-
Notifications
You must be signed in to change notification settings - Fork 8.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[data views] index negation doesn't work with cross cluster search #147926
Comments
I've reproduced the problem. I working on looking at the request between Kibana and ES to see where the bug exists. Field caps works as expected, meaning it doesn't exhibit the bug. |
Pinging @elastic/kibana-data-discovery (Team:DataDiscovery) |
Update - I was able to see the problem by turning on elasticsearch query logging in kibana. Kibana was swallowing cross cluster negation index patterns - there's code that checks for negation patterns but it fails to handle the cross cluster ones. I'm now working to determine whether I should specifically handle cross cluster patterns or whether the validation code can be removed entirely. |
## Summary The data views api examines comma delimited sections to see if there are matching indices before fetching the field list. The existing code checked for index pattern negation - patterns that started with a `-`. However, it didn't check for this in cross cluster case - `this_cluster:-kibana*`. The code now handles the cross cluster case appropriately. Still, its unclear to me whether this logic is necessary, hence why I opened #147970 - I was able to resolve the failed tests most familiar to me but I will need to work with engineers from other teams to investigate other failures. Closes #147926
## Summary The data views api examines comma delimited sections to see if there are matching indices before fetching the field list. The existing code checked for index pattern negation - patterns that started with a `-`. However, it didn't check for this in cross cluster case - `this_cluster:-kibana*`. The code now handles the cross cluster case appropriately. Still, its unclear to me whether this logic is necessary, hence why I opened elastic#147970 - I was able to resolve the failed tests most familiar to me but I will need to work with engineers from other teams to investigate other failures. Closes elastic#147926 (cherry picked from commit e3cac21)
… (#148206) # Backport This will backport the following commits from `main` to `8.6`: - [[data views] allow cross cluster index pattern negation (#147968)](#147968) <!--- Backport version: 8.9.7 --> ### Questions ? Please refer to the [Backport tool documentation](https://github.com/sqren/backport) <!--BACKPORT [{"author":{"name":"Matthew Kime","email":"matt@mattki.me"},"sourceCommit":{"committedDate":"2022-12-29T15:51:20Z","message":"[data views] allow cross cluster index pattern negation (#147968)\n\n## Summary\r\n\r\nThe data views api examines comma delimited sections to see if there are\r\nmatching indices before fetching the field list. The existing code\r\nchecked for index pattern negation - patterns that started with a `-`.\r\nHowever, it didn't check for this in cross cluster case -\r\n`this_cluster:-kibana*`. The code now handles the cross cluster case\r\nappropriately.\r\n\r\nStill, its unclear to me whether this logic is necessary, hence why I\r\nopened #147970 - I was able to\r\nresolve the failed tests most familiar to me but I will need to work\r\nwith engineers from other teams to investigate other failures.\r\n\r\nCloses https://github.com/elastic/kibana/issues/147926","sha":"e3cac218f011b3de0b39700feed1500f6129fd99","branchLabelMapping":{"^v8.7.0$":"main","^v(\\d+).(\\d+).\\d+$":"$1.$2"}},"sourcePullRequest":{"labels":["bug","release_note:fix","Feature:Data Views","Team:DataDiscovery","backport:prev-minor","v8.7.0"],"number":147968,"url":"https://github.com/elastic/kibana/pull/147968","mergeCommit":{"message":"[data views] allow cross cluster index pattern negation (#147968)\n\n## Summary\r\n\r\nThe data views api examines comma delimited sections to see if there are\r\nmatching indices before fetching the field list. The existing code\r\nchecked for index pattern negation - patterns that started with a `-`.\r\nHowever, it didn't check for this in cross cluster case -\r\n`this_cluster:-kibana*`. The code now handles the cross cluster case\r\nappropriately.\r\n\r\nStill, its unclear to me whether this logic is necessary, hence why I\r\nopened #147970 - I was able to\r\nresolve the failed tests most familiar to me but I will need to work\r\nwith engineers from other teams to investigate other failures.\r\n\r\nCloses https://github.com/elastic/kibana/issues/147926","sha":"e3cac218f011b3de0b39700feed1500f6129fd99"}},"sourceBranch":"main","suggestedTargetBranches":[],"targetPullRequestStates":[{"branch":"main","label":"v8.7.0","labelRegex":"^v8.7.0$","isSourceBranch":true,"state":"MERGED","url":"https://github.com/elastic/kibana/pull/147968","number":147968,"mergeCommit":{"message":"[data views] allow cross cluster index pattern negation (#147968)\n\n## Summary\r\n\r\nThe data views api examines comma delimited sections to see if there are\r\nmatching indices before fetching the field list. The existing code\r\nchecked for index pattern negation - patterns that started with a `-`.\r\nHowever, it didn't check for this in cross cluster case -\r\n`this_cluster:-kibana*`. The code now handles the cross cluster case\r\nappropriately.\r\n\r\nStill, its unclear to me whether this logic is necessary, hence why I\r\nopened #147970 - I was able to\r\nresolve the failed tests most familiar to me but I will need to work\r\nwith engineers from other teams to investigate other failures.\r\n\r\nCloses https://github.com/elastic/kibana/issues/147926","sha":"e3cac218f011b3de0b39700feed1500f6129fd99"}}]}] BACKPORT--> Co-authored-by: Matthew Kime <matt@mattki.me>
the index pattern
*:remote_test_index_*,*:-remote_test_index_2
will pull in fields fromremote_test_index_2
The text was updated successfully, but these errors were encountered: