On Tue, 2009-10-13 at 14:55 +0200, Benoit SIBAUD wrote:
> Hi,
>
> > I believe this bug was fixed in Debian kernel package version
> > 2.6.26-15lenny3. Can you check that this bug is no longer present in
> > the current stable kernel version?
>
> Still a kernel panic with a 2.6.26-19
> (linux-image-2.6.26-2-686_2.6.26-19_i386.deb). The stack trace seems to
> always contain CIFSFindNext+0x174/0x2dc [cifs].
>
> (after a fresh reboot)
>
> $ smbmount //xxxxxxxxx/xxxxx-xxx /tmp/crash -o
> username=xxxxxxxx,workgroup=xxxxxxx
> $ cd "/tmp/crash/__Archives/07_Communication_Information/Journées
> Techniques CORE/Journée du 27 octobre 2005 - Qualité des
> réseaux/journecore/ppt"
> $ find . -type f >/dev/null 2>&1
> [KERNEL PANIC]
>
> (in attachment, kernel panic screenshot)
We got this fix via 2.6.26.7:
commit 0752f1522a9120f731232919f7ad904e9e22b8ce
Author: Steve French <sfrench@us.ibm.com>
Date: Tue Oct 7 20:03:33 2008 +0000
[CIFS] make sure we have the right resume info before calling CIFSFindNext
But it requires these additional fixes which we're missing for some
reason:
commit b77d753c413e02559669df66e543869dad40c847
Author: Steve French <sfrench@us.ibm.com>
Date: Wed Oct 8 19:13:46 2008 +0000
[CIFS] Check that last search entry resume key is valid
commit a364bc0b37f14ffd66c1f982af42990a9d77fa43
Author: Jeff Layton <sfrench@us.ibm.com>
Date: Tue Oct 21 14:42:13 2008 +0000
[CIFS] fix saving of resume key before CIFSFindNext
Ben.
--
Ben Hutchings
Once a job is fouled up, anything done to improve it makes it worse.
Attachment:
signature.asc
Description: This is a digitally signed message part