FrankChen021 commented on issue #10566: URL: https://github.com/apache/druid/issues/10566#issuecomment-725784868
I don't know what the migration you mentioned is. But I think 1 hour is a little bit short. Default segment granularity is hour, and once there's any failure, druid has to re-ingest data from the last offset which might be one hour ago. BTW, how many replicas do you have for these topics ? For a typical kafka cluster, 3 replicas is recommended, but for such scenario, 2 replicas are enough for us to keep high availability while keeping lower disk consumption. ---------------------------------------------------------------- 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: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
