josh-mckenzie commented on pull request #1213: URL: https://github.com/apache/cassandra/pull/1213#issuecomment-953855890
> Are you proposing to refresh the cache on a single server (where denylisting JMX call comes to) and then just let other servers to catch up during "denylist_refresh_seconds"? Essentially yes. An operator would run the JMX command on a single node and, if they needed _immediate_ action across the cluster, run the loadPartitionDenylist call across all nodes so they'd pick it up. Otherwise they'd hit the one on a particular node that was underwater, let it recover, and let the rest of the cluster pick it up in the refresh period. It's functionally no different than the previous paradigm, where an operator would be expected to: 1. Insert the record on a node 2. Call loadPartitionDenylist on that node 3. Call loadPartitionDenylist on other nodes in the cluster if more than 1 is behind because of a problematic partition Just cuts out one step. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
