lbradstreet opened a new pull request #10241: URL: https://github.com/apache/kafka/pull/10241
During a slow log recovery it's easy to think that loading `.snapshot` files is a multi-second process. Often it isn't the snapshot loading that takes most of the time, rather it's the time taken to further rebuild the producer state from segment files. This PR times both snapshot load and segment recovery phases to better indicate what is taking time. Example test output: ``` [2021-03-01 22:35:28,129] INFO [Log partition=foo-0, dir=/var/folders/cb/5my51vjd1js380qcr_v245bh0000gp/T/kafka-16876782135717603479] Reloading from producer snapshot and rebuilding producer state from offset 0 [2021-03-01 22:35:28,129] INFO [Log partition=foo-0, dir=/var/folders/cb/5my51vjd1js380qcr_v245bh0000gp/T/kafka-16876782135717603479] Producer state recovery took 0ms for snapshot load and 0ms for segment recovery from offset 0 [2021-03-01 22:35:28,131] INFO Completed load of Log(dir=/var/folders/cb/5my51vjd1js380qcr_v245bh0000gp/T/kafka- ``` ---------------------------------------------------------------- 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]
