Ian.Campbell at citrix
May 4, 2012, 6:43 AM
Post #2 of 2
On Fri, 2012-05-04 at 13:46 +0100, Alexis wrote:
> I'm using the following configuration on 8 different servers:
> xen 4.1.2 from opensuse 12.1 with Kernel 3.1.10-1.9-xen
It seems like you are using both hypervisor and kernel supplied by the
distro. I would recommend reporting to the distro in the first instance
since they'll have much better insight into what the known bugs and
issues etc are for their binaries.
"noreboot" on the hypervisor command line and/or serial console might
help you get the logs they will no doubt ask you for.
> AMD Phenom(tm) II X6 1075T 8GB ram
> Dom0 : 1GB memory and 2 vcpus on core 0
> (via boot option dom0_mem=1024M dom0_max_vcpus=2)
> DomU :
> 1 PV guest (opensuse12.1) 1GB memory 2 vcpus on core 1 to 5
> 1 PV guest (opensuse12.1) 1GB memory 3 vcpus on core 1 to 5
> 1 HVM guest with gplpv drivers(XP sp3) 1GB memory 1 vcpus on core 1 to 5
> Also I'm using /usr/sbin/xm sched-credit -d Domain-0 -w 512 and
> /usr/sbin/xenpm set-max-cstate 0 for handeling a bug with AMD family 16
> I've experienced server crash on all of my 8 servers. The time between
> crashes is quite long (5-6 days) most of time it occures at night time when
> the servers are not suppose to be overload.
> I don't find anything in the logs so I don't really know how to track down
> this issue.
> Any idea / advice would be appreciate.
> Xen-users mailing list
> Xen-users [at] lists
Xen-users mailing list
Xen-users [at] lists