bshashikant edited a comment on issue #698: HDDS-3234. Fix retry interval default in Ozone client. URL: https://github.com/apache/hadoop-ozone/pull/698#issuecomment-600997186 > > @bshashikant I agree with you. This is not a permanent solution, till then it is a temporary fix. As right now default 1s, we see that the system is doing a lot of retries, and the queue limit is reaching its max size very quickly. By changing it to 15s, we have observed that the queue limit is under control and at max reached around 200. > > Do you see any issues with changing to 15s? > > In various other components outside of Ozone, I have seen a retry policy of 60s. Considering that, 15s is still reasonable for now. @bharatviswa504 , the default retry policy would make it sleep for 15 sec even when a request fails with NotLeader or LeaderNotReady or in general any intermittent IO Exception from network as well. What instead we can do for now is., enforce ExceptionBasedRetryPolicy for ratis in Ozone and make it 15s for ResourceUnavailable which can be changed to exponential backoff retry policy in ozone and for other exceptions make it 3s or so. What do you think?
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: ozone-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: ozone-issues-h...@hadoop.apache.org
