Created attachment 1475607 [details] 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:
Same as Bug 1593826 - Glusterfs being killed due to out of memory situation [NEEDINFO]
*** This bug has been marked as a duplicate of bug 1593826 ***