Diego, please raise a bug and share the periodic statedumps of processes which have memory leaks.
On Thu, Jul 18, 2019 at 5:53 PM Diego Remolina <dijuremo@xxxxxxxxx> wrote:
Are you using FUSE mounts to access the file system? I have observed the same issue all the way from 3.x series to 4.x series. I have to reboot servers monthly, at most every 45 days because gluster keeps eating the ram.As you can see, 30 days in we are right at around 30GB of RAM used.DiegoOn Thu, Jul 18, 2019 at 4:15 AM Sanju Rakonde <srakonde@xxxxxxxxxx> wrote:Christian,To debug memory leaks, we need periodic statedumps of respective processes. Please provide the statedumps.I suspect that you are hitting https://bugzilla.redhat.com/show_bug.cgi?id=1694612. This bug is addressed in release-5.6Thanks,SanjuOn Thu, Jul 18, 2019 at 1:30 PM Christian Meyer <chrmeyer@xxxxxxxxxxx> wrote:Hi everyone!
I'm using a Gluster 5.4 Setup with 3 Nodes with three volumes
(replicated) (one is the gluster shared storage).
Each node has 64GB of RAM.
Over the time of ~2 month the memory consumption of glusterd grow
linear. An the end glusterd used ~45% of RAM the brick processes
together ~43% of RAM.
I think this is a memory leak.
I made a coredump of the processes (glusterd, bricks) (zipped ~500MB),
hope this will help you to find the problem.
Could you please have a look on it?
Download Coredumps:
https://s3.eu-central-1.amazonaws.com/glusterlogs/gluster_coredump.zip
Kind regards
Christian
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
https://lists.gluster.org/mailman/listinfo/gluster-users
--_______________________________________________Thanks,Sanju
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
https://lists.gluster.org/mailman/listinfo/gluster-users
Thanks,
Sanju_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx https://lists.gluster.org/mailman/listinfo/gluster-users
