2.6.0-test6 NFS?

2.6.0-test6 NFS?

[Robert L. Harris]

[-- Attachment #1: Type: text/plain, Size: 1094 bytes --]



  I've been trying to run my NFS server on 2.6 kernels for a while now.
My desktop and my Firewall are both 2.6 already and happy.  My
fileserver though is giving me an ulcer.

  I have nfs server compiled into the kernel and the client as a module.
I've tried both with and without NFSv4 (with NFSv3 definitely).  No
matter what I try though my client machines keep getting stale
filehandles.  I can cd ~ and do an "ls -la", if I get much more complex
and try something such as 

cd ~/.spamassasin 
ls -la
ls -la

  I get the stale handles.  Reverting back to 2.4 and all is well on the
same export and mount configs.

Thoughts?

:wq!
---------------------------------------------------------------------------
Robert L. Harris                     | GPG Key ID: E344DA3B
                                         @ x-hkp://pgp.mit.edu
DISCLAIMER:
      These are MY OPINIONS ALONE.  I speak for no-one else.

Life is not a destination, it's a journey.
  Microsoft produces 15 car pileups on the highway.
    Don't stop traffic to stand and gawk at the tragedy.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: 2.6.0-test6 NFS?

[Patrick Mau]

On Wed, Oct 08, 2003 at 06:49:03PM -0400, Robert L. Harris wrote:
> 
> 
>   I've been trying to run my NFS server on 2.6 kernels for a while now.
> My desktop and my Firewall are both 2.6 already and happy.  My
> fileserver though is giving me an ulcer.

[...]

>   I get the stale handles.  Reverting back to 2.4 and all is well on the
> same export and mount configs.
> 
> Thoughts?

Hallo Robert,

you could search the archives for "NFS Problem" and "STALE", you
will probably find my post near end of September. The short solution:

Use "no_subtree_check" in /etc/exports like this:

/home \
  tony.local.net(rw,sync,no_root_squash,no_subtree_check)

I think I have found a bug in the nfs server code that always
returns a failure on subtree checks. I described my findings in
a post to this list, but nobody answered.

Patrick