-
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
Revert "Revert "util/log: more misc cleanups"" #57222
Conversation
This re-instates cockroachdb#57000, as it did not "introduce a race in crdb" (Instead it merely outlined a bug in a test, which is to be fixed in a subsequent commit.) Release note: None
The pgwire code uses goroutines whose execution extends past the server's stopper's `Stop()` call. This is arguably a mis-design which needs to be fixed -- but in the meantime the logging package needs to tolerate it. This patch ensures that asynchronous `log` calls can be performed concurrently with a `TestLogScope`'s `Close()` call. Release note: None
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.
Reviewed 59 of 59 files at r1, 1 of 1 files at r2.
Reviewable status: complete! 1 of 0 LGTMs obtained (waiting on @dt and @irfansharif)
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.
Woops, sorry about the misattribution, just friday evening laziness. LGTM (though I think we want to re-order these commits for git bisectability).
The two commits cannot be inverted - the 2nd commit works around the bug in the test within the context set up by the 1st commit. |
TFYR bors r=irfansharif |
Build succeeded: |
As discussed previously in cockroachdb#57222 - the SQL servers do not obey the server stop logic well and leak goroutines that continue to log while the TestLogScope gets closed. This commit continues to tolerate this misdesign by ensuring that some more stderr-related flags are accessed atomically. Release note: None
Reverts #57178.
This re-instates #57000, as it did not "introduce a race in crdb"
The description in #57178 was incorrect - instead #57161 / #57162 merely outlined a bug in some tests, which remains to be fixed.
This PR also includes a temporary workaround for said test bug.
Fixes #57162.
Fixes #57161 (presumably - although the symptoms there don't align).