On Fri, Feb 13, 2004 at 12:45:55PM -0800, Erik E. Fair wrote:
> 
> When it shows up, NFS will begin to fail, and the system must be 
> rebooted to make everything happy again. I tend to see it after 
> running for about six months or so. I also believe it's a namei cache 
> issue of some kind.

I can reproduce _some_ kind of namei cache problem with my metadata
torture tests I'm using to work on the buffer cache code.  I have a
client machine that runs diskless from my fileserver that I typically
use for testing.  If the client tries to page in a large executable
while the tests are running, sometimes the client hangs forever and
*other* processes on the fileserver end up sleeping in the namei
lookup path.

I haven't had time to track it down further than that.  But something
is definitely wrong.

-- 
 Thor Lancelot Simon	                                      tls@rek.tjls.com
   But as he knew no bad language, he had called him all the names of common
 objects that he could think of, and had screamed: "You lamp!  You towel!  You
 plate!" and so on.              --Sigmund Freud