-
Notifications
You must be signed in to change notification settings - Fork 3.8k
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
ccl/multiregionccl: TestMrSystemDatabase failed #98039
Comments
@jeffswenson can you put this on your radar? |
This test is flaky. See cockroachdb#98039. Epic: None Release note: None
Adding the GA-blocker label now that the test will be skipped once #98533 merges. |
98528: logictest: use a retry loop when starting testserver r=rafiss a=rafiss A similar change was already made for upgrading the node. informs #94956 Release note: None 98533: ccl/multiregion: skip TestMrSystemDatabase r=mgartner a=mgartner This test is flaky. See #98039. Epic: None Release note: None Co-authored-by: Rafi Shamim <[email protected]> Co-authored-by: Marcus Gartner <[email protected]>
I'm removing the GA-blocker label from this issue. I think there is a bug here that needs to be fixed, but it only impacts multi-region Serverless. |
I don't know that there's a bug here anymore. I've stressed the crap out of this with roachprod stress on 160 cores and can't get it to fail. I'll re-enable it. |
99858: screl: Add IndexID as a attr of UniqueWithoutIndex element r=Xiang-Gu a=Xiang-Gu Previously, ALTER TABLE stmt where we add column/drop column/alter PK and adding a unique without index is problematic in that the it can succeed even when there are duplicate values. We already had a dep rule that enforces the new primary index to be backfilled before we validate the constraint against it. Unfortunately, this rule is not enforced on unique without index constraint because IndexID was not a attr of it. This commit fixes this. Fixes #99281 Epic: None Release note (bug fix): Fixed a bug in v23.1 in the declarative schema changer where unique without index can be incorrectly added in tables with duplicate values if it's with a ADD/DROP COLUMN in one ALTER TABLE statement. 100357: sql: allow changing the number of histogram samples and buckets r=rytaft a=rytaft Informs #72418 Informs #97701 Release note (sql change): Added two new cluster settings that enable users to change the number of histogram samples and buckets collected when building histograms as part of table statistics collection: `sql.stats.histogram_samples.count` and `sql.stats.histogram_buckets.count`. While the default values should work for most cases, it may be beneficial to increase the number of samples and buckets for very large tables to avoid creating a histogram that misses important values. 100489: go.mod: bump Pebble to b84a7ec7d8dc r=RaduBerinde a=jbowens ``` b84a7ec7 db: populate return statistics for flushable ingests 5fd58365 objstorage: implement tracing 7f7451f2 db,record: add BatchCommitStats to measure total and component durations for commit 295aaab0 objstorage: implement basic refcounting ``` Epic: None Release note: None 100516: multiregionccl: reenable TestMrSystemDatabase r=ajwerner a=ajwerner I stressed this for a long time on many cores and it did not fail. Epic: none Fixes: #98039 Release note: None 100527: roachtest: skip multitenant/distsql for now r=yuzefovich a=yuzefovich Informs: #100260. Epic: None Release note: None Co-authored-by: Xiang Gu <[email protected]> Co-authored-by: Rebecca Taft <[email protected]> Co-authored-by: Jackson Owens <[email protected]> Co-authored-by: Andrew Werner <[email protected]> Co-authored-by: Yahor Yuzefovich <[email protected]>
I also ran this heavily under stress on the same commit from the original flake. I was unable to reproduce it. I'll keep an eye out for reflakes. Maybe there is some random test parameter that was required to trigger the flake. |
ccl/multiregionccl.TestMrSystemDatabase failed with artifacts on master @ cf14ad694ee562676f53e36fa8495206c3aed61f:
Parameters:
TAGS=bazel,gss
Help
See also: How To Investigate a Go Test Failure (internal)
This test on roachdash | Improve this report!
Jira issue: CRDB-25037
The text was updated successfully, but these errors were encountered: