Can you test the pre-129-2 that I just made? It fixes at least one fairly
embarrassing bug I introduced in the NFS code (and never showed on my
machine because I have ridiculous amounts of memory available). It may not
be the cause of the particular problems you see, but I'd like to know...
Anyway, if you still see the problems with pre-129-2, could you also send
me a tcpdump of the failing transactions? Something like
/usr/sbin/tcpdump -s 512 host <nfs-server> and host <nfs-client>
would help.
(the -s is there because otherwise tcpdump won't get enough of the packet
to tell what certain NFS transactions are all about)
I've verified that it works for me even with a smaller wsize, so it has to
be something that isn't completely obvious.
Finally, can you check whether it is at all rsize-related? I don't think
it should be, but just in case.. Does "rsize=1024,wsize=8192" work for
you, for example (and how about "rsize=8192,wsize=1024"?).
Thanks,
Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/