58 Commits
Author SHA1 Message Date
tellet-q b58a35004a Revert "feat(consensus): warn when applying a single entry stalls the consensus thread (#10215)" (#10241)
This reverts commit dfca67f5ed.
2026-09-03 13:19:40 +02:00
tellet-q d9d5a6e6b7 feat(consensus): warn when applying a single entry stalls the consensus thread (#10215)
* feat(consensus): warn when applying a single entry stalls the consensus thread longer than a threshold
2026-09-03 12:41:01 +02:00
tellet-q d35da46f6d test(resharding): test resharding state clears before the replica revert (#9656) 2026-09-03 12:36:30 +02:00
tellet-qandClaude Opus 4.8 16c8e53608 feat(model_testing): add --duration-sec to run for a fixed wall-clock time (#9506)
* feat(model_testing): add --duration-sec to run for a fixed wall-clock time

Bound the soak by wall-clock time instead of op count: when --duration-sec is set, the loop runs until the deadline (or Ctrl-C) and --op-num is ignored.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* Fix clippy

---------

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-08-04 11:16:53 +02:00
tellet-q df0466538d ci: bump test_many_collections timeout to avoid flaky CI timeouts (#9253) 2026-06-03 14:55:06 +02:00
tellet-q 3de12da31e Revert setting vm.max_map_count to higher value and increase pytest timeout for e2e tests (#9191) 2026-06-03 14:49:44 +02:00
tellet-q e7b5ebe893 Run Long Integration Tests against dev on schedule (#9100) 2026-05-22 10:47:48 +02:00
tellet-q 95b309b99f Fix flaky TestSnapshotsInterferenceWithConsensus (#9102) 2026-05-22 10:47:41 +02:00
tellet-q 2a3c93d57e Fix long running e2e test (#9090)
* store container logs

* increase vm.max_map_count
2026-05-22 10:45:46 +02:00
tellet-q 13c73334a1 [ai] test: wait for cluster readiness in flaky e2e test (#9012)
* [ai] test: wait for cluster readiness in flaky e2e test
2026-05-22 10:36:08 +02:00
tellet-q 33abc7a22e [ai] use readiness probe instead of a time.sleep (#8884) 2026-05-08 13:47:48 +02:00
tellet-q a18a09f5d0 test fix: add wait before starting new peer in test_cluster_rejoin (#8824)
* fix: add wait before starting new peer in test_cluster_rejoin

* Use wait_peer_added
2026-05-08 13:47:42 +02:00
tellet-q 9b4ac03a9a feat: use local cache when building e2e image locally (#8797) 2026-05-08 13:47:32 +02:00
tellet-q d3c931787e ci: improve docker build caches for e2e and consensus jobs (#8786) 2026-05-08 13:47:30 +02:00
912023968d Test if io_uring handles EINTR properly (#8578)
* Test if io_uring handles EINTR properly

* Fix unit test compilation after read_iter API change

Update test_io_uring_eintr_handling to match the new read_iter signature
that takes (Meta, ReadRange) tuples and returns Result<impl Iterator>.

Made-with: Cursor

* Install no-op SIGUSR1 handler in debug mode on Unix

Prevents SIGUSR1 from terminating the process with the default
disposition, so that io_uring EINTR tests can safely bombard
the process with signals.

Made-with: Cursor

* Enter tokio runtime context for SIGUSR1 handler, fix clippy

tokio::signal::unix::signal requires a reactor context, so enter
the runtime handle before installing the handler.

Also fix manual_let_else clippy warning in the EINTR unit test.

Made-with: Cursor

* Cleanup 🙄

---------

Co-authored-by: Cursor Agent <agent@cursor.com>
Co-authored-by: Roman Titov <ffuugoo@users.noreply.github.com>
2026-05-08 13:46:46 +02:00
tellet-q 24389577ae Adjust debug ShardTransferRestart logline format to match ShardTransfer's (#8387)
* Adjust debug logline's format of
ShardTransferRestart to match with ShardTransfer

* Convert into shardtransfer

* Deconstruct ShardTransferRestart struct
2026-03-26 18:40:24 +01:00
tellet-q b9e34b68b8 Add a test for dirty shard crashloop (#8440)
* Add a test for dirty shard crashloop
2026-03-26 18:24:29 +01:00
tellet-q c0e45c7987 Enable heap profiling in pyroscope (#8371)
* Start heap profiling if MALLOC_CONF set

* Update doc
2026-03-26 17:57:45 +01:00
tellet-q 5dfb72e0c5 Bump grafana/pyroscope-rs to v2.0.0 (#8361)
* Bump grafana/pyroscope-rs to v2.0.0
2026-03-26 17:50:58 +01:00
tellet-q 6bf5736595 Log full transfer info on aborting a shard transfer (#8345) 2026-03-26 17:43:22 +01:00
tellet-q db3867e49b Bump pyroscope version to lib-1.0.2 (#8282)
* Bump pyroscope version to lib-1.0.2

* Provide doc with pyroscope configuration
2026-03-26 17:33:51 +01:00
tellet-q fb1b45e166 Fix API response after resetting the debugger profiler (#8281) 2026-03-26 17:33:46 +01:00
tellet-q 2b579e1094 Use pyroscope lib-1.0.1 (#8247) 2026-03-26 17:23:14 +01:00
tellet-q 86775709f2 Upgrade pyroscope to 0.5.9 (jemalloc signal-safety fix) (#8225) 2026-03-26 17:16:19 +01:00
tellet-q d36ebb2a6a Upgrade reqwest to 0.13 and install ring as default rustls CryptoProvider (#8213)
* Upgrade reqwest to 0.13 and install ring as default rustls CryptoProvider

* Do not compile aws-lc-rs

* Update tests

* Revert changes
2026-03-26 17:09:43 +01:00
tellet-q 506b69297f Add test for snapshot backpressure freeze (#8197)
* Add test for snapshot backpressure freeze
2026-03-26 17:02:10 +01:00
tellet-q 0951e03c5f Add new stage in TransferStage (#8113)
* Add new stage to stream records transfers

* Report better names to users
2026-02-13 09:55:46 +01:00
tellet-q 22fb1c5bc4 Add a test to check if creating snapshots blocks other operations (#7510)
* Add a test to check if creating snapshots blocks other operations
2026-02-13 09:55:36 +01:00
tellet-q 4cf0bbd63c Add test for consensus' freeze during streaming snapshot (#8075) 2026-02-13 09:55:36 +01:00
tellet-qandtimvisee 79d0678440 Add more stages for shard transfer profiling (#8027)
* Improve stages reporting

* Remove transferring records info from snapshot transfer's comment

* Add detailed profiling for ReshardingStreamRecords

* Address AI review comment

* Remove some nesting

---------

Co-authored-by: timvisee <tim@visee.me>
2026-02-10 00:03:26 +01:00
tellet-q 55969c7e97 Track shard transfer stages and duration (#8001)
* Track shard transfer stages and duration
2026-02-09 23:36:50 +01:00
tellet-q c18f7995a8 Move TestDelay to CollectionUpdateOperations and add generic staging API (#7716)
* Move TestDelay to CollectionUpdateOperations

* Address ai review

* Modify endpoint's path

* Address review
2026-02-09 22:56:10 +01:00
tellet-qandgenerall 30a34642b9 Add a test for recovery after kill during Partial (#7762)
* Add a test for recovery after kill during Partial

* Address AI review

* simplify test

* Merge pull request #7829

* introduce a new state

* switch to ManualRecovery for user-initiated snapshot operations

* fmt

* fix test and regen api

* test fix: instead of force recovery on update failed snapshots now re…

---------

Co-authored-by: generall <andrey@vasnetsov.com>
2026-02-09 22:35:29 +01:00
tellet-q 30c0632dc7 Push master images to DockerHub (#7836) 2026-02-09 22:32:35 +01:00
tellet-q f1d54ca889 Move test dependencies to tests (#7793)
* Move pyproject and uv.lock to tests folder

* Make tests working-dir agnostic

* Fix test

* Address review
2025-12-18 17:29:18 +01:00
tellet-q 855d611539 Init snapshot metrics on collection creation (#7788) 2025-12-18 17:29:17 +01:00
tellet-q c31d165601 Download archives once (#7711) 2025-12-18 17:27:28 +01:00
tellet-qandtimvisee 2982e2bc3b Add staging endpoint for testing point operation delays (#7687)
* Add endpoint to execute delay instead of point update

---------

Co-authored-by: timvisee <tim@visee.me>
2025-12-18 17:26:41 +01:00
tellet-q c6f6224de7 Long running e2e test on many collections (#7576)
* Add test for many collections

* Reduce number of collections
2025-12-03 10:23:25 +01:00
tellet-q 24036264b2 Add a ClusterOperation to execute a sleep on the node (#7616)
* Add TestSlowDown ClusterOperation to execute a sleep on the node
2025-12-03 10:20:11 +01:00
tellet-q 362db71f2d Build dev with new feature flag (#7612)
* Build dev images with new feature flag

* Disable feature for master and tags
2025-12-03 10:18:52 +01:00
tellet-q 808aadae92 Increase timeout for e2e tests (#7611) 2025-12-03 10:18:52 +01:00
tellet-q ead5c98221 Log shard transfer abort during consensus recovery (#7373)
* Add log line about aborting transfer during consensus recovery
2025-11-14 12:28:14 +01:00
tellet-q 55d761c1c9 Refactor bash tests: storage-compatibility (#7299)
* Add e2e test for compatibility
2025-11-14 12:28:13 +01:00
tellet-q b6ec9b006e Add e2e test for snapshots (#7066)
* Add e2e test that mimics 1 crasher's scenario
2025-09-29 11:28:28 +02:00
tellet-q faed7f3174 Add hw usage to Facet response (#7189)
* Add usage stats to Facet response
2025-09-29 11:26:00 +02:00
tellet-q d26cd77180 Refactor bash test for snapshots recovery (#7008)
* Refactor bash test for snapshots recovery

* Address review: minio check

* Address review
2025-09-29 11:24:40 +02:00
tellet-qandtimvisee 38c4d624f5 Add tests for immutable text index (#7153)
* Add unit test

* Flush text index deletes before reopening them

* Fix unused warning if compiling without RocksDB

* Add e2e test

---------

Co-authored-by: timvisee <tim@visee.me>
2025-09-29 11:24:40 +02:00
tellet-q f86d8af210 Refactor bash tests: tls (#6995)
* Refactor tls bash test

* Update codespell

* Address review
2025-08-11 13:18:04 +02:00
tellet-q b33e741229 Hardcode the link (#6985)
* Hardcode the link

* Fix image build
2025-08-11 13:16:52 +02:00
tellet-q cad14965e3 Refactor bash tests: low-ram (#6911)
* Refactor low-ran bash test

* Address review
2025-08-11 13:16:20 +02:00
tellet-q cf9862949c Refactor bash tests: low-disk (#6845)
* Add e2e_tests pytest test suite

* Remove old bash test
2025-07-17 13:57:56 +02:00
tellet-q 621b4b9ba1 A new test check that payload_io_read is withing some limit for a larger amount of segments (#6301)
* Add a test on payload_io_read value
2025-04-20 23:49:29 +02:00
tellet-q 820ea5a8a9 Use common actions pattern in workflows (#6061)
* Use common actions pattern

* Update gpu workflow
2025-03-21 11:41:50 +01:00
tellet-q a71e2ffe74 Add workflow that can be triggered (#6059)
an experiment
2025-03-21 11:39:05 +01:00
tellet-q d846017015 Remove unnecessary assert (#5413)
* Remove unnecessary assert
2024-11-12 14:39:18 +01:00
tellet-q dea45cb5ab Add tests on uuid payload index queries with filters (#4971)
* Add on_disk:True for uuid index schema check

* Add tests for uuid payload index

* Fix test

* Add test comparing filtered queries between keyword index and uuid index

* Remove order by case

* Clean up
2024-08-29 15:24:25 +02:00
c99caba0a3 Add test for OOD during indexing (#4267)
* Add test for OOD during indexing

* Start indexing right after 1st OOD message

* Only send search request after insert loop

* Fail early when encountering out-of-storage during optimization (#4578)

* fs4@0.8.4

* fail early on low storage

* move `dir_size` to `common`

* move the ood bailout to `SegmentOptimizer::optimized_segment_builder`

* drop dead code

* move dir_size to common subcrate

---------

Co-authored-by: generall <andrey@vasnetsov.com>

---------

Co-authored-by: xhjkl <xhjkl@users.noreply.github.com>
Co-authored-by: generall <andrey@vasnetsov.com>
2024-06-29 23:03:43 +02:00