-
Notifications
You must be signed in to change notification settings - Fork 594
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
Fixes for migrations tests #24073
Fixes for migrations tests #24073
Conversation
otherwise we'll need to be prepared to a situation where we get no successful reply, but the migration is created
…error when waiting for a condvar with timeout
bazel build failure is https://redpandadata.atlassian.net/issues/CORE-8112 |
the below tests from https://buildkite.com/redpanda/redpanda/builds/57834#01930b9c-fc4b-4f48-a4b4-68e8ca969689 have failed and will be retried
|
this one is also known https://redpandadata.atlassian.net/issues/CORE-8093 |
ducktape was retried in https://buildkite.com/redpanda/redpanda/builds/57834#01930bde-74c5-4508-b4bb-f3ce79768742 |
Retry command for Build#57834please wait until all jobs are finished before running the slash command
|
non flaky failures in https://buildkite.com/redpanda/redpanda/builds/57834#01930bf9-31a1-44ca-85c9-373157f3f2a0:
|
…sence Finj may make things lag, so tolerate migration absence, but not wrong data.
056a9cb
to
eafe51f
Compare
} catch (const ss::timed_out_error& e) { | ||
} catch (const ss::condition_variable_timed_out& e) { |
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.
too bad they didn't inherit from timed_out_error :/
Tests/migrations: relax migrations tests: do not check for number of migrations when using high level API under failure injector.
Eventual consistency of migrations table may come late, and we'd rather not wait for it to test but move on with other actions to see how a disturbed cluster copes with active use.
Raft/consensus: catch
ss::condition_variable_timed_out
notss::timed_out_error
when waiting for a condvar with timeout. Otherwise it bubbles up unhandled appearing in logs and potentially breaking the logic too.Tests/migrations: run mount/unmount commands without finjector. Otherwise we'd need to be prepared to a situation where we get no successful reply from admin API, but the migration is nevertheless created (node killed right before it has sent a reply). When testing low level API we handle this by checking the migration is present. But high-level mount/unmount commands auto-remove migration objects on completion. Telling apart an uncreated and a completed migration in the test logic would be somewhat tricky.
Backports Required
Release Notes