> I've got a Linux box running the SGI-modified RedHat 7.1. Kernel is
> 2.4.3-SGI_XFS_1.0.1smp. I'm exporting one of the xfs partitions via NFS, and
> trying to mount it on an HP/UX 11.00 machine.
> The HP machine is able to mount nfs-exported ext2 partitions from another
> identical Linux box w/o a problem. However, when I try to mount a
> nfs-exported xfs partition, it is inaccessible after mounting. No error is
> returned while mounting, and I don't see an error in the Linux box's log.
> # mount -F nfs -o hard,intr,rw tux:/u/rd/mga /mnt
> # ls mnt
> mnt unreadable
> # df
> df: open of tux:/u/rd/mga failed
> /home (/dev/vg00/lvol5 ): 36336 blocks 4840 i-nodes
> /opt (/dev/vg00/lvol6 ): 251108 blocks 33466 i-nodes
> /tmp (/dev/vg00/lvol4 ): 96272 blocks 12819 i-nodes
> /u (/dev/vg01/lvol1 ): 4007190 blocks 534166 i-nodes
> /usr (/dev/vg00/lvol7 ): 130354 blocks 17297 i-nodes
> /var (/dev/vg00/lvol8 ): 412596 blocks 54813 i-nodes
> /stand (/dev/vg00/lvol1 ): 99986 blocks 13404 i-nodes
> / (/dev/vg00/lvol3 ): 175136 blocks 23344 i-nodes
> I am able to mount and access the partition if I specify vers=2 as one of the
> mount options. However, then I see the following behaviour: when any user
> tries to log into the HP machine, the login hangs for almost 1 minute. If I
> do a 'ps -elf' while somebody's logging in, I see that /usr/bin/quota's
> running for that user, and presumably that's what's hanging.
> Anybody have any ideas/suggestions?
Anything interesting from tcpdump when you do the df?
tcpdump -s 300 ip host <client> and <server>