[docdb] High transactions_running metric after local bootstrap #3823
Assignees
Labels
area/docdb
YugabyteDB core features
Comments
spolitov
added a commit
that referenced
this issue
Mar 6, 2020
…e stopped Summary: Before recent work on intents cleanup we could have a lot of intents stuck at transaction participant. But that cleanup does not work if write was stopped. So those transactions could hang there forever. Fixed SafeTimeForFollower to looks also at pending operations. Used it for transaction cleanup instead of last replicated hybrid time. Test Plan: Jenkins Reviewers: mikhail Reviewed By: mikhail Subscribers: ybase, bogdan Differential Revision: https://phabricator.dev.yugabyte.com/D8092
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Was looking at this with @mbautin , this is in the xdc universe. This might be some side effect of the transaction cleanup work that you did previously @spolitov , since this cluster is long running...
The side effect of this was ~3GB of unaccounted for memory. I think there's two issues here:
transactions_container?MemTracker, so we successfully account for their memory usage too?The text was updated successfully, but these errors were encountered: