On Fri, Dec 9, 2016 at 2:28 PM, Diego Castro <diego.castro@xxxxxxxxxxxxxx> wrote: > Hello, my case is very specific but i think other may have this issue. > > I have a ceph cluster up and running hosting block storage for my openshift > (kubernetes) cluster. > Things goes bad when i "evacuate" a node, which is move all containers to > other hosts, when this happens i can see a lot of map/mount commands and > suddenly the node crashes, here is the log [1]. > > > 1.https://gist.github.com/spinolacastro/ff2bb85b3768a71d3ff6d1d6d85f00a2 > > [root@n-13-0 ~]# cat /etc/redhat-release > CentOS Linux release 7.2.1511 (Core) > > [root@n-13-0 ~]# uname -a > Linux n-13-0 3.10.0-327.36.3.el7.x86_64 #1 SMP Mon Oct 24 16:09:20 UTC 2016 > x86_64 x86_64 x86_64 GNU/Linux Does this still happen with the kernel that is (soon to be) in CentOS 7.3? The Ceph kernel code got a big update in 7.3, with a rebase to the latest RBD and CephFS (libceph.ko, rbd.ko and ceph.ko) upstream. CentOS 7.3 is not GA yet, but you can get kernel-3.10.0-514.2.2.el7 from the CR repository: https://wiki.centos.org/AdditionalResources/Repositories/CR - Ken _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |