lkml.org 
[lkml]   [1997]   [Dec]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: RPC and Kernel NFS issues


At [Wed, 17 Dec 1997 15:07:16 -0500]
Bill Hawes <whawes@star.net> wrote:

> Christoph Lameter wrote:
> > I am running a Debian Hamm system (glibc) with the newest tools etc and I
> > am trying to test 2.1.72 in a production environment after having run
> > 2.1.X for awhile as a workstation.
> >
> > Some of the issues:
> >
> > 1. I keep getting strange RPC messages on bootup although I am using
> > userland NFS. (Kernel NFS compiled it but not used)
> >
> > Dec 16 14:48:40 deborah kernel: portmap: RPC call returned error 111
> > Dec 16 14:48:40 deborah kernel: RPC: task of released request still
> > queued!
> > Dec 16 14:48:40 deborah kernel: RPC: (task is on xprt_pending)
> > Dec 16 14:48:40 deborah kernel: portmap: RPC call returned error 111
>
> A number of people have reported this problem, and it seems to be caused
> by the portmapper being started too late. Error 111 is "connection
> refused". Are your startup scripts trying to bring up NFS too early?
>

The debian startup scripts try to mount the remote filesystems before
starting the portmapper.

I solved this one by removing the 'mount -a -t nfs' in /etc/init.d/boot
and placing it after rpc.portmap is started in /etc/init.d/netstd_nfs

I assume this is because mount is starting up rpc.lockd as the 'nolock'
option hasn't been used.

Hope this helps (and hope Debian get the scripts sorted before 2.2 is
released),

Adrian.

--
Adrian St. John - Linux Mailing List account
linux@oot.co.uk


\
 
 \ /
  Last update: 2005-03-22 13:40    [from the cache]
©2003-2014 Jasper Spaans. Advertise on this site