Description of problem: https://review.gluster.org/17105 brought in a change where requests are queued by event-threads and the queue is drained by rpcsvc-request-handler threads. With increasing server event threads, the single queue becomes a bottleneck due to contention involved in queueing and dequing requests. A better solution would be to let each pair of (event-thread, rpcsvc-request-handler-thread) to have its own queue to reduce contention. Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
REVIEW: https://review.gluster.org/21703 (libglusterfs: rename macros roof and floor to not conflict with math.h) posted (#1) for review on master by Raghavendra G
REVIEW: https://review.gluster.org/21703 (libglusterfs: rename macros roof and floor to not conflict with math.h) posted (#19) for review on master by Raghavendra G
REVIEW: https://review.gluster.org/19997 (rpcsvc: provide each request handler thread its own queue) posted (#26) for review on master by Raghavendra G
This bug is getting closed because a release has been made available that should address the reported issue. In case the problem is still not fixed with glusterfs-6.0, please open a new bug report. glusterfs-6.0 has been announced on the Gluster mailinglists [1], packages for several distributions should become available in the near future. Keep an eye on the Gluster Users mailinglist [2] and the update infrastructure for your distribution. [1] https://lists.gluster.org/pipermail/announce/2019-March/000120.html [2] https://www.gluster.org/pipermail/gluster-users/