anmolnar commented on PR #2185: URL: https://github.com/apache/zookeeper/pull/2185#issuecomment-2537053559
> We ran into perf issue when had sync on the ZookeeperServer object. Whenever snapshot was taken, the sync operations were queued up in the outstanding request queue and the sync() operation latency was largely increased. (edited) > There was no synchronize between snapshotting and sync() before. > The intention of adding the synchronized was to takeSnapshot() and restore() was to sync between the two operations. I didn’t realize the sync() call also requires ZookeeperServer object lock. When has synchronized been added to take/restore snapshot? Are you saying that `sync()` was already synchronized which should prevent multiple sync's running at the same time, but accidentally you introduced snapshotting into this lock? -- 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: notifications-unsubscr...@zookeeper.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
