Re: "0-order allocation failed"

[Seth Mos <knuffie@xs4all.nl>]

At 15:00 19-8-2001 -0400, tls@reefedge.com wrote:
>On Sun, Aug 19, 2001 at 07:59:39PM +0200, Seth Mos wrote:
> >
> > The error message is a general kernel error message which seems to be a
> > highmem problem.
>
>I don't see it unless I use XFS.  I don't see it, on this exact same
>system, with ext2 or reiserfs.

That's possible, but it is not limited to XFS.

> > >I see mention of this bug in a list message from April, but nothing since
> > >then.  Has it been fixed, or even analyzed?  If it's been fixed, is 
> there a
> > >simple patch against 1.0.1 or will I need to run the latest from CVS?
> >
> > I guess that using a kernel later then 2.4.5 may help but this is not
> > directly a XFS related error but XFS will help exposing this message
> > because it pushes the VM harder.
>
>What does "pushes the VM harder" mean, in this context?

If the filesystem endures high throughput will put more pressure on the 
memory that is being used. The "0 order allocation" message is trying to 
tll you that it didn't succeed in allocating memory.

> > I don't know if the highmem stuff is significantly better in 2.4.9.
> > That is the current CVS tree version which is probably your best bet for
> > testing. A 2.4.8 patch is also available on the FTP site.
>
>I noticed a later list message suggesting that a change to XFS in the 2.4.6
>timeframe should significantly improve stability on highmem machines; I also
>found a message suggesting that there were deadlocks elsewhere in the Linux
>kernel on highmem machines until 2.4.7.  So I had high hopes for the 2.4.9
>patch.

2.4.9 is not available as a patch on the FTP site yet and you will need to 
fetch it from CVS.

>Unfortunately, though I haven't gotten the machine to hang hard yet, I'm
>now running 2.4.9 and a simple cp -R of our CVS repository from one directory
>on an XFS filesystem to another produces the "0 order alloc" messages.

These are rather irritating but not related to XFS. If you push ext2 or 
reiserfs hard enough they will  show up as well.

>Is this actually believed to be fixed at the moment, or not?  Looks like
>not, but I'll leave this running for a few hours and see if I can get an
>actual hang.

The message is fairly harmless for now but the box should at least survive 
this. Stability on highmem is at least better then what it used to be.

>Thor

Cheers

--
Seth
Every program has two purposes one for which
it was written and another for which it wasn't
I use the last kind.