v0.9.0

What's Changed

• Bump vite from 6.2.4 to 6.2.5 in /ui by @dependabot[bot] in #641
• Add sharding_key for TopNAggregation source measure by @ButterBright in #636
• Bump vite from 6.2.5 to 6.2.6 in /ui by @dependabot[bot] in #644
• Fix a bug and update API by @hanahmily in #645
• Add sharding_key for measure in BanyanDB ui by @ButterBright in #647
• Restore e2e by using latest oap by @wu-sheng in #648
• Update versions.md by @wu-sheng in #649
• Change node_selector to stages for groups in UI. by @Fine0830 in #650
• Fix the issue that the etcd watcher gets the historical node registration by @hanahmily in #651
• Update lifecycle.md by @wu-sheng in #653
• Enable Dynamic Reloading of TLS Certificates and Keys by @onkar717 in #642
• Add stability checks on recreated files by @hanahmily in #655
• feat: add TLS support between Liaison and Data nodes with integration… by @wesamabed in #646
• [WIP] add lifecycle query test by @hanahmily in #657
• Fix the Measure Fields by @Fine0830 in #658
• Fix(UI): add the stages filter to yaml configuration by @Fine0830 in #659
• Add query test cases of lifecycle by @hanahmily in #660
• Feature 12893/add topn subcommand by @OmCheeLin in #656
• feat(metadata):expose auto-compaction by @raiki02 in #654
• Feature/13035 aws s3support by @Tsukilc in #652
• Bump vite from 6.2.6 to 6.3.4 in /ui by @dependabot[bot] in #661
• bug/fix Panic when creating _top_n_result #13183 by @OmCheeLin in #664
• Implement multiple groups query in one request by @hanahmily in #667
• Fix the unstopping syncing from etcd by @hanahmily in #668
• Replica: Replace Any with []byte Between Liaison and Data Nodes by @hanahmily in #669
• Fix the crash when collecting the metrics from a closed segment by @hanahmily in #671
• Replace defaultIOSize with protector memory limit by @Guo-Chenxu in #672
• Fix the data race by @hanahmily in #673
• Support configurable replica count on Group by @hanahmily in #675
• fix potential oom by @OmCheeLin in #674
• Move the TopN pre-calculation flow from the Data Node to the Liaison by @hanahmily in #677
• Add a wait and retry to write handlers to avoid the local metadata cache not being loaded by @hanahmily in #678
• Add e2e test to lifecycle management by @hanahmily in #679
• Implement versioning properties and replace physical deletion with the tombstone mechanism for the property database by @mrproliu in #681
• Add Load Balancer Feature to Liaiso by @hanahmily in #685
• Physics delete property when merging segment by @mrproliu in #684
• Implement primary block cache for measure by @ButterBright in #670
• Implement fadvise for large files to prevent page cache pollution by @Eternity1824 in #680
• Fix topN parsing panic when the criteria is set by @hanahmily in #689
• Implement dictionary encoding by @ButterBright in #682
• Support delete properties when query by @mrproliu in #686
• Introduce the Trace data model to store the trace/span data by @hanahmily in #691
• Implement skipping index for stream by @ButterBright in #665
• Support skipping index and tree index in BanyanDB ui by @ButterBright in #693
• Add documentation for skipping index by @ButterBright in #666
• Encode each column based on its data type by @OmCheeLin in #688
• Update TSDB version to v1.3.0 in documentation by @ButterBright in #696
• Remove useless console print by @mrproliu in #699
• Push down aggregation for topN query by @sollhui in #697
• Fix panic in Bloom filter decoding by @ButterBright in #701
• Support save repair state tree in property by @mrproliu in #692
• Support dictionary encoding for low cardinality columns by @ButterBright in #695
• Introduce write queue mechanism in liaison nodes by @hanahmily in #705
• Remove the indexed_only field in TagSpec by @ButterBright in #703
• Add the property background repair strategy documentation by @mrproliu in #702
• Remove the indexed_only field in BanyanDB ui by @ButterBright in #707
• Add Trace module metadata management by @hanahmily in #708
• Enhance debugging experience for e2e tests by @ButterBright in #706
• Feature/azure support by @Eternity1824 in #700
• fix(UI): avoid shardingKey.tagNames is undfined by @Fine0830 in #710
• Add chunked data sync to improve memory efficiency and performance during data transfer operations by @hanahmily in #711
• Introduce gossip propagation protocol by @mrproliu in #698
• Support background property repair by @mrproliu in #712
• Implement migration system for both measure and stream data with file-based approach by @hanahmily in #716
• Improves the handling of node-specific data in the top N processing. by @hanahmily in #718
• Enable Login with Username/Password in BanyanDB by @OmCheeLin in #704
• Push down min/max aggregation operation to data nodes by @sollhui in #694
• Updated migration visitor constructors to accept target stage interval by @hanahmily in #719
• Feature/Google Cloud Storage Support by @Eternity1824 in #717
• Introduce Trace System in background repair property feature by @mrproliu in #714
• fix: correct decode bug in tag.go and column.go by @hanahmily in #720
• Enhance flusher and introducer loops to support merging operations by @hanahmily in #721
• Update storage cache and protector implementations by @hanahmily in #723
• Trace sidx design by @hanahmily in #724
• Fix memory leak in measure package by @hanahmily in #726
• New storage engine for trace by @ButterBright in #713
• A patch to the block cache by @hanahmily in #727
• Secondary index for trace(1/6) by @hanahmily in #728
• Secondary index for trace: interface by @hanahmily in #729
• Secondary index for trace: interface mocking by @hanahmily in #730
• Secondary index for trace: memory management by @hanahmily in #731
• Secondary index for trace: snapshot by @hanahmily in #732
• implement dynamic reload of credential files by @OmCheeLin in #734
• Implement traceID-based query by @ButterBright in #735
• Add benchmark and fix some issue of background property repair by @mrproliu in #725
• Add BDSL documentation for BanyanDB Query Language by @hanahmily in #737
• Secondary index for trace: query process by @hanahmily in #738
• Implement atomic counters for total queries and writes in sidx by @hanahmily in #740
• Enhance access log functionality with sampling option and add tests by @hanahmily in #739
• Add integration test for trace by @hanahmily in #741
• Integrate sidx writing, flushing and merging into core trace storage by @ButterBright in #742
• Add query by trace id test case to the trace model by @hanahmily in #743
• Fix unit test TestQueryResult in trace by @ButterBright in #744
• Add filter and order test cases to the trace model integration test suit by @hanahmily in #746
• Change to using pod name as Grafana variable by @mrproliu in #747
• Improve the trace model for the limit by @hanahmily in #748
• Implement sidx flushing and merging by @ButterBright in #745
• Add unit tests for sidx merge operation by @ButterBright in #749
• Implement a resilient publisher by @hanahmily in #750
• Write index before core trace storage by @ButterBright in #751
• feat: optimize gRPC message size limits for better time-series data handling by @hanahmily in #752
• Add query access log support by @hanahmily in #754
• Fix panic when repair property by @mrproliu in #753
• Enhance query access logging across gRPC services by @hanahmily in #755
• Fix query access log Unmarshal error by @mrproliu in #756
• Fix bugs time query range will not automatically update and add some new time query ranges by @peachisai in #757
• Fix the merging panic by @hanahmily in #759
• Refactor BydbQL document by @hanahmily in #760
• feat: improve trace query functionality by @hanahmily in #762
• Bump vite from 6.3.4 to 6.3.6 in /ui by @dependabot[bot] in #763
• Fix variable assignment in OpenDB function to use group name by @hanahmily in #765
• Refactor OpenDB functions to use 'Name' field from groupSchema.Metadata by @hanahmily in #766
• Implement versioning support in API by @hanahmily in #768
• Introduce the disk monitor to force cleanup disk by @hanahmily in #771
• reduce the repair cron time for the UT by @mrproliu in #772
• Fix sorting trace issue by @hanahmily in #773
• Fix the unexpected result when querying by traceID and tag belongs to a index rule by @hanahmily in #774
• Fix returning empty result when using IN operator on array type tags by @hanahmily in #775
• Fix etcd prefix matching any key that starts with this prefix by @hanahmily in #776
• Fix memory leaks and OOM issues in streaming processing by @hanahmily in #777
• Implement cluster mode for trace by @ButterBright in #767
• update trace e2e by @wankai123 in #779
• Fix the property repair test failure by @mrproliu in #780
• Fix the sorting timestamps issue of the measure model by @hanahmily in #781
• Update Go version to 1.25 and upgrade dependencies by @hanahmily in #783
• Refactor readData method in block.go by @hanahmily in #782
• Fix data race by @hanahmily in #785
• Fix the OOM when Merging Trace by @hanahmily in #787
• fix(trace): prevent tagType corruption in partMergeIter by @hanahmily in #789
• Add Having method for bloom filter by @hanahmily in #786
• Update timeout configurations for liaison services to 1 minute by @hanahmily in #790
• Add TraceIDFormat constants and update trace handling logic by @hanahmily in #791
• Refactor mergeBlocks function to simplify block handling by @hanahmily in #792
• Fix the bug that sidx doesn't support array value tags by @hanahmily in #793
• Fix syncSnapshot for trace by @ButterBright in #788
• Compressing the Spans of Trace by @hanahmily in #795
• Fix the bug of node selector in newLiaisonCmd by @hanahmily in #798
• Fix the unset partID by @hanahmily in #799
• Add trace query command in bydbctl by @mrproliu in #797
• Implement multi-group query for trace by @ButterBright in #800
• feat: add QL protocol by @kezhenxu94 in #778
• Fix long latency of trace merge by @hanahmily in #801
• Allocate Bloom filter from memory pool by @ButterBright in #802
• Deduplicate spans across replicas using span_id by @ButterBright in #794
• Fix trace query with empty span_id list by @ButterBright in #803
• Enhance block reading and writing with raw block handling by @hanahmily in #804
• Update required and missing selection in bydbQL documentation by @mrproliu in #805
• Introduce Streaming Query Pipeline for Sidx and Trace by @hanahmily in #806
• Add test case and update document of query measure from bydbQL by @mrproliu in #807
• Add property test cases and related QL by @mrproliu in #808
• Add trace related QL test cases by @mrproliu in #811
• Adding span_id and trace_id to Trace Proto by @hanahmily in #812
• feat: Implement Trace views by @Fine0830 in #809
• Avoid tag sorting in unmarshalTagFromSeqReaders by @ButterBright in #814
• Add stream related QL test cases by @mrproliu in #810
• Add top-N related QL test cases by @mrproliu in #816
• Remove unnecessary SNP ID check in property UT by @mrproliu in #818
• fix(UI): correct the tagProjection for trace query by @Fine0830 in #817
• Bump vite from 6.3.6 to 6.4.1 in /ui by @dependabot[bot] in #819
• Apply traceIDFilter in the query process to perform filtering by @ButterBright in #813
• Use Fetch request to instead of Axios request and remove Axios by @Fine0830 in #820
• Fix Several Bugs Related to Trace Query Performance by @hanahmily in #815
• Refactor CI Workflows to Use Centralized Preparation Step by @hanahmily in #822
• feat(UI): implement Trace Tree for debug mode by @Fine0830 in #823
• Introduce bydbQL for query data by @mrproliu in #821
• Remove TSB of BydbQL in document menu by @mrproliu in #824
• fix(UI): moved "Debug Trace" button from below the table to above it and correct debug trace data by @Fine0830 in #826
• fix missing HTTP endpoint of bydbQL by @mrproliu in #827
• feat(UI): Implement the Query Page for BydbQL by @Fine0830 in #829
• Fix the property repair UT by @mrproliu in #825
• Fix Several Flaws of Trace Query by @hanahmily in #830
• Suppport backup and restore trace data by @mrproliu in #831
• Fix dist usage error and add trace related metrics into storage selection in the Grafana dashboard by @mrproliu in #833
• Implement Handoff Queue for Trace by @hanahmily in #832
• refactor(UI): Optimize router for better usability by @Fine0830 in #834
• Add command-line tool for dumping BanyanDB trace data by @hanahmily in #835
• Implement Backoff Retry Mechanism for Sending Queue Failures by @hanahmily in #836
• Fix restore delete wrong file issue by @mrproliu in #838
• Fix comparison issues in TopN test cases by @peachisai in #839
• Delete empty files directory when restoring by @mrproliu in #841
• feat(ui): implement the prompt and highlight for BydbQL by @Fine0830 in #840
• Skip trace catalog and empty directory in the lifecycle process by @mrproliu in #842
• Fix the time range error in the lifecycle by @mrproliu in #843
• feat(ui): introduce the TanStack Query and optimize build configuration by @Fine0830 in #845
• Improve sidx scan efficiency by @hanahmily in #844
• Remove Unnecessary Slice Allocation of Sidx Query by @hanahmily in #847
• Refactor lifecycle logical by @mrproliu in #848
• Fix missing sidx data when take snapshot of trace data by @mrproliu in #850
• Update handoff controller to use maxTotalSizeBytes by @hanahmily in #853
• Fix empty tag of SIDX by @hanahmily in #854
• Fix trace syncer flush handling by @hanahmily in #855
• Liaison Memory Shedding by @hanahmily in #857
• Update dependencies and improve revive configuration by @hanahmily in #859
• Implement Additional Dump Commands by @OmCheeLin in #858
• Refactor logging in localIndexCallback to use debug level by @hanahmily in #860

New Contributors

• @onkar717 made their first contribution in #642
• @wesamabed made their first contribution in #646
• @raiki02 made their first contribution in #654
• @Tsukilc made their first contribution in #652
• @Guo-Chenxu made their first contribution in #672
• @Eternity1824 made their first contribution in #680
• @peachisai made their first contribution in #757

Full Changelog: v0.8.0...v0.9.0