Merge main into lucene_snapshot #132263
Merged
Merge main into lucene_snapshot #132263
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…stic#131517) For CPS S2D9, we'd need field caps to refresh potentially lost connections before executing a query by explicitly establishing a connection with a short timeout to avoid waiting for large duration. This is similar to what we recently did with `_search` and will help for ES|QL as well.
… testTimeoutSetting elastic#132179
Current knn search over nested vectors only supports filtering on parent's metadata. This adds support for filtering over nested metadata. Closes elastic#106994 Closes elastic#128803
We set a 120s timeout on the repo analysis that runs in these tests, but the REST client in use has a read timeout of 60s so any analysis that lasts longer than that will yield an HTTP timeout that contains very limited information to help further investigation. With this commit we extend the HTTP timeout to be a few seconds longer than the analysis timeout so we should get a proper response in these cases.
…UpdateIT testDenseVectorMappingUpdate {initialType=bbq_flat updateType=bbq_disk #2} elastic#132184
The number of plugin archives changed and triggered the assertion. Changing the assertion to check for `analysis-icu` plugin archive instead of asserting on number of archives. This will always fail when new plugin is added (like microsoft-graph-authz). Resolves elastic#116866
…UpdateIT testDenseVectorMappingUpdate {initialType=bbq_hnsw updateType=bbq_disk} elastic#132188
…UpdateIT testDenseVectorMappingUpdate {initialType=int8_flat updateType=bbq_disk #2} elastic#132189
This commit presents a hierarchical layer on top of the DiskBBQ centroids to reduce the number of centroids scored at search time.
Support for (the deprecated) path.shared_data was previously not covered in any REST test and broke unnoticed when we introduced entitlements. The problem only surfaced when eventually enabling entitlements on internalClusterTests. This PR adds a basic test to check indexing using a relative (to path.shared_data) data_path index setting succeeds with sufficient entitlements for path.shared_data. Relates to ES-12447
A failure of a write to a register does not indicate that the write has definitely completed - it may eventually succeed at some later point in time.
…131937) It is possible for a linked remote to get unlinked in between the containsKey() and get() calls in collectNodes(). This change adds a test that produces the NullPointerException and adds a fix.
In some contexts ESQL's HeapAttack tests were failing because the TranslogWriter keeps a map with the contents of the documents in it around. But only when assertions are enabled. Fun. This flushes the map away, preventing the odd OOM we were seeing in the HeapAttack tests.
Speeds up the spec tests by: * Randomize `async`/`sync` instead of both * Check the took time rarely * Don't check if index exists every time * Reduce transfer when checking breaker * Cache capabilities checks The first one is most of the speed up but the rest save another couple of minutes. Time drops from 13m45s to 6m or so on my laptop.
…UpdateIT testDenseVectorMappingUpdate {initialType=int4_hnsw updateType=bbq_hnsw} elastic#132211
…UpdateIT testDenseVectorMappingUpdate {initialType=int8_hnsw updateType=bbq_disk #2} elastic#132213
…UpdateIT testDenseVectorMappingUpdate {initialType=int8_hnsw updateType=bbq_disk} elastic#132214
…Tests testCacheCleanupOnProjectDeletion elastic#132218
If index throttling is enabled such that it pauses all indexing threads that try to index into a shard, this can starve other tasks such as relocation that try to acquire all indexing permits. This PR addresses this by suspending throttling to allow the indexing threads that are holding the permits to pass. Addresses ES-11770.
…UpdateIT testDenseVectorMappingUpdate {initialType=int4_flat updateType=bbq_hnsw} elastic#132219
…UpdateIT testDenseVectorMappingUpdate {initialType=hnsw updateType=bbq_disk} elastic#132220
…UpdateIT testDenseVectorMappingUpdate {initialType=int4_flat updateType=bbq_disk} elastic#132221
…dIntoLogsDbChallengeRestIT testTermsQuery elastic#132225
…xedLogsDbChallengeRestIT testTermsQuery elastic#132226
…UpdateIT testDenseVectorMappingUpdate {initialType=int4_hnsw updateType=bbq_disk #2} elastic#132228
…#131981) The ml-cpp PR elastic/ml-cpp#2846 introduces changes to how memory values are calculated and reported for Linux platforms. This PR adjusts test case values accordingly.
… built in Editor and Viewer roles (elastic#129662) * Adding asset criticality and entity store permissions to built in roles * Update docs/changelog/129662.yaml * [CI] Auto commit changes from spotless * Corrects entity store index pattern Updates the entity store index pattern to ensure it matches the minimum necessary index name and narrow it down to the correct use case --------- Co-authored-by: elasticsearchmachine <[email protected]>
…2157) This is the same failure as observed in elastic#129644 for which the original fix elastic#129680 did not really work. It did not work because the the ordering of checks. The shutdown marker is removed after the cluster passes ready check so that new shards can be allocated. But the cluster cannot pass the ready check before the shards are allocated. Hence the circular dependency. In hindsight, there is no need to put shutdown record for all nodes. It is only needed on the node that upgrades the last to prevent snapshot from completion during the upgrade process. This PR does that which ensures there are always 2 nodes for hosting new shards. Resolves: elastic#132135 Resolves: elastic#132136 Resolves: elastic#132137
…UpdateIT testDenseVectorMappingUpdate {initialType=int4_flat updateType=bbq_disk #2} elastic#132234
The listener is expected to be called on failure case. This PR does that. Resolves: elastic#131979
…rderWithMissingOrderSupportQueryingNewNode elastic#132249
* Finalize docs for v9.1.0 release
Reenable INLINESTATS telemetry test that was disabled in elastic#116005. The test query is now planned and accounted for correctly.
elastic#115813 refactoring disabled a couple of INLINESTATS tests. These are working correctly now.
…stic#132154) This scorer gives a better tradeoff on scoring precision and speed and only requiring two additional bytes per vector comparing to int4.
ChrisHegarty
added
the
:Search/Search
|
Pinging @elastic/es-search (Team:Search) |
carlosdelest
approved these changes
Jul 31, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
23 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Merge main into lucene_snapshot