Fix missing completion when no buckets are in checkpoint #214
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Outdated suggestions cannot be applied.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.
For clients, a full sync iteration generally looks like this:
checkpointmessage describing the target state of each bucket.datalines are sent for data.checkpoint_completeline is sent and allows clients to make the lines they've received public if it matches the target state from earlier.After the changes to the sync service for bucket priorities and partial completions, we also send a
partial_checkpoint_completeline as part of step 2 after sending all data lines in a specific priority (except for the lowest priority, which results in the traditionalcheckpoint_completeline).These changes introduced a regression observable when performing a sync iteration with no changed buckets (which can happen when only write checkpoints are changed). Since there are no buckets and no priorities, there is no lowest priority for which a
checkpoint_completeline would be sent. This breaks clients, which only consider the sync complete after receiving that line.Fix this by ensuring we always send a
checkpoint_completeafter a finished sync operation.