| 6050bd5 | David Roberts | 24 May 2021, 12:21:03 UTC | Muting search/110_field_collapsing/field collapsing and search_after (#73320) Due to https://github.com/elastic/elasticsearch/issues/73316 | 24 May 2021, 12:21:03 UTC |
| 0216cf0 | David Roberts | 24 May 2021, 10:13:24 UTC | [ML] Switch ML internal index templates to composable templates (#73232) Legacy index templates are deprecated but ML was still using them for its hidden indices. This PR switches the legacy ML index templates to use the new composable index template framework. The composable index templates get installed once the master node is on a version that understands them. For templates that need to be up-to-date in mixed version clusters where the master might still be on a version that doesn't understand composable index templates we still ship the legacy template too, and install this if required in the mixed version cluster. (The notifications index template falls into this category.) This makes a couple of places in the code a little messy, as the new style template definitions don't contain a dummy _doc level (where the type used to be), but the legacy template definitions do - hopefully we can tidy this up in master once 8.0 is released. There is one more change of note in this PR that is not strictly related to switching to composable templates, but which was shown up during the testing. We used to wait for all templates to be installed by the master node before running tests in mixed version clusters. I do not believe we should have been doing this, as other upgrade orchestration systems, e.g. Cloud, will not be doing this. Our production code needs to install templates and/or mappings before any operation that requires them if there's a chance that the elected master won't have done this in time. Fixes #65437 | 24 May 2021, 10:13:24 UTC |
| 7080aef | Tim Vernum | 24 May 2021, 03:53:16 UTC | Add more logging to SAML Realm (#73279) This change adds more debug/trace logging to the SAML realm so that its behaviour can be traced more easily at runtime. Some of the information being added is also available from the role mapping log messages, however the runtime behaviour of the SAML realm should be observable without an over reliance on the logging of another package. | 24 May 2021, 03:53:16 UTC |
| 6c5ab8c | James Rodewig | 20 May 2021, 19:42:54 UTC | [DOCS] Fix data stream ref in index template docs (#73292) Updates an outdated reference to empty `data_stream` objects. The create index template API's `data_stream` parameter now supports the `hidden` property. | 20 May 2021, 19:42:54 UTC |
| f5aa824 | Armin Braun | 20 May 2021, 19:25:44 UTC | Stricter Parsing Shard Level Repository Metadata (#73269) Similar to #73268 we should be stricter here, especially when we are super-strict about additional fields anyway. Also, use our parser exception utils to get better exceptions if parsing fails. | 20 May 2021, 19:25:44 UTC |
| 41a9eae | James Rodewig | 20 May 2021, 18:40:10 UTC | [DOCS] Update `hidden` anchor (#73287) | 20 May 2021, 18:40:10 UTC |
| cb943db | James Rodewig | 20 May 2021, 16:21:20 UTC | [DOCS] Reuse index template body params in simulate template API (#73177) Changes: * Reuses and reorders the index template API's body parameters in the simulate template API docs. * Replaces several includes with a shorter xref. * Reformats a sidebar on naming collisions with built-in index templates. | 20 May 2021, 16:21:20 UTC |
| a11ea87 | James Rodewig | 20 May 2021, 16:08:40 UTC | [DOCS] Update exists API for data streams (#73180) Updates the exists API docs to better reflect its support of data streams and aliases. Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com> | 20 May 2021, 16:08:40 UTC |
| 522283c | Bogdan Pintea | 20 May 2021, 14:56:02 UTC | SQL: Add ODBC proxy support documentation (#73263) This documents how to configure the proxy support for ODBC. It also removes the documentation of the connection string values, these are now all covered by the GUI settings. Co-authored-by: Andrei Stefan <astefan@users.noreply.github.com> | 20 May 2021, 14:56:02 UTC |
| b13f43b | Armin Braun | 20 May 2021, 13:51:58 UTC | Refactor RestoreService Restore Path (#73258) Make the restore path a little easier to follow by splitting it up into the cluster state update and the steps that happen before the CS update. Also, document more pieces of it and remove some confusing redundant code. | 20 May 2021, 13:51:58 UTC |
| 81de369 | James Rodewig | 20 May 2021, 12:38:37 UTC | [DOCS] Fix query parameters for restore API (#73015) | 20 May 2021, 12:38:37 UTC |
| 25dcc62 | Armin Braun | 20 May 2021, 12:16:12 UTC | Fix SnapshotInfo.fromXContentInternal not Fully Consuming Parser (#73268) The parsing here was causing trouble with the new streaming deserialization because it did not fully consume the parser so if the internal buffer of the parser was just enough to finish reading the `"snapshot"` field but missed the closing bracket, then the stream behind the parser would not have been consumed fully. Also it was strangely lenient and would just read a broken in-progress `SnapshotInfo` if it ran into SMILE that contained any object field under any key that isn't "snapshot". I made it a little stricter now to enforce that we have a "snapshot" field and not just an object field by any name. | 20 May 2021, 12:16:12 UTC |
| 0f0cdcb | Henning Andersen | 20 May 2021, 09:18:42 UTC | [DOCS] Use partially mounted in autoscaling docs (#73229) Fixed autoscaling docs to no longer call partially mounted indices or shards for frozen indices/shards, now uses partially mounted indices or shards. Closes #73132 Co-authored-by: James Rodewig <40268737+jrodewig@users.noreply.github.com> | 20 May 2021, 09:18:42 UTC |
| 0b03f69 | Rene Groeschke | 20 May 2021, 06:46:20 UTC | Fix Spotless failing with InvocationTargetException on JDK 16 (#73246) | 20 May 2021, 06:46:20 UTC |
| b23dfe2 | Julie Tibshirani | 19 May 2021, 23:24:23 UTC | Adjust REST tests now that #73023 is backported | 19 May 2021, 23:24:23 UTC |
| f85a9dd | Julie Tibshirani | 19 May 2021, 21:21:18 UTC | Support field collapsing with search_after (#73023) This change adds support for using `search_after` with field collapsing. When using these in conjunction, the same field must be used for both sorting and field collapsing. This helps keep the behavior simple and predictable. Otherwise it would be possible for a group to appear on multiple pages of results. Currently search after is handled directly in `CollapsingTopDocsCollector`. As a follow-up, we could generalize the logic and move support to the Lucene grouping framework. Closes #53115. | 19 May 2021, 21:21:18 UTC |
| 9b99234 | Mark Tozzi | 19 May 2021, 20:04:36 UTC | Javadoc for how aggs work (#73214) Based a tech talk Nik gave, I just typed up the notes. | 19 May 2021, 20:04:36 UTC |
| 4162a57 | Gordon Brown | 19 May 2021, 19:09:35 UTC | Handle the existence of system data streams in Get Aliases API (#73244) This commit adjusts the behavior of the Get Aliases API to more thoroughly prevent errors and warnings from being emitted unnecessarily from the Get Aliases API by retrieving all indices including system ones and only warning in the post processing of the action. Additionally, the IndexAbstractionResolver has been updated to properly handle system data streams when evaluating visibility. Closes #73218 Co-authored-by: jaymode <jay@elastic.co> | 19 May 2021, 19:09:35 UTC |
| caffbd5 | Joe Gallo | 19 May 2021, 16:19:34 UTC | Remove obsolete ilm BWC checks (#73220) | 19 May 2021, 16:19:34 UTC |
| 70cfcf8 | Joe Gallo | 19 May 2021, 16:18:26 UTC | Remove obsolete datastream BWC checks (#73247) | 19 May 2021, 16:18:26 UTC |
| 8170376 | David Turner | 19 May 2021, 15:16:41 UTC | Restore global state does not merge (#73226) Today the docs indicate that restoring a snapshot with `include_global_state` set will merge the ingest pipelines, ILM policies, settings etc in the snapshot with those already in the cluster. This isn't the case, we simply replace all the things. This commit corrects the docs. | 19 May 2021, 15:16:41 UTC |
| 25dcfae | James Rodewig | 19 May 2021, 13:44:51 UTC | [DOCS] Fix reindex API intro (#73188) The destination doesn't need to pre-exist. | 19 May 2021, 13:44:51 UTC |
| fb212e4 | James Rodewig | 19 May 2021, 12:50:19 UTC | [DOCS] Fix typo | 19 May 2021, 12:50:19 UTC |
| c9a09e3 | James Rodewig | 19 May 2021, 12:37:16 UTC | [DOCS] Remove `_all` examples from 'Fix common cluster issues' (#73217) We plan to deprecate the `_all` wildcard pattern. This removes several `_all` examples from the 'Fix common cluster issues' guide. | 19 May 2021, 12:37:16 UTC |
| a6830ed | István Zoltán Szabó | 19 May 2021, 12:34:11 UTC | Fixes URL in security enroll node API json. (#73228) | 19 May 2021, 12:34:11 UTC |
| f273bcf | Martijn van Groningen | 19 May 2021, 12:15:34 UTC | Disable watcher debug logging in full cluster restart qa tests, maybe enabling it again when investigating watcher full cluster restart qa tests (#48381) | 19 May 2021, 12:18:17 UTC |
| 35460a5 | Przemyslaw Gomulka | 19 May 2021, 11:23:47 UTC | [Rest Api Compatibility] REST Terms vector typed response (#73117) Enabling the tests and adds a type field for termvector response the commit that enabled typed endpoints but missed to update the response #72155 | 19 May 2021, 11:23:47 UTC |
| 4b2c3ab | Martijn van Groningen | 19 May 2021, 08:07:11 UTC | The get aliases api should not return entries for data streams with no aliases (#72953) The get alias api should take into account the aliases parameter when returning aliases that refer to data streams and don't return entries for data streams that don't have any aliases pointing to it. Relates to #66163 | 19 May 2021, 08:07:11 UTC |
| e125cb1 | Benjamin Trent | 18 May 2021, 16:08:12 UTC | [ML] adds latest record timestamp to model snapshot deprecation warning (#73066) This adds the latest record timestamp of the deprecated model snapshot in the warning. This hopefully gives the user a better view into whether they would want to delete the snapshot or to upgrade it. | 18 May 2021, 16:08:12 UTC |
| 7a0eaab | David Turner | 18 May 2021, 15:59:45 UTC | Improve BlobStoreFormatTests#randomCorruption (#73201) This method today corrupts bytes until the checksum changes, but (a) it's comparing the checksum vs one computed before even reading the file, and (b) changing a single byte will always invalidate a CRC-32 checksum so the loop is unnecessary as is the checksum calculation. It also doesn't ever try truncating the file which is a realistic kind of corruption that we must be able to detect. This commit addresses all that. | 18 May 2021, 15:59:45 UTC |
| 8e46acf | Marten | 18 May 2021, 14:58:44 UTC | Fix typo in Rectangle() error message (#73124) Co-authored-by: James Rodewig <40268737+jrodewig@users.noreply.github.com> | 18 May 2021, 14:58:44 UTC |
| dfa5903 | Benjamin Trent | 18 May 2021, 14:23:37 UTC | [ML] fix empty body on post issue for datafeed _preview (#73205) It is common practice to pass an empty body on a `POST` call. Consequently, a `POST _ml/datafeed/<ID>/_preview` could fail if the body is empty as we try to parse it while IGNORING the ID in the URL. This commit fixes that bug. closes #73206 | 18 May 2021, 14:23:37 UTC |
| 814839e | Armin Braun | 18 May 2021, 14:19:00 UTC | Fix Edge-Case Threading Bug in TransportMountSearchableSnapshotAction (#73196) The callback to loading the repository-data may not run on generic in the uncached case because of the repo data deduplication logic. The same issue was fixed for the snapshot status API in https://github.com/elastic/elasticsearch/pull/68023 | 18 May 2021, 14:19:00 UTC |
| 06fc62f | Armin Braun | 18 May 2021, 13:43:34 UTC | Fix UpdateThreadPoolSettingsTests (#73199) Small and obvious oversight from #73172 | 18 May 2021, 13:43:34 UTC |
| 6a038b2 | Yang Wang | 18 May 2021, 13:19:27 UTC | Service Accounts - Ensure valid random service token name (#73098) The random token name is a base64 UUID. It can sometimes vilolate the validation rules. The base64 UUID is now prefixed with a string token_ to ensure the name is always valid. | 18 May 2021, 13:19:27 UTC |
| da24285 | Armin Braun | 18 May 2021, 12:40:39 UTC | Introduce SNAPSHOT_META Threadpool for Fetching Repository Metadata (#73172) Adds new snapshot meta pool that is used to speed up the get snapshots API by making `SnapshotInfo` load in parallel. Also use this pool to load `RepositoryData`. A follow-up to this would expand the use of this pool to the snapshot status API and make it run in parallel as well. | 18 May 2021, 12:40:39 UTC |
| 77d756b | Ryan Ernst | 18 May 2021, 12:38:35 UTC | Deprecate shared and index data path settings (#73178) This commit adds deprecation warnings for use of the path.shared_data setting as well as the index setting index.data_path. relates #73168 | 18 May 2021, 12:38:35 UTC |
| 6dd2a2a | Armin Braun | 18 May 2021, 12:33:11 UTC | Deserialize BlobStore Metadata Files in a Streaming Manner (#73149) We were reading the full file contents up-front here because of the complexity of verifying the footer otherwise. This commit moves the logic for reading metadata blobs (that can become quite sizable in some cases) in a streaming manner by manually doing the footer verification as Lucene's utility methods don't allow for verification on top of a stream. | 18 May 2021, 12:33:11 UTC |
| d5bf72a | Przemyslaw Gomulka | 18 May 2021, 11:07:13 UTC | [Rest Api Compatibility] Typed endpoints for RestUpdateAction and RestDeleteAction (#73115) the previously removed typed enpotins for Update and Delete are retrofitted in this commit the commit that removed them https://github.com/elastic/elasticsearch/pull/47671 relates main meta issue #51816 relates types removal issue #54160 | 18 May 2021, 11:07:13 UTC |
| 963c226 | Martijn van Groningen | 18 May 2021, 10:20:37 UTC | Adjust get alias api with aliases pointing to data streams (#73140) Change the get alias api to not return a 404 when filtering by alias name that refers to data streams. Originated from #72953 Relates to #66163 | 18 May 2021, 10:20:37 UTC |
| f041a02 | Martijn van Groningen | 18 May 2021, 10:09:39 UTC | Take data stream aliases into account with snapshot and restore (#72970) Data stream aliases are stored separately from the data streams in the cluster state. Currently snapshot/restore only takes data streams into account during snapshotting and restoring, this change changes snapshot/restore to also capture and restore data stream aliases. Which alias instances to use depends on the actual data streams that are included in a snapshot or restored from a snapshot. Relates to #66163 | 18 May 2021, 10:09:39 UTC |
| 809af34 | Philip Krauss | 18 May 2021, 09:05:31 UTC | Incorrect Parameter value in watcher.stats tests (#73006) * correct param value * remove metric param * Change the expected warning Co-authored-by: Joe Gallo <joegallo@gmail.com> | 18 May 2021, 09:05:31 UTC |
| d748852 | Michael Bischoff | 18 May 2021, 07:04:48 UTC | Dissect parsing: An `%` occurring in the delimiter causes incorrect capture of the `${key}` that follows (#72876) * Extending parser Regex Patterns to take into account % appearing in the input. * Adding example failure from the ER to the tests | 18 May 2021, 07:04:48 UTC |
| 009cbb8 | Yang Wang | 18 May 2021, 01:57:30 UTC | Service Accounts - enforcing token secret min length at authc time (#72519) The secret value of a service account token generated using either the CLI or API is a time-based UUID of length 22 which provides sufficient entropy. But file-based service account tokens can be created with external tools. It is therefore possible that a token is created with too short secret value. Since there is no way to detect it at the token creation or load (from the file) time, this PR adds a check at authentication time to reject such tokens, i.e. it returns an error if the service token (decoded from the bearer string) has a secret value with length less than 10. | 18 May 2021, 01:57:30 UTC |
| 7ed32f6 | Joe Gallo | 17 May 2021, 23:42:12 UTC | Searchable snapshots mounted in the 'hot' phase should be pinned to hot nodes (#72696) | 17 May 2021, 23:42:12 UTC |
| 6c29374 | Nhat Nguyen | 17 May 2021, 23:17:27 UTC | Adjust Lucene version after upgraded to 8.9 on 7.14 (#73182) Relates #73130 | 17 May 2021, 23:17:27 UTC |
| e36adb3 | Nhat Nguyen | 17 May 2021, 22:09:41 UTC | Disable BWC to upgrade Lucene on 7.x | 17 May 2021, 22:09:41 UTC |
| eeb1098 | James Rodewig | 17 May 2021, 16:18:50 UTC | [DOCS] Use query parameters in search API example (#73158) The current search API documentation doesn't include any examples of query parameter usage. This updates the docs to include a simple syntax example using the `from` and `size` query parameters. | 17 May 2021, 16:18:50 UTC |
| e9add28 | James Rodewig | 17 May 2021, 15:39:12 UTC | [DOCS] Update anchor and xrefs for `alias` glossary entry (#73169) | 17 May 2021, 15:39:12 UTC |
| 25421df | James Rodewig | 17 May 2021, 15:08:07 UTC | [DOCS] Add `alias` to glossary (#73065) https://github.com/elastic/elasticsearch/pull/72613 adds data stream support to aliases. This adds an `alias` glossary entry and removes out the current `index alias` entry. | 17 May 2021, 15:08:07 UTC |
| 6f46ae4 | Greg Back | 17 May 2021, 15:05:03 UTC | Fix comment about number of roles (#73163) This comment is out of date since #71710. | 17 May 2021, 15:05:03 UTC |
| dbc37f9 | markharwood | 17 May 2021, 14:54:53 UTC | TermsEnum api - allow null search strings (#73144) Allow null search strings (matches all) Closes #73141 | 17 May 2021, 14:54:53 UTC |
| 7b64cae | James Rodewig | 17 May 2021, 14:02:08 UTC | [DOCS] Clarify when SLM deletes expired snapshots (#73155) Changes: * Removes an error in the create SLM policy API's `schedule` parameter def. `schedule` is not used to delete expired snapshots. * Updates the `expire_after` parameter def to mention the `slm.retention_schedule` cluster setting. | 17 May 2021, 14:02:08 UTC |
| d9b39d3 | James Rodewig | 17 May 2021, 13:49:41 UTC | [DOCS] Fix typo (#73137) (#73150) Co-authored-by: Ernesto Reig <ernesto.reig@gmail.com> | 17 May 2021, 13:49:41 UTC |
| 1764e8b | Nhat Nguyen | 17 May 2021, 13:37:20 UTC | Upgrade to Lucene-8.9.0-SNAPSHOT-efdc43fee18 (#73130) Upgrades to Lucene-8.9 snapshot which includes: - LUCENE-9507: Custom order for leaves (/cc @mayya-sharipova) - LUCENE-9935: Enable bulk merge for stored fields with index sort | 17 May 2021, 13:37:20 UTC |
| ff406a8 | Seth Michael Larson | 17 May 2021, 13:08:06 UTC | Rename 'termsenum' API to 'terms_enum' for better readability | 17 May 2021, 13:08:06 UTC |
| 4268c76 | David Turner | 17 May 2021, 12:57:49 UTC | Clean up FsHealthService after MDP removal (#73136) Following #72432 we now no longer need a `Set` of unhealthy paths, we can just track the individual path directly. | 17 May 2021, 12:57:49 UTC |
| 9061e59 | David Turner | 17 May 2021, 12:56:39 UTC | Complete backport of #72931 (#73143) Adjusts wire-compat and re-enables BWC tests | 17 May 2021, 12:56:39 UTC |
| a9229ab | Yang Wang | 17 May 2021, 11:17:01 UTC | [Test] Service account tokens should work when TokenService is disabled (#72518) Add a test to ensure that service account tokens always work with basic license for which oauth2 token service is disabled. | 17 May 2021, 11:17:01 UTC |
| bf23f76 | David Turner | 17 May 2021, 10:30:19 UTC | Disable BWC tests for backport of #72931 | 17 May 2021, 10:30:19 UTC |
| b2956b3 | David Turner | 17 May 2021, 10:02:50 UTC | Identify cancelled tasks in list tasks API (#72931) This commit adds a `cancelled` flag to each cancellable task in the response to the list tasks API, allowing users to see that a task has been properly cancelled and will complete as soon as possible. Closes #72907 | 17 May 2021, 10:02:50 UTC |
| eabe2d1 | David Turner | 17 May 2021, 09:52:18 UTC | Increase PeerFinder verbosity on persistent failure (#73128) If a node is partitioned away from the rest of the cluster then the `ClusterFormationFailureHelper` periodically reports that it cannot discover the expected collection of nodes, but does not indicate why. To prove it's a connectivity problem, users must today restart the node with `DEBUG` logging on `org.elasticsearch.discovery.PeerFinder` to see further details. With this commit we log messages at `WARN` level if the node remains disconnected for longer than a configurable timeout, which defaults to 5 minutes. Relates #72968 | 17 May 2021, 09:52:18 UTC |
| 09d1b97 | Przemyslaw Gomulka | 17 May 2021, 07:11:53 UTC | Fix deprecation logs throttling for deprecated routes (#73051) So far when a deprecated route was executed it only emitted deprecation warning once. All subsequent deprecated routes (even when path and method were different) were throttled because the key was the same - deprecated_route This commit suffixes the deprecation key with path and method. closes #73002 | 17 May 2021, 07:11:53 UTC |
| 5a9e655 | Rene Groeschke | 14 May 2021, 20:56:54 UTC | Update Gradle wrapper to 7.0.2 (#73120) Keep Gradle wrapper up to date | 14 May 2021, 20:56:54 UTC |
| 5770107 | Rory Hunter | 14 May 2021, 19:40:27 UTC | Allow container restarts with file logging (#73101) Closes #72702. It wasn't possible to restart an Elasticsearch Docker container when using `ES_LOG_STYLE=file`, and now it is. | 14 May 2021, 19:40:27 UTC |
| e394250 | Jack Conradson | 14 May 2021, 17:50:22 UTC | Add LinkageError to the errors we catch as part of the Painless sandbox (#73116) This changes the Painless sandbox to be more encompassing of possible compiler bugs including JVM bugs. This prevents any single script from crashing a node under a wider array of circumstances that in theory should be recoverable with possible changes to a user-defined script. | 14 May 2021, 17:50:22 UTC |
| c9223a2 | James Rodewig | 14 May 2021, 16:35:25 UTC | [DOCS] Update anchor for alias write index (#73108) Updates the anchor for our alias write index content. Changes are no-op, but this helps prepare for future work on alias docs. | 14 May 2021, 16:35:25 UTC |
| 34d5f6c | James Rodewig | 14 May 2021, 16:27:19 UTC | [DOCS] Consolidate `routing` parameter definitions (#73107) Combines duplicate definitions for the `routing` parameter. | 14 May 2021, 16:27:19 UTC |
| fa60dfd | James Rodewig | 14 May 2021, 15:53:34 UTC | [DOCS] Update single index APIs reference (#73103) Many single index APIs, such as the index document API, can now target data streams. This updates updates references to single index APIs. | 14 May 2021, 15:53:34 UTC |
| 8360a53 | Luca Cavanna | 14 May 2021, 15:14:15 UTC | Consolidate parsing runtime fields from source (#72920) Every runtime field type allows users to omit its script, in which case the field values will be loaded at runtime from _source. This is implemented by having each field type expose a parse from source script factory that extracts the values and converts them to the appropriate type that can then be emitted. The extraction logic from source is though always the same, what changes between the different types is the factory type that is needed and how the object values are converted to their appropriate type. This commit moves the common bits to AbstractFieldScript. Especially the conversion from Object to the appropriate type is handy in a specific method as it will be reused to emit multiple fields from a single script. | 14 May 2021, 15:14:15 UTC |
| 34ab0c0 | Joe Gallo | 14 May 2021, 14:40:28 UTC | [REST Compatible API] transformations for keys in do (#72823) | 14 May 2021, 14:40:28 UTC |
| c3fb60d | Tim Vernum | 14 May 2021, 14:02:40 UTC | Don't run Node Enrollment Tests on FIPS (#73078) The Node Enrollment API requires the use of a KeyStore, but neither PKCS#12 nor JKS keystores are supported when running in FIPS mode. For this reason, the Enrollment APIs are not supported on FIPS mode, and we therefore shouldn't run the tests under FIPS either. Resolves: #73012 | 14 May 2021, 14:02:40 UTC |
| ebb113a | markharwood | 14 May 2021, 14:00:21 UTC | Add "search_after" support to new termsEnum api (#72933) Adds an optional parameter to the _terms_enum request designed to allow paging. The last term from a previous result can be passed as the search_after parameter to a subsequent request, meaning only terms after the given term (but still matching the provided string prefix) are returned Relates to #72910 | 14 May 2021, 14:00:21 UTC |
| a5a20ae | Marios Trivyzas | 14 May 2021, 13:00:48 UTC | SQL: Disallow non-collapsable subselects with ORDER BY (#72991) Ordering an already pre-ordered and limited subselect is not allowed, as such queries cannot be collapsed and translated into query DSL, but the require an extra ordering step on top of the results returned internally by the search/agg query. Fixes: #71158 | 14 May 2021, 13:00:48 UTC |
| 891642b | sasukerui | 14 May 2021, 12:36:06 UTC | [DOCS] Fix typo (#72936) Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com> | 14 May 2021, 12:36:06 UTC |
| acd7e5e | sasukerui | 14 May 2021, 12:21:29 UTC | [DOCS] Fix typo (#72930) Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com> | 14 May 2021, 12:21:29 UTC |
| b564650 | James Rodewig | 14 May 2021, 11:41:11 UTC | [DOCS] Rename `Bulk index alias` API to `Aliases` API (#73077) In 7.14, the API supports both index and data stream aliases. | 14 May 2021, 11:41:11 UTC |
| 955d4f5 | David Roberts | 14 May 2021, 06:04:55 UTC | [ML] Make ml_standard tokenizer create single tokens for email addresses (#73052) The ml_classic tokenizer creates two (or more) tokens for email addresses, at minimum splitting on the @ symbol. This change makes the new ml_standard tokenizer preserve email addresses as a single token. Tokens that contain an @ symbol but are otherwise purely numeric are ignored as though they were just numbers. Additionally @ symbols are ignored at the beginning and end of tokens. | 14 May 2021, 06:04:55 UTC |
| d6e9d18 | Mark Vieira | 13 May 2021, 19:08:44 UTC | Revert "Reduce concurrent workers on Arm CI builds" This reverts commit 998bb148 | 13 May 2021, 19:08:52 UTC |
| 1bb91c7 | Mark Vieira | 13 May 2021, 17:55:01 UTC | Bump bundled JDK to 16.0.1 (#73057) | 13 May 2021, 17:55:01 UTC |
| 43417af | Ryan Ernst | 13 May 2021, 16:51:43 UTC | Move security transport implementation into security module (#73013) The security transport currently still exists in x-pack core, even though the rest of security was split out when x-pack was split into modules. It appears this was due to some tests indirectly relying (and not needing to) rely on adding the netty4plugin. This commit moves the implementation, and the netty module copy that is made in x-pack, into the security module, which is what actually registers the transport implementation. | 13 May 2021, 16:51:43 UTC |
| 05256fa | James Rodewig | 13 May 2021, 15:39:19 UTC | [DOCS] Add missing anchor for script contexts (#73056) | 13 May 2021, 15:39:19 UTC |
| 0bad8a4 | Alan Woodward | 13 May 2021, 15:07:03 UTC | Fix edge case in ExistsQueryBuilderTests for object exists queries (#73037) After #72674 the logic for building object exists queries changed subtly, and the random tests in ExistsQueryBuilderTests can occasionally fail when presented with an exists query for an object. This commit reworks the test to expect the correct query. Fixes #73045 | 13 May 2021, 15:07:03 UTC |
| 5c6dae6 | Jack Conradson | 13 May 2021, 14:49:29 UTC | [DOCS] Add runtime fields contexts to Painless execute API docs (#72131) * add runtime fields contexts to execute docs * Changes for formatting throughout * Add missing context and context_setup * Updating runtime field context * Moving parameters and adopting a more standard API layout * Update several examples * Update more examples for runtime context * Fix links * Add boolean_field example and remove extraneous headings * Add example for date_time context * Remove extra space in TEST * Updating date_time example * Incorporating review feedback * Adding cross links * Tweaking some language based on feedback Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com> Co-authored-by: Adam Locke <adam.locke@elastic.co> | 13 May 2021, 14:49:29 UTC |
| 464dc36 | Marios Trivyzas | 13 May 2021, 14:40:56 UTC | SQL: Add Verification for HAVING on TopHits with subquery (#72967) Previously, when a TopHits aggregation function was used (FIRST/LAST, or MIN/MAX on keyword field) in a subquery and on an outer query this aggregation was filtered (with WHERE/HAVING) the Verification was passed successfully and the query was planned and translated resulting into an unsupported query DSL - since bucket selector on a TopHits agg is not currently supported, and the user received a weird error msg. Verify this case of subselects with TopHits aggs and throw an appropriate error message to the user. Closes: #71441 | 13 May 2021, 14:40:56 UTC |
| 8ec893a | James Rodewig | 13 May 2021, 13:32:36 UTC | [DOCS] Change field alias anchor (#73043) | 13 May 2021, 13:32:36 UTC |
| 81c666e | James Rodewig | 13 May 2021, 12:36:14 UTC | [DOCS] Fix typo in SLM docs (#73025) (#73040) Co-authored-by: Daisuke Harada <1519063+dharada@users.noreply.github.com> | 13 May 2021, 12:36:14 UTC |
| 576fe75 | Yang Wang | 13 May 2021, 11:55:42 UTC | Service Accounts - add beta version documentation (#71729) * wip * Service Accounts - add beta documentation * consistent names * fix test * Update service accounts overview and token creation files. * Rename get service tokens to get service credentials * fix tests * Changes for create and get service tokens. * Changes for get token creds, delete token, clear token cache, and token auth. * add manage_service_account privilege to list * List service accounts APIs * Move xpack setting to Security API page, plus other cleanup. * Shorten secret tokens in examples, add cross links, plus other cleanup. * Clarifying parameter descriptions. * Clarify language for authenticating with a token. * Tweaks * Typo fix * Adding redirects to work around CI build checks * Revert "Adding redirects to work around CI build checks" This reverts commit 20a1b53591fee9cf80d352818572f47f8c637cd7. * Remove redirects that were implemented to satisfy CI checks in master branch * Move note about not supporting basic auth * Clarify what service accounts are specifically for * Apply suggestions from code review Co-authored-by: Tim Vernum <tim@adjective.org> * Addressing review feedback * tweak * Improve doc tests * fix test Co-authored-by: Adam Locke <adam.locke@elastic.co> Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com> Co-authored-by: Tim Vernum <tim@adjective.org> | 13 May 2021, 11:55:42 UTC |
| 3bd594e | Alan Woodward | 13 May 2021, 10:35:23 UTC | Replace simpleMatchToFullName (#72674) MappingLookup has a method simpleMatchToFieldName that attempts to return all field names that match a given pattern; if no patterns match, then it returns a single-valued collection containing just the pattern that was originally passed in. This is a fairly confusing semantic. This PR replaces simpleMatchToFullName with two new methods: * getMatchingFieldNames(), which returns a set of all mapped field names that match a pattern. Calling getFieldType() with a name returned by this method is guaranteed to return a non-null MappedFieldType * getMatchingFieldTypes, that returns a collection of all MappedFieldTypes in a mapping that match the passed-in pattern. This allows us to clean up several call-sites because we know that MappedFieldTypes returned from these calls will never be null. It also simplifies object field exists query construction. | 13 May 2021, 10:35:23 UTC |
| a5e39ce | Felix Barnsteiner | 13 May 2021, 06:34:34 UTC | Add missing word in data streams checkpoints docs (#73005) | 13 May 2021, 06:34:34 UTC |
| 2aecba8 | Yang Wang | 13 May 2021, 06:01:33 UTC | Service Accounts - Fix delete token status code (#73021) The delete token response now returns status code 404 instead of 200 when the token does not exist. | 13 May 2021, 06:01:33 UTC |
| 716b06f | Yang Wang | 13 May 2021, 06:00:28 UTC | Service Accounts - rename service account realm (#73024) Add a leading underscore for both realm type and nane since they are synthetic. | 13 May 2021, 06:00:28 UTC |
| 44fc661 | Nhat Nguyen | 12 May 2021, 21:59:25 UTC | Add point in time to HLRC (#72167) Closes #70593 | 12 May 2021, 21:59:25 UTC |
| ce41fd7 | James Rodewig | 12 May 2021, 20:25:39 UTC | [DOCS] Mark `filter` param as optional | 12 May 2021, 20:25:39 UTC |
| 85ed910 | Przemyslaw Gomulka | 12 May 2021, 19:00:19 UTC | [Rest Api Compatibility] Typed endpoints for search and related endpoints (#72155) Implements a V7 compatible typed endpoints for REST for search related apis retrofits the REST layer change removed in #41640 relates main meta issue #51816 relates types removal issue #54160 | 12 May 2021, 19:00:19 UTC |
| b3a1977 | James Rodewig | 12 May 2021, 18:26:22 UTC | [DOCS] Update allowed operations on data stream write index (#72994) With #70908, you can now close a write index in 7.12.1 and later versions. This removes an outdated reference from the data stream overview docs. | 12 May 2021, 18:26:22 UTC |
| 7909d3b | James Rodewig | 12 May 2021, 16:57:56 UTC | [DOCS] Fix JSON spec link for terms enum API (#72996) | 12 May 2021, 16:57:56 UTC |
| 8dddca7 | James Rodewig | 12 May 2021, 16:54:20 UTC | [DOCS] Remove and redirect frozen index overview content (#72990) Changes: * Removes and adds redirects for the frozen indices [overview][0], [best practices][1], [search][2], and [monitoring][3] pages. * Removes glossary terms related to frozen indices. * Updates several xrefs to point to the freeze index API docs. Relates to elastic/elasticsearch#72946 and elastic/elasticsearch#70192. [0]: https://www.elastic.co/guide/en/elasticsearch/reference/7.12/frozen-indices.html [1]: https://www.elastic.co/guide/en/elasticsearch/reference/master/best_practices.html [2]: https://www.elastic.co/guide/en/elasticsearch/reference/master/searching_a_frozen_index.html [3]: https://www.elastic.co/guide/en/elasticsearch/reference/master/monitoring_frozen_indices.html | 12 May 2021, 16:54:20 UTC |
| 785ce7e | Adam Locke | 12 May 2021, 16:43:21 UTC | [DOCS] Add missing SSL settings for Metricbeat (#72987) | 12 May 2021, 16:43:21 UTC |
| 0937c3f | Ryan Ernst | 12 May 2021, 16:01:45 UTC | Separate encrypted repository plugin integ tests (#72959) The encrypted repository module supports encryption with existing repository plugins. Currently the integration tests with these plugins pull in all of the plugins into the same test classpath. That can cause problems when the dependencies of those repositories need to differ. This commit splits the tests into separate qa subprojects. | 12 May 2021, 16:01:45 UTC |
