On 9/10/2017 2:02 AM, Pavel Szalbot wrote:
WK: I use bonded 2x10Gbps and I do get crashes only in heavy I/O situations (fio). Upgrading system (apt-get dist-upgrade) was ok, so this might be even related to amount of IOPS. -ps
Well, 20Gbps of writes could overwhelm a lot of DFS clusters. We saw VMs being overwhelmed by Gigabit speed large file copy ops into the VM with early versions of Ceph and MooseFS (MFS 1.6.x. was really vulnerable to that, MFS 3.x is much improved in that regard).
Your success with the fuse client may be because it has a more complicated path than libgfapi, thus slowing down/buffering data enough to let the cluster catch up.
Also you are on simple replica. With a more advanced Gluster setup involving distribute that write load would be spread out more.
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://lists.gluster.org/mailman/listinfo/gluster-users
