-
Notifications
You must be signed in to change notification settings - Fork 25k
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
Upgrade to lucene 8.0.0-snapshot-ff9509a8df #39350
Conversation
Pinging @elastic/es-search |
Pinging @elastic/es-distributed |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Now that FilterTermsEnum#seekExact(BytesRef, TermState)
delegates to the wrapper instance rather than to FilterTermsEnum.seekExact(BytesRef)
, I think it'd be safer if we checked whether the term is accepted in FieldNamesTermsEnum#seekExact(BytesRef, TermState)
.
Looks like there might be a BWC issue in the Lucene80 Codec for norms:
|
OK, BWC tests fail here because it's targetting 7.0.0, which currently is on a previous lucene snapshot without jump tables, so the indexes are not readable across both versions. Should I disable BWC tests while this PR is merged and backported? |
…a non-filtered reader
I updated to throw an |
Failure in ci/2 is due to #32299 @elasticmachine run elasticsearch-ci/2 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the ISE, LGTM.
Packaging tests were still running with BWC enabled @elasticmachine test elasticsearch-ci/packaging-sample |
@elasticmachine run elasticsearch-ci/packaging-sample |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I left a question.
boolean bwc_tests_enabled = true | ||
final String bwc_tests_disabled_issue = "" /* place a PR link here when committing bwc changes */ | ||
boolean bwc_tests_enabled = false | ||
final String bwc_tests_disabled_issue = "https://github.com/elastic/elasticsearch/pull/39350" /* place a PR link here when committing bwc changes */ |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What is the reason for disabling all BWC tests with this PR?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
BWC tests are currently being run against the 7.0.0 branch, which is using a previous lucene 8.0.0 snapshot. There has been a change in the Codec format between these two snapshots, which means that an elasticsearch running with this change cannot read lucene indexes built with the previous snapshot; upgrade tests between the current 7.0.0 branch head and this PR fail because the indexes cannot be opened.
If there's a finer-grained BWC suspension I should be doing instead, please let me know!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This also means that people using 7.0-alpha and 7.0-beta won't be able to do a rolling upgrade to 7.0-GA, because they're using unreleased lucene snapshots that do not contain this Codec change.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This relates to #38919. The issue here being that that change is still under discussion. Let me see if we can accelerate it.
This also means that people using 7.0-alpha and 7.0-beta won't be able to do a rolling upgrade to 7.0-GA, because they're using unreleased lucene snapshots that do not contain this Codec change.
That's okay, we do not support that to begin with. 😇
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I merged and back-ported #38919 . The discussion is still ongoing, but we should not hold anything up for it.
It looks as though I need to separately disable BWC on the packaging tests? |
Side note, when we release a version, it would be helpful to think about updating this page? Not sure if this should be done before the release though as for example 7.0 branch has been yet released with |
Backport of #39350 Contains the following: * LUCENE-8635: Move terms dictionary off-heap for non-primary-key fields in `MMapDirectory` * LUCENE-8292: `TermsEnum` is fully abstract * LUCENE-8679: Return WITHIN in `EdgeTree#relateTriangle` only when polygon and triangle share one edge * LUCENE-8676: Nori tokenizer deals correctly with large buffers * LUCENE-8697: `GraphTokenStreamFiniteStrings` better handles side paths with gaps * LUCENE-8664: Add `equals` and `hashCode` to `TotalHits` * LUCENE-8660: `TopDocsCollector` returns accurate hit counts if the total equals the threshold * LUCENE-8654: `Polygon2D#relateTriangle` fix for when the polygon is inside the triangle * LUCENE-8645: `Intervals#fixField` can merge intervals from different fields * LUCENE-8585: Create jump-tables for DocValues at index time
Backport of #39350 Contains the following: * LUCENE-8635: Move terms dictionary off-heap for non-primary-key fields in `MMapDirectory` * LUCENE-8292: `TermsEnum` is fully abstract * LUCENE-8679: Return WITHIN in `EdgeTree#relateTriangle` only when polygon and triangle share one edge * LUCENE-8676: Nori tokenizer deals correctly with large buffers * LUCENE-8697: `GraphTokenStreamFiniteStrings` better handles side paths with gaps * LUCENE-8664: Add `equals` and `hashCode` to `TotalHits` * LUCENE-8660: `TopDocsCollector` returns accurate hit counts if the total equals the threshold * LUCENE-8654: `Polygon2D#relateTriangle` fix for when the polygon is inside the triangle * LUCENE-8645: `Intervals#fixField` can merge intervals from different fields * LUCENE-8585: Create jump-tables for DocValues at index time
* master: Upgrade to lucene 8.0.0-snapshot-ff9509a8df (elastic#39350)
Contains the following:
MMapDirectory
TermsEnum
is fully abstractEdgeTree#relateTriangle
only when polygon and triangle share one edgeGraphTokenStreamFiniteStrings
better handles side paths with gapsequals
andhashCode
toTotalHits
TopDocsCollector
returns accurate hit counts if the total equals the thresholdPolygon2D#relateTriangle
fix for when the polygon is inside the triangleIntervals#fixField
can merge intervals from different fields