From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Cooper Subject: Re: Xen host crash Date: Thu, 29 Aug 2013 17:25:13 +0100 Message-ID: <521F75E9.7060006@citrix.com> References: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============8469493985538215134==" Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Rushikesh Jadhav <2rushikeshj@gmail.com> Cc: "xen-devel@lists.xen.org" List-Id: xen-devel@lists.xenproject.org --===============8469493985538215134== Content-Type: multipart/alternative; boundary="------------090408060108050400080104" --------------090408060108050400080104 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit On 29/08/13 17:22, Rushikesh Jadhav wrote: > Hi People, > > I had a crash of Xen [ 3.4.2 ] host today and the crash log was dumped > in /var/crash/ > > While analyzing the crash log & call trace on this 24 PCPU host, I > found that some of PCPUs were in idle state & many were having same > call trace as > > PCPU7 > Call Trace: > [ffff828c8010e310] dump_domains+0x4d0 > ffff828c80175b7c crash_nmi_callback+0x2c > ffff828c8015f2f9 do_nmi+0x39 > ffff828c801d6877 handle_ist_exception+0x52 > ffff828c801787b2 acpi_safe_halt+0x2 > > > Only one PCPU has got call trace as > > PCPU6 > Call Trace: > [ffff828c8010e310] dump_domains+0x4d0 > ffff828c8010eeb7 kexec_crash+0x57 > ffff828c80127b36 panic+0x136 > ffff828c8011b7da __print_symbol+0x8a > ffff828c8019b4ab vmx_asm_vmexit_handler+0x6b > ffff828c80100000 __per_cpu_shift+0x800ffff4 > ffff828c8015eb75 show_stack+0x155 > ffff828c8015eeba fatal_trap+0x6a > ffff828c801567a1 nmi_watchdog_tick+0x131 > ffff828c8015f37f do_nmi+0xbf > ffff828c801d6877 handle_ist_exception+0x52 > ffff828c8011ab02 _spin_lock+0x12 > > Can anyone please help me understand this & try to find out crash cause ? > There are no error logs in messages or kernel at the time of crash. > > I checked for C-States and it is set to 2. > > Thanks. This is a spinlock deadlock, resulting in the NMI watchdog timing out and killing the host. Do you have Stack and register dump for PCPU6 ? ~Andrew > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel --------------090408060108050400080104 Content-Type: text/html; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit

On 29/08/13 17:22, Rushikesh Jadhav wrote:

Hi People,

I had a crash of Xen [ 3.4.2 ] host today and the crash log was dumped in /var/crash/

While analyzing the crash log & call trace on this 24 PCPU host, I found that some of PCPUs were in idle state & many were having same call trace as

PCPU7

Call Trace:

[ffff828c8010e310] dump_domains+0x4d0

ffff828c80175b7c crash_nmi_callback+0x2c

ffff828c8015f2f9 do_nmi+0x39

ffff828c801d6877 handle_ist_exception+0x52

ffff828c801787b2 acpi_safe_halt+0x2

Only one PCPU has got call trace as

PCPU6

Call Trace:

[ffff828c8010e310] dump_domains+0x4d0

ffff828c8010eeb7 kexec_crash+0x57

ffff828c80127b36 panic+0x136

ffff828c8011b7da __print_symbol+0x8a

ffff828c8019b4ab vmx_asm_vmexit_handler+0x6b

ffff828c80100000 __per_cpu_shift+0x800ffff4

ffff828c8015eb75 show_stack+0x155

ffff828c8015eeba fatal_trap+0x6a

ffff828c801567a1 nmi_watchdog_tick+0x131

ffff828c8015f37f do_nmi+0xbf

ffff828c801d6877 handle_ist_exception+0x52

ffff828c8011ab02 _spin_lock+0x12

Can anyone please help me understand this & try to find out crash cause ?

There are no error logs in messages or kernel at the time of crash.

I checked for C-States and it is set to 2.

Thanks.

This is a spinlock deadlock, resulting in the NMI watchdog timing out and killing the host. Do you have Stack and register dump for PCPU6 ?

~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

--------------090408060108050400080104-- --===============8469493985538215134== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel --===============8469493985538215134==--