Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Wed, Aug 09, 2017 at 04:45:56PM +1200, Brady, Mike wrote: > I currently have a three node cluster running 9.0.8 that I have tried > upgrading to 9.0.9rc1. > > A couple of minutes after I start a KVM VM on one of the nodes, the node > becomes unresponsive and reboots. The VM is the third node in the cluster > and is configured with no storage and shows as being back on line in the > cluster before the node that it is running on goes unresponsive. > > I have rolled back to 9.0.8 and all is well again. > > On one of the crashes I did manage to catch some "Out of Memory" messages on > the console, but the system had around 13G free out of 16G as shown by atop. > I have tried increasing vm.min_free_kbytes to 131072 but it made no > difference. > > All nodes are running up to date CentOS 7.3. > > I am happy to provide any additional info that is needed if you let me know > what is required. IO stack (e.g. lsblk ; lsblk -t ; lsblk -D) may be interesting, as well as the drbd configuration (drbdadm dump ), the kernel logs around the "crash" if possible, and your "best guess" as to what the box may have been doing when it became unresponsive. -- : Lars Ellenberg : LINBIT | Keeping the Digital World Running : DRBD -- Heartbeat -- Corosync -- Pacemaker DRBD® and LINBIT® are registered trademarks of LINBIT __ please don't Cc me, but send to list -- I'm subscribed