-
Notifications
You must be signed in to change notification settings - Fork 3.7k
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
roachtest: schemachange/invertedindex failed #91977
Comments
This had a weird stall for more than 11s in the workload which is not obviously attributable, and then the workload gets an error about a nonsensical deadline and exits. The deadline error propagation is somewhat tracked in #76727. |
roachtest.schemachange/invertedindex failed with artifacts on master @ 6723e00a46aaa3ea575093bd82a02b7d6f6b131b:
Parameters: |
roachtest.schemachange/invertedindex failed with artifacts on master @ 286b3e235171a39b8f9910555affcc7ce310741a:
Parameters: |
roachtest.schemachange/invertedindex failed with artifacts on master @ e028ce5b14505dfd17ef8b13001c0ab8ac811e3c:
Parameters: |
roachtest.schemachange/invertedindex failed with artifacts on master @ 0d3393b0623a5c258b25725f64f3689e2f54667b:
Parameters: |
roachtest.schemachange/invertedindex failed with artifacts on master @ 39c06b5a438c01c93ffbfeeefe702d3f9b620eaf:
Parameters: |
roachtest.schemachange/invertedindex failed with artifacts on master @ 21786aa112e6b822858f281c1cc59608987c5c0a:
Parameters: |
@cockroachdb/sql-queries, I'm tagging y'all because I think this has something to do with query execution. The workload which is doing a single json workload, seems to be constructing bad batch requests. |
roachtest.schemachange/invertedindex failed with artifacts on master @ cf14ad694ee562676f53e36fa8495206c3aed61f:
Parameters: |
I've been unable to reproduce this so far. |
Stack trace:
|
|
Cool this is great. I think we can take it from here. |
Do you have the rest of the artifacts from the failing run? |
I think what's going on here is that the merge phase of the backfill has some incorrect logic to determine whether it is done. It'll be a bit icky to test this. |
@msirek how many iterations did it take for this to flake? |
It's too large for github. Here's the first one that failed:
|
#98262 is a speculative fix. Thanks for doing the repro exercise. Much appreciated. |
I tried running the test against #98262 and it still fails. |
98262: sql/backfill: consult the ResumeSpan when scanning in index merge r=ajwerner a=ajwerner The existing logic could construct an invalid batch request which would result in spurious errors. Fixes #91977 Release note (bug fix): Fixed a bug which rarely could result in some CREATE INDEX statements to fail with an error `failed to verify keys for Scan`. Co-authored-by: ajwerner <awerner32@gmail.com>
roachtest.schemachange/invertedindex failed with artifacts on master @ d98d195dfa7a3480ce3e07657aa870692fe71cea:
Parameters:
ROACHTEST_cloud=gce
,ROACHTEST_cpu=4
,ROACHTEST_encrypted=false
,ROACHTEST_ssd=0
Help
See: roachtest README
See: How To Investigate (internal)
This test on roachdash | Improve this report!
Jira issue: CRDB-21504
The text was updated successfully, but these errors were encountered: