[Bugs] [Bug 1615443] New: Glusterfs client Memory Leak

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1615443

            Bug ID: 1615443
           Summary: Glusterfs client Memory Leak
           Product: GlusterFS
           Version: 3.12
         Component: libglusterfsclient
          Severity: medium
          Assignee: bugs at gluster.org
          Reporter: okane at khm.de
                CC: bugs at gluster.org



Created attachment 1475607
  --> https://bugzilla.redhat.com/attachment.cgi?id=1475607&action=edit
Gluster dumps, version info, gluster info

Description of problem:

glusterfs process using too much Memory 


Version-Release number of selected component (if applicable):

glusterfs-3.12  from ovirt-4.2 repositories

rpm -qa | grep gluster
glusterfs-rdma-3.12.11-1.el7.x86_64
glusterfs-client-xlators-3.12.11-1.el7.x86_64
libvirt-daemon-driver-storage-gluster-3.9.0-14.el7_5.6.x86_64
glusterfs-cli-3.12.11-1.el7.x86_64
python2-gluster-3.12.11-1.el7.x86_64
glusterfs-geo-replication-3.12.11-1.el7.x86_64
glusterfs-server-3.12.11-1.el7.x86_64
glusterfs-fuse-3.12.11-1.el7.x86_64
glusterfs-gnfs-3.12.11-1.el7.x86_64
glusterfs-events-3.12.11-1.el7.x86_64
glusterfs-3.12.11-1.el7.x86_64
vdsm-gluster-4.20.35-1.el7.x86_64
glusterfs-libs-3.12.11-1.el7.x86_64
glusterfs-api-3.12.11-1.el7.x86_64



How reproducible:

This happens on all "non brick" hosts,  3 hosts with bricks + 3 without 
Since Ovirt-4.2.0  most likely since gluster-3.12 , 3.7 (3.9?) was OK.


Steps to Reproduce:
1. Activate Host (mount gluster filesystem) 

2. let run

I realize now I have not simply tried an active host without VMs.. I will test
that
and add to this post.


Actual results:

currently using 64GB Ram (and growing ca. 150MB every 1/2 hour. 4 days uptime,
maximum 7 VMs on host.

Expected results:

use less memory?


Additional info:

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.