Test: "ack committed prepare": Remove extra t.run() workaround #1366
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
As part of the Scan API code review, we found that
Groove.prefetch()
would invoke its callback synchronously if all of the objects could be prefetched from cache. That was fixed as part of the same PR.But fixing it caused an unrelated test to fail:
replica_test.zig
's"Cluster: repair: ack committed prepare"
.During the
Change views. B1/B2 participate. Don't allow B2 to repair op=21.
step, even though only B1/B2 participated in the view change, A0 was still allowed to send SVC messages. Deferringprefetch()
's callback to next tick made commits take slightly longer. This permutation meant that B1/B2 would finish their view change (as before), but then A0 would prod them into kicking off another view change. So after thet.run()
, B1/B2 were in status=view_change instead of status=normal.(Also add some additional assertions that were useful while troubleshooting.)