Retry flaky unified tests #1565
Draft
Conversation
This file contains 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
katcharov
requested review from
a team and
vbabanin
and removed request for
a team
katcharov
force-pushed
the
JAVA-5393-unified-retries
branch
from
6282ce7 to
8e8d7a3
Compare
katcharov
force-pushed
the
JAVA-5393-unified-retries
branch
from
8e8d7a3 to
6f3b56d
Compare
katcharov
force-pushed
the
JAVA-5393-unified-retries
branch
from
d24e003 to
6b53427
Compare
katcharov
commented
Dec 5, 2024
Comment on lines
330
to
+331
| public void shouldPassAllOutcomes( | ||
| final String testName, |
There was a problem hiding this comment.
This method contains the try-catch with the new retry logic.
katcharov
force-pushed
the
JAVA-5393-unified-retries
branch
from
86681ad to
3b2c915
Compare
katcharov
commented
Dec 5, 2024
Comment on lines
+43
to
+55
|
|
||
| // TODO reasons for retry | ||
| // Exception in encryption library: ChangeCipherSpec message sequence violation | ||
| def.retry("TODO reason") | ||
| .whenFailureContains("ChangeCipherSpec message sequence violation") | ||
| .test("client-side-encryption", "namedKMS-createDataKey", "create datakey with named KMIP KMS provider"); | ||
|
|
||
| def.retry("TODO reason") | ||
| .whenFailureContains("Number of checked out connections must match expected") | ||
| .test("load-balancers", "cursors are correctly pinned to connections for load-balanced clusters", "pinned connections are returned after a network error during a killCursors request"); | ||
|
|
||
| def.retry("TODO reason") | ||
| .test("client-side-encryption", "namedKMS-rewrapManyDataKey", "rewrap to kmip:name1"); |
There was a problem hiding this comment.
These are flaky failures from evergreen. Reasons should be specified, or these 3 retries can be removed and added in a separate PR, after examining the failures.
katcharov
commented
Dec 6, 2024
| } | ||
| if (!testDef.matchesThrowable(e)) { | ||
| // if the throwable is not matched, test definitions were not intended to apply; rethrow it | ||
| throw e; |
There was a problem hiding this comment.
I tested this by failing an assertion within the try of shouldPassAllOutcomes. When the assertion message matches what is specified in a def, it will ignore, ignore, and then fail with that assertion failure. When it does not match, the first instance will fail, and the remaining will be ignored.
katcharov
force-pushed
the
JAVA-5393-unified-retries
branch
from
ce1918c to
784dc51
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
JAVA-5393 (part of).
See WRITING-28651
This creates N copies (attempts) of the unified tests, and places a try-catch around shouldPassAllOutcomes. If an attempt passes, ensuing attempts are ignored. Failures are ignored, except the last attempt.
This is not intended to fully resolve the issue, but is a low-cost fix that should reduce failing tests.