Well this is still happening ... I have tried to isolate what could be causing but not much luck yet. Thought the VMs may have been IO bound but that not the case and even tried upping the vCPU allocation from one to two as plenty of head room. When it locks up I see this on a strace:
[pid 1343] read(14, 0x7fff82aeb360, 4096) = -1 EAGAIN (Resource temporarily unavailable)
[pid 1343] read(7, "\0", 512) = 1
[pid 1343] read(7, 0x7fff82aec160, 512) = -1 EAGAIN (Resource temporarily unavailable)
[pid 1343] select(26, [7 10 13 14 16 17 22 25], [], [], {1, 0}) = 1 (in [16], left {0, 999981})
[pid 1343] read(16, "\16\0\0\0\0\0\0\0\376\377\377\377\0\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0"..., 128) = 128
[pid 1343] rt_sigaction(SIGALRM, NULL, {0x7f210b2c0510, ~[KILL STOP RTMIN RT_1], SA_RESTORER, 0x7f210ac22500}, 8) = 0
[pid 1343] write(8, "\0", 1) = 1
[pid 1343] write(15, "\1\0\0\0\0\0\0\0", 8) = 8
[pid 1343] read(16, 0x7fff82aec2d0, 128) = -1 EAGAIN (Resource temporarily unavailable)
[pid 1343] timer_gettime(0x1, {it_interval={0, 0}, it_value={0, 0}}) = 0
[pid 1343] timer_settime(0x1, 0, {it_interval={0, 0}, it_value={0, 656000000}}, NULL) = 0
[pid 1343] select(26, [7 10 13 14 16 17 22 25], [], [], {1, 0}) = 2 (in [7 14], left {0, 999998})
[pid 1343] read(14, "\1\0\0\0\0\0\0\0", 4096) = 8
[pid 1343] read(14, 0x7fff82aeb360, 4096) = -1 EAGAIN (Resource temporarily unavailable)
[pid 1343] read(7, "\0", 512) = 1
[pid 1343] read(7, 0x7fff82aec160, 512) = -1 EAGAIN (Resource temporarily unavailable)
Does that shed any light ? Trying to find a how to for upgrading to the latest KVM/QEMU.
Thanks.
----- Original Message -----
From: "Gleb Natapov" <gleb@xxxxxxxxxx>
To: "Phil Daws" <uxbod@xxxxxxxxxxxx>
Cc: kvm@xxxxxxxxxxxxxxx
Sent: Wednesday, 10 April, 2013 3:10:27 PM
Subject: Re: KVM Guest Lock up (100%) again!
On Mon, Apr 08, 2013 at 09:02:04PM +0100, Phil Daws wrote:
> Hello all,
>
> Another lock up again this evening :( am wondering whether should consider upgrading the kernel to 3.7.10 and the latest version of KVM. Thoughts ?
>
Please try with latest kvm/qemu. This is not the list to report vendor specific problems.
> Thanks.
>
> ----- Original Message -----
> To: kvm@xxxxxxxxxxxxxxx
> Sent: Thursday, 4 April, 2013 3:36:11 PM
> Subject: KVM Guest Lock up (100%) again!
>
> One of my KVM guests locked up again at 100% CPU! Any thoughts on how I can diagnose it ? We would love to put into production but am very concerned about the current stability. I have tried to re-direct the console, through screen, to see whether there is a spin lock that is causing the problem; though all I got in the log file was a login prompt. What is the correct way of redirecting the console in KVM on a CentOS 6.4 system please ? Thanks.
> --
> To unsubscribe from this list: send the line "unsubscribe kvm" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe kvm" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
Gleb.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
