[OpenAFS-devel] bonnie++ problem on Linux w/ OpenAFS client
Alf Wachsmann
alfw@SLAC.Stanford.EDU
Thu, 29 Nov 2001 15:32:54 -0800 (PST)
On Wed, 28 Nov 2001, Derek Atkins wrote:
> Try running under strace(1) and see where it fails?
>
> > > Alf Wachsmann <alfw@SLAC.Stanford.EDU> writes:
> > > > When I run "bonnie++ -d /afs/slac/test -s 2g" on this machine I get after
> > > > a short while the following error:
> > > > Writing with putc()...done
> > > > Writing intelligently...Can't write block.
> > > > Bonnie: drastic I/O error (write(2)): No such file or directory
Here are some more test results:
1. strace shows that bonnie++ fails in a write() function which returns
"-1" which lets bonnie++ gracefully terminate. It even cleans up
after itself by deleting the files it has created in the very same
AFS partition it claims it cannot access.
2. During all bonnie++ runs I had a second xterm opened on the client
machine. I could access ("/bin/ls -l") the directory bonnie++ was
writing to all the time.
3. I had the AFS cache on the RH7.2 client on a ext3 partition. I changed
that to ext2. Same bonnie++ failure.
4. I tried different bonnie++ parameters: "-s 1g -r 500M" instead of
"-s 2g". Same failure.
5. Tried the same hardware but RedHat6.2 (2.2.19-6.2.7smp kernel) with
OpenAFS-1.2.2. That works.
6. I ran some other (home grown) filesystem test from my RH7.2 client
against the AFS filesystem. These tests are not really sophisticated.
But none failed.
I am out of ideas what else to test/check/modify to get this working.
Thanks,
Alf.
-----------------------------------------------------------------------
Alf Wachsmann | e-mail: alfw@slac.stanford.edu
SLAC Computing Service | Phone: +1-650-926-4802
2575 Sand Hill Road, M/S 97 | FAX: +1-650-926-3329
Menlo Park, CA 94025, USA | Office: Bldg. 50/323
-----------------------------------------------------------------------
http://www.slac.stanford.edu/~alfw (PGP)
-----------------------------------------------------------------------