[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] cpuidle and un-eoid interrupts at the local apic

Thimo E. wrote on 2013-09-17:
> Hello Yang,
> The problem that we are discussing here seems to come from the
> interaction of my raid controller (Areca, ARC-1212) and the hardware.
> When I enable MSI on that controller the server crashes with the known
> error messages between 1 day and 7 days.  When I disable MSI I don't
> see these crashes anymore.
> Andrew doesn't have this controller but he could also observe these
> type of crashes with another MSI enabled card, I think the networc
> card. But less often, I think 2 times in the last 4 months.
> My cpuinfo nor dmesg show any microcode information, perhaps Xen hides
> that info ?!
'cat /proc/cpuinfo' will show the microcode version.

> Best regards
>    Thimo
> Am 17.09.2013 04:09, schrieb Zhang, Yang Z:
>> Hi, Thimo,
>> I cannot reproduce this issue in my box after running about two weeks:
>> I started four guests (two PV guests and two HVM guests). And each
>> guest
> runs a simple workload (ping a remote machine). After two weeks, the
> machine still works no crash and panic happen. Are there any special
> workload required to reproduce this issue?
>> Attached the cpuinfo and pci info in my box. Please compare it with
>> yours to see whether it is same. Especially, the microcode version.
>> Best regards,
>> Yang

Best regards,

Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.