Josh McKenzie created CASSANDRA-17830: -----------------------------------------
Summary: Read Unavailables due to GossipStage backlog with high number of pending tasks Key: CASSANDRA-17830 URL: https://issues.apache.org/jira/browse/CASSANDRA-17830 Project: Cassandra Issue Type: Improvement Components: Cluster/Gossip Reporter: Josh McKenzie Assignee: Josh McKenzie In CASSANDRA-16759, like we changed from iterating over liveMembers + unreachableMembers to all keys in endpointStateMap [here|https://github.com/apache/cassandra/commit/2fba5c80ce7bf71d04c62043ffa1088b9e832d83#diff-99267a2170b04fd7dd24d6c6bf2ba1fc26d6dc896cd74f8c5bd56c476e2540e4R208]. We can have an endpoint in quarantine (in justRemovedEndpoints) which has no release version information. In this case we don't memoize the value and instead recalculate the min version for every endpoint for every incoming gossip message... which can be bad. -- This message was sent by Atlassian Jira (v8.20.10#820010) --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org