openSUSE 11.1 not mounting NFS at boot

I just installed openSUSE 11.1 on one of the computers here at work. One of the problems I am having is that NFS is not mounting at boot.

fstab /atlas nfs noatime,nosuid,rsize=32768,wsize=32768,soft,nolock,rw 0 0

I know from the past that there has been a change in the way that openSUSE handels NFS in the fstab. In 10.2 all I had to do was put the above entry in the fstab and it was mounted at boot. Starting in 10.3 that no longer worked. I had to use the NFS Client in yast2 to mount the NFS share, and then it would mount at boot. openSUSE 11.0 worked like 10.3.

Now on this new 11.1 the NFS share does not mount at boot. Mount /atlas after boot does work. I do not see any mount or NFS errors in /var/log/messages.

Is anyone else having problems with NFS during boot? Thanks.

Is the “nfs” service enabled? Also the “rpcbind” service? (Used to be called portmap.)

Yes they are enabled. If they were not then the manual mount would fail to.

Well no rpcbind would prevent manual mounting, but if “nfs” is not enabled, then it simply doesn’t mount any NFS partitions at boot. If it is but it can’t find any nfs lines in fstab, then it prints:

Not starting NFS client services - no NFS found in /etc/fstab.

/var/log/boot.msg contains a log of all the services that were started. Also by watching the console (you may have to hit ESC and later Ctrl-Alt-F1 to get away from the GUI VT) you should see whether nfs ran and if it did, what was the reason it failed.

Because mount /atlas works, both nfs and rpc are working. So this is a problem in the boot process. I have been using the ip address for my nfs fstab entries because other distro’s have a problem with name lookup during nfs boot mounts.

grep -i nfs boot.msg
Starting NFS client services:<notice>checkproc: /sbin/rpcbind 2696

grep -i atlas boot.msg

grep -i mount boot.msg
<4>Mount-cache hash table entries: 256
<5>XFS mounting filesystem sda1
<7>Ending clean XFS mount for filesystem: sda1
<5>XFS mounting filesystem sda5
<7>Ending clean XFS mount for filesystem: sda5
fsck succeeded. Mounting root device read-write.
Mounting root /dev/disk/by-id/ata-ST3120814A_4LS4N7MJ-part1
Mounting local file systems…
Mounting fuse control filesystemdone
Mounting securityfs on /sys/kernel/security done
Mount CIFS File Systems unused
<notice>startproc: execve (/usr/sbin/automount) /usr/sbin/automount -p /var/run/ ], CONSOLE=/dev/console ROOTFS_FSTYPE=xfs SHELL=/bin/sh TERM=linux ROOTFS_FSCK=0 LC_ALL=POSIX INIT_VERSION=sysvinit-2.86 REDIRECT=/dev/tty1 COLUMNS=156 PATH=/sbin:/usr/sbin:/bin:/usr/bin vga=0x31a DO_CONFIRM= RUNLEVEL=5 PWD=/ SPLASHCFG=/etc/bootsplash/themes/openSUSE/config/bootsplash-1280x1024.cfg PREVLEVEL=N LINES=60 HOME=/ SHLVL=2 splash=silent SPLASH=yes ROOTFS_BLKDEV=/dev/disk/by-id/ata-ST3120814A_4LS4N7MJ-part1 _=/sbin/startproc DAEMON=/usr/sbin/automount ]
Starting automount done

The automount is for the users /home, which is working fine. Home directories are on

There is a suggestion in /etc/init.d/nfs that if run too soon after the network comes up, it may not work. I regard the nfs script as a bit unsatisfactory since it doesn’t try to deal with the network state.

I’m having the same problem. When my SuSE 11 client computer boots up, it does not read all of /etc/fstab, which has an entry to mount an nfs volume (SuSe 11 Server). This is the same situation described by dwestf in this post. Has this been resolved?

After booting up, I can issue a ‘mount -a’ command, and it also mounts successfully (as dwestf explained).

I have a Fedora Core 4 cluster with the same setup, except it does mount the nfs volume correctly at boot up.

This thread, NFS mount works manually, but not during boot up … why? - openSUSE Forums
suggests that the NFS client service is not on. I checked mine and it is “enabled”. It also suggests that I use yast2 to make the fstab entry instead of doing in manually. I tried both and it will still not mount at boot.

This thread, CIFS Mounts Ignored in fstab - openSUSE Forums
suggests making a startup script with a delay with the ‘mount -a’ command so that after the network has been established, it will re-issue the mount command. I believe this is what ken_yap has suggested (although he didn’t like it).

If so, is the script simple, like:
#! /bin/bash
mount -a

How do I add the delay and where do I put this startup script? In Fedora I put them in /etc/rc3.d/
However, this directory doesn’t exist in SuSE

I’m running everything in root by the way.

Thanks for the help.

Here is the excerpt from /etc/init.d/nfs that I think ken_yap was talking about.

# Mount all auto NFS devices (-&gt; nfs(5) and mount(8) )
#  NFS-Server sometime not reachable during boot phase.
#  It's sometime usefull to mount NFS devices in
#  background with an ampersand (&) and a sleep time of
#  two or more seconds, e.g:
#   sleep 2 && mount -at nfs,nfs4 &
#   sleep 2 
if test "$nfs" = yes ; then
        mount -at nfs,nfs4 &gt; /dev/null 2&gt;&1

To add the delay, do I just change the command:

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


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

The computers are in the middle of simulations so I can’t restart them right now to test this out.

I had the same problem. That’s how I solved it:
First, I changed the mount line as suggested above; note that there is a typo, correctly it should read (two lines):

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

sleep 2

Second, I had to change the network setting from DHCP to manual configuration. Seems that with DHCP, the network is not yet up and running when nfs is started. Maybe it is enough to increase the sleep 2 to sleep 5 or something (and keep DHCP turned on) – or to turn off DHCP and keep the nfs mount line :wink:


I am having the same problem - but with SLES 10 sp2…
But adding sleep doesnt help me… :\

until nfs requests are completed i dont get reply back for ping!! :open_mouth: so i have to login and mount manually…

also, i noticed that keeping the firewall disabled makes the nfs mount @ boot! without the necessity to mount manually after loggin!! :’(

Im a bit out of my depth, grateful for pointers

Last night I updated the system as there were some security patches (red triangle in systray), Subsequently, I also updated some KDE 4.2.1 stuff from the repositories, but I don’t think that is relevant.

It’s just a desktop system, with broadband, nothing amazing.

This morning, the boot sequence chunters away until it gets to

starting rpcbind

then it reports “no NFS found in fstab”

after that it seems to continue on as normal, e.g., run level 5 seemed to have been reached, but instead of starting X, it goes to a command line login prompt

after logging in, it’s unimpressed by startx

May I ask for some slow class help, please?