ashishkumar50 commented on PR #11571: URL: https://github.com/apache/ozone/pull/11571#issuecomment-6345777243
@hani-fouladgar My main concern is the case where one of the SCM followers is consistently lagging due to some underlying issue. If the follower is temporarily behind and can eventually catch up, this approach makes sense. However, if the follower continues to lag, we could end up throttling block deletions indefinitely, even when the SCM leader(and one follower) and OM are otherwise capable of processing deletions at a higher rate. This would cause deleted-key entries to accumulate in the OM DB, potentially leading to significant DB growth, especially when deleting directories containing millions of keys. A problem with a single SCM follower could become a bottleneck for the entire deletion pipeline. We need to distinguish temporary lag from persistent lag and ensure that backpressure does not cause unbounded growth in the OM DB. -- 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]
