estimates the benchmark job's minutes on c6id.2xlarge
from the cell times of earlier runs. It refuses a plan that does not
fit the job timeout.
The summary keeps the Raft Engine ratios within each image. With
several images it also reports, per group, concurrency and config:
- every round's rows/s and client p50/p99;
- each image's ratio to the first, with the geometric mean, minimum
and maximum over the rounds.
Every cell records the host load averages at its start and end.
Signed-off-by: jeremyhi <fengjiachun@gmail.com>
* ci: scale the S3 benchmark time estimate with the batch size
The per-cell rates were measured with 1000-row batches. A smaller batch
keeps about the request rate, not the row rate, so the estimate now
shrinks rows/s with the batch. A one-row durable plan is refused instead
of passing a timeout it cannot fit.
Signed-off-by: jeremyhi <fengjiachun@gmail.com>
* ci: pin the S3 benchmark runtime image without pulling it in validate
An anonymous pull from ECR Public on a hosted runner can exceed the
registry's data limit, and it failed the validate job of several
dispatches with "toomanyrequests: Data limit exceeded". Validate now
reads only the registry manifest to pin the runtime image's digest. A
reference that names a digest must be served under that digest.
The check that the image holds the loader and the generator moves to
the benchmark host. It runs right after the runtime pull and before any
GreptimeDB image is pulled or any cell starts.
Signed-off-by: jeremyhi <fengjiachun@gmail.com>
* test: run the S3 benchmark runtime content check against both files
The fake docker now runs the pull step's own check in a directory that stands in for the runtime image. It covers both files present, the loader missing and the generator missing, so dropping either file check fails the test.
Signed-off-by: jeremyhi <fengjiachun@gmail.com>
* ci: pin a digest-named S3 benchmark runtime image without the registry
ECR Public's anonymous data limit now refuses hosted runners even the
manifest read, and validate failed in "Pin the runtime image digest".
A runtime reference that names its sha256 digest, like the default one,
is already pinned, so validate takes that digest with no network call.
The format checks stay. Only a tag-only reference is resolved through
the registry. The content check on the benchmark host is unchanged.
Signed-off-by: jeremyhi <fengjiachun@gmail.com>
* ci: check the full S3 benchmark runtime reference before pinning it offline
A named digest no longer reaches the registry, so nothing parsed the rest of the reference. Validate now checks the whole reference against Docker's reference grammar first. It refuses an empty repository, a doubled '@' or a bad tag before any pin is emitted or any instance is allocated.
Signed-off-by: jeremyhi <fengjiachun@gmail.com>
---------
Signed-off-by: jeremyhi <fengjiachun@gmail.com>
|
||
|
jeremyhi
|
93c3b116da |
fix(mito2): re-check a filled write as sent after a schema change (#9701)
* fix(mito2): re-check a filled write as sent after a schema change The engine fills the columns a write leaves out with their defaults when it accepts the write, appending them to the rows without adding them to `name_to_index` and `has_null`. If the region schema changes before the worker handles the write, the worker checks the request again and `check_schema` looks up a filled column in `name_to_index`, which panics and stops the region worker. If the filled column was dropped, the write fails with an unknown column that the client never sent. Remove the filled columns before checking, so the request is checked as sent and filled again for the current schema. Nothing is filled yet when the engine checks the request, so that check does no extra work. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * test(mito2): rename put_request helper to build_put_request Signed-off-by: jeremyhi <fengjiachun@gmail.com> --------- Signed-off-by: jeremyhi <fengjiachun@gmail.com> |
||
|
dennis zhuang
|
da88ff8653 |
fix(partition): route NULLs like row inserts and accept signed-minimum bounds (#9697)
* fix(partition): route NULLs like row inserts and accept signed-minimum bounds Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(partition): keep equality two-valued for NULL rows and accept unsigned -0 Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(partition): treat NULL masks as false in batch routing and keep comparisons bare Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * test(compat): update partition filter in analyze verbose expected output Signed-off-by: Dennis Zhuang <killme2008@gmail.com> --------- Signed-off-by: Dennis Zhuang <killme2008@gmail.com> |
||
|
jeremyhi
|
7397ab6b23 |
perf(datanode): send the mito puts of one request to each worker at once (#9692)
* perf(datanode): send the mito puts of one request to each worker at once The region server handled the puts of a request to several regions as separate requests. A Mito worker woke on the first of them and wrote it alone; in the durable object store WAL mode its append then blocked the worker until the next seal, so the rest of the request went into the next object, and clients writing a partitioned table locked into one phase that wastes a tick. When every put of a request goes to a ready mito region, the region server now hands them to `MitoEngine::put_regions_batch`, which submits the puts of each worker in one message, so the worker writes its regions of the request in one batch and one WAL append. Requests to one region and to other engines are handled as before. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * perf(datanode): await each region of a batched mito put on its own task The batched path waited for the regions of a request in order and sent the worker messages one after another, so an error or a full worker queue of one region held back the others, and its metrics counted the request as a whole. `MitoEngine::put_regions_batch` now sends the worker messages concurrently on their own task and returns each region's result. The region server awaits and accounts every region on its own ingest task, as the per-region path does: the first error returns at once, and each region records its latency, changed rows or failure and is named in its error. A request with several puts to one region keeps the per-region path. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * perf(datanode): spawn the accounting of batched mito puts from the batch task The request spawned the per-region accounting tasks only after the batch task returned, so a request dropped in between left accepted writes unaccounted. The batch task now spawns them and the request only joins their handles. Also state in the architecture invariants that the region server's batched puts to the metric and mito engines are the one request dispatch that does not go through the RegionEngine trait. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * perf(datanode): batch mito puts through the RegionEngine trait Add `RegionEngine::handle_batch_put_requests`, which returns each region's response of the puts of one request. The default handles each put with `handle_request`, so other engines keep their behaviour; Mito overrides it with the one message per worker submission. The region server calls the trait method instead of downcasting to `MitoEngine`, and the architecture invariants no longer need an exception. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * perf(datanode): trace each region of a batched mito put Attach the per-region `RegionServer::handle_region_request` span of the per-region path to each region's task of a batched mito put, so the region ids stay in distributed traces. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * perf(datanode): let a batch put response answer for several regions `RegionEngine::handle_batch_put_requests` now returns each response with the regions it answers for. The default and Mito answer each region on its own; an engine that writes several puts together, such as the metric engine's batched puts, can answer them with one response. The region server accounts a response for each of its regions. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * perf(datanode): batch any request with several mito puts Take the batch path for a request with at least two puts to ready mito regions, also when they all target one region: a request of many puts to one region otherwise still sends one worker message per put. Each put keeps its own response, affected rows and error. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * perf(datanode): account batched mito puts in one task Await the responses of a batched mito put in the batch task with a FuturesUnordered instead of one spawned task per put. The task accounts each response as it completes, hands the first error back to the caller at once and keeps accounting the rest, so a request of many puts no longer spawns a task per put. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * refactor(mito2): name the batched worker message BatchWrite Signed-off-by: jeremyhi <jiachun_feng@proton.me> * refactor(mito2): submit batched writes on the ingest runtime Grouping and submitting the writes of a request is part of ingestion, so the task runs on the ingest runtime and takes the write task class when the workload scheduler is enabled. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * refactor(datanode): share ingest result accounting across put paths The per-region path, the metric batch path and the mito batch path each recorded the changed rows or the failures of an ingest response. They now call one helper, so a metric or label change happens in one place. A batch put response lists the region of each put it answers for, so the failure count stays per put when two puts go to one region; the trait documentation now says so. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * fix(mito2): give batched puts of a stopped worker the worker's error A batched write message that could not reach a stopped worker was dropped, so its puts failed with "Failed to recv result" while a put submitted on its own fails with "Worker <id> is stopped". The worker now returns the message it could not take, and each of its puts receives the WorkerStopped error. Signed-off-by: jeremyhi <jiachun_feng@proton.me> * refactor(datanode): look up the batch latency histogram once The label of the request latency histogram is the same for every put of a batch, so the batch task looks it up once instead of per observation. Signed-off-by: jeremyhi <jiachun_feng@proton.me> --------- Signed-off-by: jeremyhi <jiachun_feng@proton.me> |
||
 fengyue-xve
|
1e29e8d4b6 |
fix(meta): validate persisted partition bounds before altering regions (#9487)
* fix(meta): validate persisted partition bounds before altering regions
A partition bound is persisted with the type of the column at the time
it was written. When an ALTER TABLE changes the type of a column
referenced by a partition expression (widening the time index unit is
the common case), each region rewrites the bound literal to the new
type while building its scan predicate. A millisecond bound like year
3000 fits milliseconds but overflows nanoseconds, so every scan of the
region fails once the table schema is updated -- and some regions may
already be altered when the failure surfaces.
Validate every region's persisted expression on prepare, before any
region is modified: parse the persisted JSON (mirrored types, since
common-meta cannot depend on the partition crate which depends on it),
replay the same value -> ScalarValue conversion and checked cast the
region performs, and reject the ALTER naming the offending region when
a bound cannot be represented with the new column type. Incompatible
changes are rejected atomically; compatible widening proceeds unchanged.
Adds meta-level tests (multi-region rejection naming the second region,
compatible widening completing) and an end-to-end sqlness case.
Fixes #9454
Signed-off-by: fengyue-xve <1759519137@qq.com>
* fix: access ColumnSchema fields directly and use plain format args
ColumnSchema exposes name and data_type as public fields, not methods,
and the telemetry macros only accept a format string after the error.
Signed-off-by: fengyue-xve <1759519137@qq.com>
* test: access ColumnSchema data_type field in alter table tests
Same fix as the previous commit: data_type is a public field on
ColumnSchema, not a method.
Signed-off-by: fengyue-xve <1759519137@qq.com>
* test: drop borrows now that data_type is read as a field
Signed-off-by: fengyue-xve <1759519137@qq.com>
* test: drop useless format! in nested expression test
Signed-off-by: fengyue-xve <1759519137@qq.com>
* test: cover the partition tail in change_col_type sqlness case
Signed-off-by: fengyue-xve <1759519137@qq.com>
* test: drop spurious blank line in change_col_type result
Signed-off-by: fengyue-xve <1759519137@qq.com>
* fix(meta): compare partition bounds with the scan comparison coercion
The preflight validator cast the persisted bound directly to the altered
column type, which rejected ALTERs the scan itself accepts: the region
coerces both the column and the bound to the common comparison type
(ExprSimplifier::coerce in PredicateGroup::new), so narrowing a field
column with an out-of-range bound still scans. Compare through
comparison_coercion(target_type, bound_type) instead and only require the
bound to be castable to the common type.
Also add a round-trip test in the partition crate that serializes real
PartitionExprs and runs them through the validator, so drift between the
Persisted* mirror and the partition types fails CI, and fix the error
message to match the comparison semantics.
Signed-off-by: fengyue-xve <1759519137@qq.com>
* test: update the sqlness fixture for the new incompatibility message
The error message now reflects the comparison semantics: the persisted
bound can no longer be compared with the new column type.
Signed-off-by: fengyue-xve <1759519137@qq.com>
* build(deps): refresh Cargo.lock for the common-meta datafusion-expr dependency
common-meta gained a datafusion-expr dependency in
|
||
|
dennis zhuang
|
e0032534d2 |
fix(mito2): build indexes for encoded bulk parts flushed in sync mode (#9722)
* fix(mito2): build indexes for encoded bulk parts flushed in sync mode Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(mito2): skip reading the SST when an index build has nothing to create Signed-off-by: Dennis Zhuang <killme2008@gmail.com> --------- Signed-off-by: Dennis Zhuang <killme2008@gmail.com> |
||
|
jeremyhi
|
660c4efe1f |
fix(mito2): fence writes before a graceful downgrade (#9716)
* test(mito2): cover graceful role changes racing with writes A graceful role change flips the region state in a background task while the worker keeps handling writes. Add bulk and row write tests that check a write handled after the role change request is either rejected or covered by the returned last_entry_id. Add tests for DowngradingLeader while an ALTER keeps the region in Altering, alone and in the downgrade sequence the datanode runs (DowngradingLeader, Flush, Follower) with a migration candidate on its own local WAL. Add a test that the downgrade sequence covers a write in flight when a lease renewal sets Downgrading. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * fix(mito2): fence writes before a graceful downgrade set_region_role_state_gracefully hands the role change to a background task that waits for the manifest lock. The worker keeps handling writes until the task flips the state, so a write could be acknowledged with an entry id above the returned last_entry_id. Set Downgrading in the worker before spawning the task for Follower and DowngradingLeader requests, as lease renewal already does, so every write the worker handles afterwards is rejected. DowngradingLeader also did nothing and returned success when the region was in another leader state such as Altering. The region became writable again when the operation finished, and the datanode then flushed and turned it into a follower in one step. Return a retryable error instead, so metasrv retries the downgrade. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * test(mito2): assert the retryable downgrade error during ALTER Check that DowngradingLeader during an ALTER returns a retryable error, and release the paused ALTER on that path. The local WAL test now retries the downgrade after the ALTER, as metasrv does. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * test(mito2): write before the retried downgrade in the local WAL test Rows written before an ALTER were already flushed, so the candidate check did not depend on the downgrade's final flush. Write rows after the ALTER and before the downgrade is retried, which only that flush can carry to the candidate. Signed-off-by: jeremyhi <fengjiachun@gmail.com> --------- Signed-off-by: jeremyhi <fengjiachun@gmail.com> |
||
|
Yingwen
|
b3dc5674e9 |
feat(mito2)!: account SST decoding against shared scan memory limits (#9101)
* feat(mito2): add shared estimated SST scan memory budget Signed-off-by: evenyag <realevenyag@gmail.com> * feat: unify scan memory limits with configurable accounting modes Signed-off-by: evenyag <realevenyag@gmail.com> * refactor(mito2): reuse query runtime memory pool for series merges Signed-off-by: evenyag <realevenyag@gmail.com> * test(mito2): wait for cancelled scan reservations before exhaustion check Signed-off-by: evenyag <realevenyag@gmail.com> * docs(mito2): explain parallel scan cancellation cleanup Signed-off-by: evenyag <realevenyag@gmail.com> * fix(mito2): deprecate waiting for scan memory quota Signed-off-by: evenyag <realevenyag@gmail.com> * refactor(recordbatch): remove scan memory wait handling Signed-off-by: evenyag <realevenyag@gmail.com> * fix(table): initialize memory pool in scan test fixtures Signed-off-by: evenyag <realevenyag@gmail.com> --------- Signed-off-by: evenyag <realevenyag@gmail.com> |
||
|
Lei, HUANG
|
52554db52f |
perf(mito2): keep short series unsplit in compaction merges (#9706)
* perf(mito2): keep short series unsplit in compaction merges Compaction writes every merged batch to the output SST. Splitting short series before the merge made it emit one small batch per series and input file, which the SST writer and indexer then handled one by one. Compaction now splits only when series are long enough to fill most of a batch. With 16 L0 files of 10k series x 30 rows, one output went from 160,560 merged batches to 592, and merging plus writing took 1.66s instead of 2.67s. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> * test(mito2): benchmark compacting L0 files into one output The benchmark writes 16 L0 files through the engine, each holding every series over its own time span as consecutive flushes do, and times a strict window compaction that merges them into one output. Every iteration compacts a fresh copy of the same files. With 10k series of 30 rows per file, keeping short series unsplit in compaction merges cut the time from 830ms to 520ms. With 60 series of 5,000 rows per file, which compaction still splits, both took 266ms. Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> --------- Signed-off-by: Lei, HUANG <ratuthomm@gmail.com> |
||
|
jeremyhi
|
791f429246 |
fix(mito2): validate bulk insert batches against the region schema (#9698)
* fix(mito2): validate bulk insert batches against the region schema Bulk inserts were only checked for the time index column, unlike row writes, which WriteRequest::check_schema checks against the region schema. A batch with columns of another type, nulls in the time index or in non-nullable columns, duplicate or unknown columns reached the region worker: a time index that is not a timestamp or holds only nulls made the worker fail, and other mismatches failed after the WAL append or were stored as written, depending on the SST format. Move the column checks of check_schema into a core shared by row writes and bulk inserts, and check bulk insert batches at the same points as rows: when the engine builds the worker request, and again in the worker if the region schema changed since. check_schema keeps its behavior. Bulk insert columns must have the column's Arrow type, except that tags may use any string encoding, as the memtables already convert them. Sparse batches must carry one encoded primary key column of binary without nulls. A batch that misses columns gets them filled by the worker even if it claims to match the current schema. The metric engine checks logical bulk inserts against the logical region before rewriting them into the physical region, as the rewrite copies columns without checking them and would encode a duplicated tag into the primary key twice. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * fix(mito2): report an error when a write is interrupted by a panic WriteNotify answers its waiter when it is dropped and reported success unless an error had been set, also when it was dropped while unwinding from a panic in the middle of the write. Report an error in that case. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * fix(mito2): tolerate a poisoned series lock when dropping a series map SeriesMap::drop reads every series to update the memtable metrics and unwrapped the lock, so dropping the map while unwinding from a panic that poisoned a series lock panicked again. Read through the poisoned lock instead. Signed-off-by: jeremyhi <fengjiachun@gmail.com> --------- Signed-off-by: jeremyhi <fengjiachun@gmail.com> |
||
 yihong
|
4684b95f72 |
ci: do not run these actions on fork (#9721)
Signed-off-by: yihong0618 <zouzou0208@gmail.com> |
||
|
fys
|
4e8bc6df86 |
fix(prom): align remote read label matcher semantics (#9633)
* fix(prom): align remote read label matcher semantics Signed-off-by: fys <fengys1996@gmail.com> * chore: optimize comment of validate_matchers * fix(prom): preserve existing binary label regex handling Signed-off-by: fys <fengys1996@gmail.com> * refactor: condition * test(prom): simplify remote read matcher tests and clarify comments * fix(prom): return binary remote read labels as UTF-8 * fix(prom): reject binary labels in remote read * refactor(prom): align remote read null label handling with PromQL Signed-off-by: fys <fengys1996@gmail.com> * refactor(prom): simplify null handling in label matcher branches Signed-off-by: fys <fengys1996@gmail.com> * fix(prom): restrict remote read labels to string types Signed-off-by: fys <fengys1996@gmail.com> * test(prom): use string label in default value column test Signed-off-by: fys <fengys1996@gmail.com> * fix: code review --------- Signed-off-by: fys <fengys1996@gmail.com> |
||
|
jeremyhi
|
cca95b2637 |
fix(datanode): stop the object store WAL before the region server (#9639)
* fix(mito2): cancel flushes waiting for the WAL when the engine stops A flush waits for the WAL to be durable through the entry id it records before it writes its manifest edit. With the object store WAL in the `enqueued` mode that wait lasts until an upload succeeds, while the log store is stopped only after the engines. A graceful shutdown during an outage of the WAL's object store therefore never finished: closing a region does not cancel its running flush, stopping the worker group waits for running flushes and runs the queued ones, and a region holding writes that skipped the WAL is flushed on close. The worker group now owns a cancellation token, raised before its schedulers stop and kept raised. `RegionEngine::begin_stop`, which the region server calls before it closes regions, raises it earlier, so that a close-time flush cannot wait either. A flush polls its durability barrier first and gives up at the barrier once the token is raised: it writes no manifest edit, deletes its SSTs and fails as cancelled. A flush whose entries are already durable commits as before. The region server reports a failed close, which loses the region's writes that skipped the WAL, in its stop result. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * test(mito2): wait for the outage on the store under test Signed-off-by: jeremyhi <fengjiachun@gmail.com> * fix(mito2): give a healthy WAL a grace period before a stopping engine cancels flushes A flush of a stopping engine no longer gives up its wait for the WAL as soon as the stop begins. It gives up at once while the log store reports that it fails to make entries durable, through the new `LogStore::is_failing`, and otherwise 5 s after the engine began to stop, so a healthy object store WAL uploads its last entries first and a close-time flush keeps the writes that skipped the WAL. The object store WAL reports that it fails while the last create of its backlog failed, which covers a refused and an unreachable object store, and once it is poisoned or stopped. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * fix(mito2): drain the WAL for a stopping flush and cancel its failed waits A flush of a stopping engine now asks the log store to drain before it waits out the grace period, through a new `LogStore::begin_drain` that does nothing by default. The object store WAL then seals its open batch at once and, until it stops, as soon as a durability wait needs a region it holds, so a healthy store makes the entries durable after about one upload whatever its `flush_interval`, which has no upper bound. A barrier wait that fails while the engine stops, for instance on a poisoned or stopped log store, now fails the flush as cancelled like a wait that gives up. The healthy-store test no longer depends on the seal timer: its store never seals on its own, and the close must finish well within the grace period. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * fix(mito2): drain the WAL when the engine begins to stop The drain started only when a stopping flush reached its durability barrier, so a close behind slow flushes that reached it after the grace period had no time left to upload its entries. The worker group now asks the log store to drain as soon as it begins to stop, and the object store WAL seals its open batch then and every later admission right away, so its entries become durable while the flushes still write their SSTs. `LogStore::begin_drain` becomes synchronous for that. The healthy-store test now writes after the stop began and closes the region only after the grace period. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * fix(log-store): refuse appends once the object store WAL drains A WAL write admitted late in a stop could still be uploading when the close-time flush of its region reached an expired grace period, so the writes that skipped the WAL in that region were lost although the store was healthy. A draining object store WAL now refuses every later append, also one held back, with the stopped error, so the entries a stopping engine waits for are fixed when the stop begins and become durable while the regions still close; writes that skip the WAL are still taken. Tests cover a close-time flush whose wait completes within the grace period once a parked create is released, a close that reaches its wait after the grace period, and a drain refusing a held-back append. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * fix(datanode): word a failed region close neutrally in the shutdown log A close can fail for reasons other than its flush, such as a stopped worker, and for regions without writes that skipped the WAL, so the error line no longer states that such writes were lost: it says they may be. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * fix(datanode): stop the object store WAL before the region server The datanode stopped the object store WAL only after the region server, whose stop waits for flushes that wait for the WAL to become durable, so a shutdown never finished while the WAL's uploads kept failing. The WAL now stops first: its stop uploads the backlog or drops it and answers every wait for its entries, so a running, queued or close-time flush either commits or fails with the WAL-stopped error without writing a manifest edit. This replaces the stop signal and grace period the engine used to give up those waits, and the log store and engine hooks they needed. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * test(datanode): check rows after a healthy shutdown and wait for the WAL stop Signed-off-by: jeremyhi <fengjiachun@gmail.com> * test: keep only the shutdown tests that prove the new stop order The engine-level tests pinned the log store's stop contract, which the log store's own tests already cover, and the stalled-create test asserted a limitation rather than a guarantee. Drop them with the testing hooks only they used; the datanode and region server tests cover the change end to end. Signed-off-by: jeremyhi <fengjiachun@gmail.com> --------- Signed-off-by: jeremyhi <fengjiachun@gmail.com> |
||
|
jeremyhi
|
baa5630119 |
perf(log-store): keep the object store WAL reclaim boundary independent of the region count (#9644)
* perf(log-store): keep the object store WAL reclaim boundary independent of the region count The WAL actor recomputed the reclaim boundary after every watermark update, release and indexed object by scanning the footers of the objects from the previous boundary, checking each segment against the obsolete watermarks. Each scan started at the first segment of the boundary object, so when one object holds segments of many regions whose watermarks advance one region at a time, every update inspected the segments already known to be obsolete again, about R^2/2 footer entries in total for R regions. The catalog now keeps the obsolete watermarks itself and counts the live segments of every object. A watermark move visits only the segments of its region that it newly reaches, through the per-region index, and the boundary is the first object at or above the floor that still holds a live segment. The boundary is unchanged: segments of release records never hold it, a region without a watermark does, a watermark of EntryId::MAX releases every segment of its region, and it never moves below the floor. The store records each watermark in the catalog together with the one reads use. The per-region index is now keyed by the largest entry id of each segment, which is also object sequence order because the entry ranges of a region increase strictly with the sequence. Indexing an object already enforced that; it now checks the neighbours by entry id instead of by sequence, which accepts and rejects the same objects. A property test checks the boundary against a scan of the footers over random appends, watermark moves, releases, drops, collections and recoveries, and ignored measurement tests time watermark updates and tail reads through the store. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * perf(log-store): seek the object store WAL objects of a region by entry id A read found the objects of its region by filtering every object of the region that the catalog retains, so a read of the last few objects cost time in proportion to the whole retained history of the region, under the catalog read lock. The per-region index is keyed by the largest entry id of each segment, and the entry ranges of a region increase strictly with the object sequence, which the catalog enforces when it indexes an object, whatever entry ids the objects carry. So the lookup seeks to the first segment that ends at or after the start of the range and stops at the first that starts after its end, and the result is still in sequence order. The property test of the catalog now also checks range lookups against a scan of the footers. Signed-off-by: jeremyhi <fengjiachun@gmail.com> --------- Signed-off-by: jeremyhi <fengjiachun@gmail.com> |
||
 DeviousCardi
|
095c288f16 |
fix(promql): honor @ modifier on subqueries (#9446)
* fix(promql): honor @ modifier on subqueries `prom_subquery_expr_to_plan` never read the `at` field of a subquery, so `sum_over_time(m[20s:10s] @ 60)` was evaluated like `m[20s:10s]` and followed the outer evaluation grid. Handle `@` the way Prometheus does: fold the anchor into the subquery offset (`offset + (eval_start - anchor)`, as `setOffsetForAtModifier` does), evaluate the subquery once at the start of the evaluation, and replay the folded window over the outer grid, reusing the anchored range selector machinery. A call whose range argument is an anchored subquery is promoted as step-invariant, like one over an anchored range selector. `@ start()`/`@ end()` and `offset` combine as for selectors, and an unrepresentable anchor is rejected with the existing error. Closes #9411 Signed-off-by: Aarav <aaravsjadav@gmail.com> * test(promql): add sqlness cases for @ on subqueries Cover the #9411 reproducer end to end: `@ <ts>`, `@ start()`/`@ end()`, `@` combined with `offset` in either order, an instant query away from the anchor, `*_over_time` functions and `rate` over an anchored subquery, a computed inner expression, and an aggregation above the anchored call. Two series per table so a dropped or merged series shows up as a wrong row. Refs #9411 Signed-off-by: Aarav <aaravsjadav@gmail.com> --------- Signed-off-by: Aarav <aaravsjadav@gmail.com> |
||
|
jeremyhi
|
e68e8c46d8 |
perf(log-store): reuse the release records the object store WAL recovery already read (#9645)
* perf(log-store): reuse the release records the object store WAL recovery already read Recovery reads every object it lists to verify its header and footer: an object of at most 64 KiB whole, a larger one as its header and the last 64 KiB. It then read the segment of release records of every indexed object that holds one again with a range read, although the segment, the last one before the footer, is almost always among the bytes already read. A prefix of many dropped regions, each released in an object of its own, paid two reads per such object. The footer read now keeps a copy of that segment when the bytes it read hold it, and recovery decodes the copy, or reads the segment as before when the copy is missing, only for the objects on the selected chain. The segment is copied so that the read window is not held until the releases are decoded. Chain selection, the merge of the largest released entry id per region, and the failure of recovery on a release segment that cannot be fetched or decoded, whatever on_corrupted_segment says, are unchanged. Tests cover the reads of a recovery whose release segments are all in the read bytes, a release segment that starts before the tail window, and that a damaged release segment fails recovery on the chain and is never decoded off it. An ignored measurement test counts the reads and times recovery over a prefix of dropped regions. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * perf(log-store): bound the release segments object store WAL recovery keeps Recovery keeps the segment of release records of every object it fetches before it selects the chain, so an object off the chain could hold up to the tail window until recovery ends. Keep a segment only up to 4 KiB, far above the records of the few regions a release seals at once, and read a larger one again only when its object is on the chain, as before. Tests cover a release segment inside the tail window of an object read as its header and its tail, which exercises the offset into the window, and a segment above the kept size in an object read whole. Signed-off-by: jeremyhi <fengjiachun@gmail.com> --------- Signed-off-by: jeremyhi <fengjiachun@gmail.com> |
||
|
dennis zhuang
|
822c90c15f |
fix(test): download darwin release assets for compat runs on macOS (#9630)
Signed-off-by: Dennis Zhuang <killme2008@gmail.com> |
||
|
Weny Xu
|
98f79d0cf9 |
fix(meta): allow affinity fallback during failover (#9641)
Signed-off-by: WenyXu <wenymedia@gmail.com> |
||
|
jeremyhi
|
0e82b29b23 |
ci: add crash, restart, multi-region and drop cells to the S3 WAL benchmark (#9632)
* ci: add crash, restart, multi-region and drop cells to the S3 WAL benchmark Extend the object store WAL S3 benchmark beyond one load per configuration: - Inputs for a full image reference, the configurations to run (R0 adds Raft Engine with the standalone default sync_write = false), extra environment for every server container, a restart after each load, and a JSON list of named cell groups that override those inputs. - Restart cells SIGKILL the loaded server, restart it on the same prefix and data home, and time /health, the first correct count and the log store recovery; min_objects fails a cell that left too few objects indexed. - Crash groups write striped 1000-row batches from several writers, kill the server at a random point, restart it and check every acknowledged batch against the table (durable: present exactly once; enqueued: no duplicate, no partial batch, losses counted). A pinned variant keeps one row of a second table unflushed and samples the indexed objects while it runs. - Idle windows scrape the server before and after a quiet period. - A load group may split the loaded table into equal event_time ranges, so every batch writes to every region, and checks that each region took rows. - Drop groups write a table, drop it, kill and restart the server, then write and flush another table and wait for the dropped table's objects to be collected. - The summary adds WAL traffic, objects per second, mean object size, entries and appends per object, S3 requests per operation, server and host CPU, the loader's own request latency, rows per region, crash cycles, and each restart's replay per region from the server log. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * ci: report the object sequences a replay spans instead of objects replayed The width of a region's replayed entry range counts every object sequence in it, including objects that hold no entry of the region and are skipped, so the summary labels it as a span. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * ci: fail a crash cycle that loses a batch an earlier restart recovered The oracle accepted every missing acknowledged batch in the enqueued mode, so a batch that one restart recovered and a later restart no longer found passed. Every cycle keeps what its restart found, and a cycle now fails in either mode unless every batch an earlier restart recovered is still present, so only the current cycle's backlog that was not durable at its kill may be missing. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * ci: keep the pinned WAL pinned and label OpenDAL operations as such The pin table inherited the engine's 10-minute auto flush, so a longer pinned cycle could flush its row and release the WAL before the kill. It now sets its own auto_flush_interval of one day, longer than any crash group can run, and a pinned cycle fails unless its indexed objects never fall while it runs. The summary's request counts come from opendal_operation_duration_seconds, which counts OpenDAL calls rather than HTTP requests (a paginated list or a multipart write is one operation), so the fields and columns now say S3 operations. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * ci: check a pinned cycle's retention by the collected-object counter Net growth of the indexed objects can hide collection, and short cycles take no periodic sample, so a pinned cycle now fails unless deleted_objects_total is the same in the scrape before its writers start and in the one before the kill, both of the same server process. The operation columns name OpenDAL. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * ci: let a pinned cell collect only the objects below its pin The first write after a start lets the store collect the empty start object asynchronously, so a deletion after a cycle's baseline does not mean the pin let go. A pinned cell now scrapes the server before it writes the pin row and fails only once its servers together have collected more objects than were indexed then, counting the first server from that scrape and every restarted one from its start. Signed-off-by: jeremyhi <fengjiachun@gmail.com> * ci: test that a pinned cell's allowance spans its server restarts Two servers that each collect fewer objects than lie below the pin but more together now fail the cell in the second cycle. Signed-off-by: jeremyhi <fengjiachun@gmail.com> --------- Signed-off-by: jeremyhi <fengjiachun@gmail.com> |
||
|
dennis zhuang
|
8b0a155702 |
perf(otlp): chunk trace spans across resource groups (#9621)
* perf(otlp): chunk trace spans across resource groups Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix(otlp): plan spans of a mixed binary/JSON chunk individually Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * docs(otlp): describe cross-group trace chunking Signed-off-by: Dennis Zhuang <killme2008@gmail.com> --------- Signed-off-by: Dennis Zhuang <killme2008@gmail.com> |
||
|
dennis zhuang
|
672e94f61d |
feat: add column encoding option and use it for trace span ids (#9629)
* feat: add column encoding option and set it for trace span ids COMPRESSION WITH accepts encoding = 'plain' | 'delta_binary_packed'. Either disables the dictionary for the column; delta_binary_packed is limited to integer and timestamp columns. OTLP trace v1/v2 tables now create span_id and parent_span_id with plain encoding and 64KiB pages, and timestamp_end with delta_binary_packed. Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * test: add compatibility cases for column encoding options Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * test: move column encoding downgrade checks into downgrade_compatibility Signed-off-by: Dennis Zhuang <killme2008@gmail.com> --------- Signed-off-by: Dennis Zhuang <killme2008@gmail.com> |