nfs timing out during boot on 11.3

I just upgraded Suse using zypper dup and have run into problems during boot. While the system will boot properly the boot process now takes about 15 minutes to complete. The issue is that during boot its trying to load the nfs mounts but these fail showing connection timeout (which appears to be 3 minutes each mount).


   948  <notice -- Aug 23 21:50:28.454938000> service boot.apparmor donedone
   949  done
   950  System Boot Control: The system has been set up
   951  Skipped features: boot.open-iscsi boot.cycle
   952  System Boot Control: Running /etc/init.d/boot.local
   953  mount.nfs: Connection timed out
   954  mount.nfs: Connection timed out
   955  mount.nfs: Connection timed out
   956  mount.nfs: Connection timed out
   957  mount.nfs: Connection timed out
   958  failed
   959  <notice -- Aug 23 22:05:28.760584000> killproc: kill(447,3)

This did not occur in the boot sequence before I upgraded 11.2 to 11.3. Also, interestingly the mounts are there and usable once the system boots into KDE. Is there a timing issue in the boot process now that causes this? What can I modify to avoid this hang in the boot process in the future?

Sample entry in fstab for one of these mounts is the following

baalnfs:/storage2/apps /nfs/apps nfs defaults 0 0

All the nfs mounts come from the same server (which was up and running when boot occured). My nfs server is running as a headless 11.2 command line system (will upgrade it in a week or two to 11.3 when I get my new drives for it).

Looks like you have the same problem in this thread:
openSUSE 11.3 NFS problem (hang) during boot process

There’s a reply at the end which solved this matter for my friend flark.

Thanks switching to version 3 resolved the problem of it taking 3 minutes to timeout for each share, but doesnt really fix the underlying issue in that the drives are mounted right before the network is enabled.

Also I now get other failure messages on nfs at boot, but once again doesnt impact me being able to use the drives once the system is completely loaded.

New Errors

   928  mount.nfs: rpc.statd is not running but is required for remote locking.
   929  mount.nfs: Either use '-o nolock' to keep locks local, or start statd.
   930  mount.nfs: an incorrect mount option was specified
   931  mount.nfs: rpc.statd is not running but is required for remote locking.
   932  mount.nfs: Either use '-o nolock' to keep locks local, or start statd.
   933  mount.nfs: an incorrect mount option was specified
   934  mount.nfs: rpc.statd is not running but is required for remote locking.
   935  mount.nfs: Either use '-o nolock' to keep locks local, or start statd.
   936  mount.nfs: an incorrect mount option was specified
   937  mount.nfs: rpc.statd is not running but is required for remote locking.
   938  mount.nfs: Either use '-o nolock' to keep locks local, or start statd.
   939  mount.nfs: an incorrect mount option was specified
   940  mount.nfs: rpc.statd is not running but is required for remote locking.
   941  mount.nfs: Either use '-o nolock' to keep locks local, or start statd.
   942  mount.nfs: an incorrect mount option was specified

On 08/24/2010 04:36 PM, baaldemon wrote:
> Thanks switching to version 3 resolved the problem of it taking 3
> minutes to timeout for each share, but doesnt really fix the underlying
> issue in that the drives are mounted right before the network is
> enabled.
>
> Also I now get other failure messages on nfs at boot, but once again
> doesnt impact me being able to use the drives once the system is
> completely loaded.

Find the mount statements in /etc/init.d/nfs and have them run in the
background. On my systems, adding the & at the end of the lines below fixed the
problem:

mount -o nolock $where &

and

mount -at nfs,nfs4 > /dev/null 2>&1 &