Messages in this thread | | | From | Chuck Lever <> | Subject | Re: NFS oops in 2.6.26rc4 | Date | Fri, 30 May 2008 15:37:01 -0400 |
| |
On May 30, 2008, at 3:03 PM, Dave Jones wrote: > On Fri, May 30, 2008 at 11:31:48AM -0700, Trond Myklebust wrote: >> On Fri, 2008-05-30 at 14:21 -0400, Dave Jones wrote: >> >>> mount point in the fstab is .. >>> >>> gelk:/mnt/data /mnt/nfs/gelk nfs >>> nfsvers=3,tcp 0 0 >>> >>>> What "brand" of server were you trying to mount? >>> >>> It's just another linux box. A no-name core2 duo, running 2.6.25. >>> >>>> How often can you reproduce this? >>> >>> Seems to do it every time I ask it to. >> >> Could you provide us with a binary tcpdump in that case? I'd love to >> have a look at the actual filehandle the server is producing. > > This is from the client side: http://www.codemonkey.org.uk/junk/ > tcp.out > Wireshark picks up some of those packets as being 'malformed', which > could be a clue ?
Wireshark is overly cautious, and sometimes throws spurious warnings.
> Something else of note which I hadn't seen before, usually things lock > up just after that first oops. For some reason, today it survived > a little longer, but things really went downhill fast. > It survived a 'dmesg ; scp dmesg davej@gelk', and then wedged solid. > So as well as the oops, it seems we're corrupting memory too. > For reference, this kernel has both SLUB_DEBUG and PAGEALLOC_DEBUG > enabled.
I haven't seen this kind of problem here with .26, but yes, it does look like something is clobbering memory during an NFS mount.
I introduced some NFS mount parsing changes in this commit range:
2d767432..82d101d5
A quick bisect should show which, if any of these, is the guilty party. If any of these are the problem, I suspect it's 3f8400d1.
-- Chuck Lever chuck[dot]lever[at]oracle[dot]com
| |