Re: OOo 1.1 in sid with rootfs mounted readonly via NFS

[Chris Halls <halls@debian.org>]

On Wed, 2004-02-25 at 22:32, Gavin Hamill wrote:
> Hm, let's see...
> 
> 'Make sure NFS lockd is running on the client, or mount the NFS export which 
> contains /usr with the 'nolock' option.
> 
> If /usr gets mounted from an initrd, check 'ps' output once the system is 
> fully booted to ensure that you see a [lockd] kernel process, or an 
> 'rpc.lockd' userspace process. 
> 
> If not, just issue your mount command again (no need to even -o remount it 
> seems!) and run /etc/init.d/nfs-common restart.

Thanks.  Now, I need to describe this in such a way that other people
will find it ;)

Re-reading the beginning of the thread, it seems you were seeing
something like this?

  running openoffice.org setup...
  setup failed.. abort

Is that right?  If so, please can you find out which exit code setup was
producing - if the code is unique we can try checking for that in the
wrapper.

You can get the exit code by modifying /usr/bin/openoffice:

Change the echo line at line 247 from:
        echo "setup failed.. abort" >&2
to:
        echo "setup failed with code $?.. abort" >&2

Hmm, might be a good idea to do this in the packages too..

Chris