WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-bugs

[Xen-bugs] [Bug 1698] Xen dom0 crashes immediately after boot

To: xen-bugs@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-bugs] [Bug 1698] Xen dom0 crashes immediately after boot
From: bugzilla-daemon@xxxxxxxxxxxxxxxxxxx
Date: Mon, 20 Dec 2010 09:57:53 -0800
Delivery-date: Mon, 20 Dec 2010 09:58:21 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <bug-1698-3@xxxxxxxxxxxxxxxxxxxxxxxxxxx/bugzilla/>
List-help: <mailto:xen-bugs-request@lists.xensource.com?subject=help>
List-id: Xen Bugzilla <xen-bugs.lists.xensource.com>
List-post: <mailto:xen-bugs@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-bugs>, <mailto:xen-bugs-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-bugs>, <mailto:xen-bugs-request@lists.xensource.com?subject=unsubscribe>
Reply-to: bugs@xxxxxxxxxxxxxxxxxx
Sender: xen-bugs-bounces@xxxxxxxxxxxxxxxxxxx
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=1698





------- Comment #3 from konrad.wilk@xxxxxxxxxx  2010-12-20 09:57 -------
(In reply to comment #2)
> Another R710, being an exact clone of this configuration crashes as well. 
> 
> Some older machines (~ 2 months), again being R710 but having a slight
> different configuration (< 2TB disk thus using DOS partition tables, slightly
> different CPU, RAID controller and no SAS disks) don't have this problem. They
> run Debian stable (thus and old style dom0+hypervisor with Xen 3.2). The 
> output
> may 

The issue here seems from the fact that the NUMA configuration steers the
hypervisor (not the Linux kernel) in allocating memory differently. I am not
sure what the allocation code in the hypervisor does but it looks that the
node 0 ends up having no DMA32 memory.

The older machines might not have the SRAT table, or the older hypervisor =
either which causes the hypervisor to have more DMA32 type memory available.

So what we need to try to see is if you can limit the amount of memory that
Dom0 gets so there is enough space in the hypervisor to swap with DMA32 type
memory. Try dom0_mem=max:4GB for example on your Hypervisor line. 

The other thing to try is to disable the Hypervisor from parsing the NUMA
information (the SRAT table). I am not sure what the command line argument for
that is (I would think 'numa=off')

So try either one or both and see what the 'xm info' has for node_to_dma32_mem 
    value.


-- 
Configure bugmail: 
http://bugzilla.xensource.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

_______________________________________________
Xen-bugs mailing list
Xen-bugs@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-bugs