Hi,
Could you please paste the crash back trace.
On 2017/9/27 16:15, net...@tango.lu wrote:
> Hello All,
>
> I wrote earlier about our OCFS2 crash issue in KVM due to bug in the SMP
> code.
>
> For this we come up with a solution:
>
> Instead of using multiple vcpus
> 8
>
> using a
Hi,
It seems that your cluster has something wrong with connection between
nodes. So no dlm message can be sent out.
This may cause a node being fenced, thus to crash.
Please check your network condition including switch, Ethernet HBA card,
etc.
Thanks,
Changwei
On 2017/11/28 18:07,
Hi Jim,
From the log you provided, it seems that one node died.
If I remember correctly, you are using kernel-4.9 in which a bug resides
causing cluster hang if a node dies.
You can refer to a fix in kernel mainline.
commit 1c01967116a678fed8e2c68a6ab82abc8effeddc
Author: Changwei Ge
On 2017/12/28 3:02, Jim Okken wrote:
> Peter,
>
> I did not want to flood my first email with details and make it 3 pages long.
> i gladly will provide more details. first I'd like to ask that you be less
> condescending. You have no idea the journey I took toward using ocfs2 in this
>