Hello NFS hackers,
I'm running into a problem with NFS stale file handles... I seem to get
them even after just stopping and restarting NFS services on our file
server.
The server is running RedHat 7.2 and kernel 2.4.7-10 (no SMP support).
Clients include some RedHat 7.1 through 7.3 machines, and Solaris 2.6
through 8 machines.
My first suspect is the fairly ancient kernel. Are there known issues
with kernels from that era? I'd like to upgrade kernels but am kind of
stuck -- this is a Dell server, we like being able to monitor the
hardware, and last I heard they didn't support OpenManage on newer
kernels.
If there aren't problems with these kernels and there is a possible
work-around ... am I doing something wrong? My /etc/exports contains only
/export *(rw,sync,wdelay,no_subtree_check)
I'm not exporting anything else (especially subdirectories), but clients
may mount subdirectories (for example /export/home as /home). I set
no_subtree_check in the hopes that it would eliminate any state from the
file handles.
Thanks in advance for any info,
-Joel
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs
I think I figured this out. It turns out that while stopping and
restarting NFS services I also deleted /var/lib/nfs/xtab . Don't ask
how... it was pilot error on my part.
-Joel
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs