Re: NFS regression? Odd delays and lockups accessing an NFS export.

[Roger Heflin <rogerheflin@gmail.com>]

Ian Campbell wrote:
> (added some quoting from previous mail to save replying twice)
> 
> On Sun, 2008-08-24 at 15:19 -0400, Trond Myklebust wrote:
>> On Sun, 2008-08-24 at 15:17 -0400, Trond Myklebust wrote:
>>> >From the tcpdump, it looks as if the NFS server is failing to close the
>>> socket, when the client closes its side. You therefore end up getting
>>> stuck in the FIN_WAIT2 state (as netstat clearly shows above).
>>>
>>> Is the server keeping the client in this state for a very long
>>> period?
> 
> Well, it had been around an hour and a half on this occasion. Next time
> it happens I can wait longer but I'm pretty sure I've come back from
> time away and it's been wedged for at least a day. How long would you
> expect it to remain in this state for?
> 
>> BTW: the RPC client is closing the socket because it detected no NFS
>> activity for 5 minutes. Did you expect any NFS activity during this
>> time?
> 
> It's a mythtv box so at times where no one is watching anything and
> there isn't anything to record I expect NFS activity is pretty minimal.
> 
> Ian.
> 

Ian,

Do you have a recording group setup on the NFS partition that mythtv is
going to be accessing?

I have seen similar funny stuff happen, it used to happened around 2.6.22*
(on each end), and quit happening around 2.6.24* and now has started happening
again with 2.6.25* on both ends.

Similar to what you have the only thing I see is "NFS server not responding" and 
  restarting the NFS server end (/etc/init.d/nfs restart) appears to get things to
continue on the NFS client.    No other messages appear on either end that 
indicate that anything is wrong, other non-nfs partitions on the client work 
find, the machine is still up, and the NFS server is still up and fine, and 
after a restart things will again work for a while (hours or days).

                             Roger