Backport of eval delete: move batching of deletes into RPC handler and state into release/1.4.x #15247
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Outdated suggestions cannot be applied.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.
Backport
This PR is auto-generated from #15117 to be assessed for backporting due to the inclusion of the label backport/1.4.x.
The below text is copied from the body of the original PR.
During unusual outage recovery scenarios on large clusters, a backlog of millions of evaluatons can appear. In these cases, the
eval delete
command can put excessive load on the cluster by listing large sets of evals to extract the IDs and then sending larges batches of IDs. Although the command's batch size was carefully tuned, we still need to be JSON deserialize, reserialize to messagepack, send the log entries through raft, and get the FSM applied.To improve performance of this recovery case, move the batching process into the RPC handler and the state store. The design here is a little weird, so let's look at the failed options first:
A naive solution here would be to just send the filter as the raft request and let the FSM apply delete the whole set in a single operation. Benchmarking with 1M evals on a 3 node cluster demonstrated this can block the FSM apply for several minutes, which puts the cluster at risk if there's a leadership failover (the barrier write can't be made while this apply is in-flight).
A less naive but still bad solution would be to have the RPC handler filter and paginate, and then hand a list of IDs to the existing raft log entry. Benchmarks showed this blocked the FSM apply for 20-30s at a time and took roughly an hour to complete.
Instead, we're filtering and paginating in the RPC handler to find a page token, and then passing both the filter and page token in the raft log. The FSM apply recreates the paginator using the filter and page token to get roughly the same page of evaluations, which it then deletes. The pagination process is fairly cheap (only about 5% of the total FSM apply time), so counter-intuitively this rework of the pagination ends up being much faster. A benchmark of 1M evaluations showed this blocked the FSM apply for 20-30ms at a time (typical for normal operations) and completes in less than 4 minutes.
Note that, as with the existing design, this delete is not consistent: a new evaluation inserted "behind" the cursor of the pagination will fail to be deleted.