At 13:40 19-8-2001 -0400, tls@xxxxxxxxxxxx wrote:
We're building a server for a CVS repository consisting of several tens
of thousands of files; about 1.5GB of data total. It's a two-processor
Pentium-III machine with a Mylex RAID controller and 4GB of RAM.
We installed from the RH7.1/XFS 1.0.1 media, built a 2.4.5 kernel with
the XFS 1.0.1 patches, copied the repository into place and ran a small
script that copied the repository four times, copied the resulting directory,
then removed both copies.
I let it run overnight -- came in this morning to find that the first time
I ran any command that generated any disk I/O on the box, I got dozens of
"0 order alloc failed" messages on the console and then a hard hang.
The error message is a general kernel error message which seems to be a
highmem problem.
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.
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.
Cheers
Thor
--
Seth
Every program has two purposes one for which
it was written and another for which it wasn't
I use the last kind.
|