-
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
*: de-flake TestPaginatedBackupTenant and remove data race #74926
*: de-flake TestPaginatedBackupTenant and remove data race #74926
Conversation
This test readily fails under stress with span configs enabled (cockroachdb#73876). Now that we split within the tenant keyspace for tenant tables, the export requests generated for each BACKUP is liable to be retried if concurrent with range splits. The splits causes KV to error out with RangeKeyMismatchErrors, at which point the request is internally retried. The test, which wants to assert on what ExportSpan requests are intercepted, previously did not need to consider these retries/need to de-dup -- it now does. Release note: None
This test got a lot slower under the span configs infrastructure, something we're investigating as part of cockroachdb#74919. Until then, prevent this test from spoiling builds. Release note: None
Fixes cockroachdb#74932; we were reading from the replicas map without having acquired the prerequisite lock. Release note: None
3aaf50b
to
3218353
Compare
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.
The "sqush race" commit needs to be backed out.
bors r+ |
Build succeeded: |
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 3 of 3 files at r4, 1 of 1 files at r5, 1 of 1 files at r6, all commit messages.
Reviewable status: complete! 0 of 0 LGTMs obtained (waiting on @adityamaru, @ajwerner, @irfansharif, @nvanbenschoten, and @tbg)
See individual commits; this PR squashes a data race in KV among other things.
This test readily fails under stress with span configs enabled (#73876).
Now that we split within the tenant keyspace for tenant tables, the
export requests generated for each BACKUP is liable to be retried if
concurrent with range splits. The splits causes KV to error out with
RangeKeyMismatchErrors, at which point the request is internally
retried. The test, which wants to assert on what ExportSpan requests are
intercepted, previously did not need to consider these retries/need to
de-dup -- it now does.
Release note: None