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-devel

Re: [Xen-devel] 0-order allocation failed

To: stevegt@xxxxxxxxxxxxx
Subject: Re: [Xen-devel] 0-order allocation failed
From: Keir Fraser <Keir.Fraser@xxxxxxxxxxxx>
Date: Wed, 11 Feb 2004 08:13:39 +0000
Cc: xen-devel@xxxxxxxxxxxxxxxxxxxxx
Delivery-date: Wed, 11 Feb 2004 08:18:52 +0000
Envelope-to: steven.hand@xxxxxxxxxxxx
In-reply-to: Your message of "Tue, 10 Feb 2004 20:12:30 PST." <20040211041230.GZ3272@pathfinder>
List-archive: <http://sourceforge.net/mailarchive/forum.php?forum=xen-devel>
List-help: <mailto:xen-devel-request@lists.sourceforge.net?subject=help>
List-id: List for Xen developers <xen-devel.lists.sourceforge.net>
List-post: <mailto:xen-devel@lists.sourceforge.net>
List-subscribe: <https://lists.sourceforge.net/lists/listinfo/xen-devel>, <mailto:xen-devel-request@lists.sourceforge.net?subject=subscribe>
List-unsubscribe: <https://lists.sourceforge.net/lists/listinfo/xen-devel>, <mailto:xen-devel-request@lists.sourceforge.net?subject=unsubscribe>
Sender: xen-devel-admin@xxxxxxxxxxxxxxxxxxxxx
This means that a number of GFP_ATOMIC allocations failed. This is no
surprise in a low-memory location where your root filesystem is
mounted over NFS: Linux isn't able to launder and evict pages quickly
enough to satisfy all non-blocking page-allocation requests.

I can practically guarantee exactly the same behaviour in native Linux
under the same circumstances (we saw various NFS-root weirdnesses on
native Linux under high load when stress-testing the MM code).

 -- Keir


> Looks like yesterday's paging problems were an artifact of something on
> that particular dom0 disk image -- I haven't been able to reproduce it
> on other nodes, and after I re-imaged the same node (with the same
> image), the problem has gone away there too, so that rules out hardware.  
> 
> However, something else did pop up -- while trying to break things with
> "perl -e '$a="a"x100000000'", I got the following messages; do we care?
> This is in mainstream linux mm/page_alloc.c, and it's hard for me to
> tell from the code whether these are outright errors or whether they
> were recoverable.  Does anyone know?
> 
>   DOM4: __alloc_pages: 0-order allocation failed (gfp=0xf0/0)
>   DOM4: __alloc_pages: 0-order allocation failed (gfp=0xf0/0)
>   DOM4: __alloc_pages: 0-order allocation failed (gfp=0xf0/0)
>   DOM4: __alloc_pages: 0-order allocation failed (gfp=0xf0/0)
> 
> This was still on last Monday's version of 1.2; I'll see if I can
> reproduce it in today's 1.2; I'm deploying that later tonight.
> 
> Steve


-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/xen-devel