[Gluster-users] gluster client 4.1 memory leak

Richard Neuboeck hawk at tbi.univie.ac.at
Wed Feb 6 13:38:57 UTC 2019

Hi Gluster-Group,

I've stumbled upon a memory leak in the gluster client 4.1. It
manifests itself the same way the last one [1] did in 3.12. Memory
consumption of the glusterfs process climbs until the system is out
of memory and the process gets killed.

Excerpt from the system log:
rnel: Out of memory: Kill process 77419 (glusterfs) score 505 or
sacrifice child
rnel: Killed process 77419 (glusterfs) total-vm:71549476kB,
anon-rss:70730944kB, file-rss:196kB, shmem-rss:0kB

I didn't find a bug report for this problem for version 4.1 on
Bugzilla and I'm unsure if I should open a bug report there or
somewhere else.

I'm running gluster 4.1 from the CentOS repo on the client and the


The files are transfered with rsync:
Number of files: 15,143,321 (reg: 13,735,846, dir: 1,283,846, link:
123,471, special: 158)
Total file size: 1,232,120,420,136 bytes

I've created statedumps every three hours. I hope that helps to
track down the problem. The dumps are here:

Is there any thing else I can help with to solve this problem?




-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20190206/805f3605/attachment.sig>

More information about the Gluster-users mailing list