Re: Unable to kill processes in D-state

Re: Unable to kill processes in D-state

[Jeremy Sanders]

This sounds a similar problem (well it has similar symptoms) to what we
had with RedHat's kernel for (null) and their rawhide kernel. I think this
turned out to be an ext3 interaction bug.

See https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=70561

(also try ctrl+scrolllock to get a trace of the nfsds on your system).
This will provide useful info over where it is stuck.

Jeremy

-- 
Jeremy Sanders <jss@ast.cam.ac.uk>   http://www-xray.ast.cam.ac.uk/~jss/
X-Ray Group, Institute of Astronomy, University of Cambridge, UK.
Public Key Server PGP Key ID: E1AAE053





-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

Re: Re: Unable to kill processes in D-state

[Thomas Langås]

Jeremy Sanders:
> This sounds a similar problem (well it has similar symptoms) to what we
> had with RedHat's kernel for (null) and their rawhide kernel. I think this
> turned out to be an ext3 interaction bug.

We don't use ext3 on any of our systems...

> See https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=70561
> (also try ctrl+scrolllock to get a trace of the nfsds on your system).
> This will provide useful info over where it is stuck.

It's the client systems that have problems, how would it help to get a trace
of the nfsd's?

-- 
Thomas


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

Re: Re: Unable to kill processes in D-state

[Jeremy Sanders]

On Sun, 6 Oct 2002, Thomas Lang=E5s wrote:

> Jeremy Sanders:
> > This sounds a similar problem (well it has similar symptoms) to what =
we
> > had with RedHat's kernel for (null) and their rawhide kernel. I think=
 this
> > turned out to be an ext3 interaction bug.
>
> We don't use ext3 on any of our systems...

Okay so it's unlikely to be what I saw.

> It's the client systems that have problems, how would it help to get a =
trace
> of the nfsd's?

Probably not. It might be useful to get a trace of the processes stuck in
the 'D' state (ctrl+scroll lock is logged to dmesg), and rpciod, lockd
(and any of the rpc.xxx processes).

Also it might be useful to get a trace using tcpdump of the communication
between the server and client (options -vv -u -s 192 to get extra NFS
information).

Jeremy

PS I'm not an NFS expert, so please don't expect me to solve the problem!

--=20
Jeremy Sanders <jss@ast.cam.ac.uk>   http://www-xray.ast.cam.ac.uk/~jss/
X-Ray Group, Institute of Astronomy, University of Cambridge, UK.
Public Key Server PGP Key ID: E1AAE053






-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs