In WT-7788 recent activity suggests a change to checkpoint cleanup. MongoDB sometimes uses use_timestamp=false on closing the connection and sometimes that takes a very long time. strace results indicate a huge read load.
We should avoid checkpoint cleanup anytime we're shutting down, not just when timestamps are in use. Possibly consider whether this should be a configuration setting or just an internal action.
- related to
-
WT-7788 Ideas on improving checkpoint cleanup
- Backlog