|
Home > Archive > Red Hat Topics > March 2005 > Can't mount nfs filesystems after this mornings FC3 updates
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Can't mount nfs filesystems after this mornings FC3 updates
|
|
| Rincewind 2005-03-31, 8:45 pm |
| I updated the following packages this morning, using up2date:
[Thu Mar 31 10:40:29 2005] up2date installing packages:
['ImageMagick-6.2.0.7-2.fc3', 'gdk-pixbuf-0.22.0-16.fc3',
'gdk-pixbuf-devel-0.22.0-16.fc3', 'gdk-pixbuf-gnome-0.22.0-16.fc3',
'gtk2-2.4.14-3.fc3', 'gtk2-devel-2.4.14-3.fc3', 'initscripts-7.93.7-1',
'util-linux-2.12a-21']
Before updating, I had a couple of nfs filesystems mounted. They are set
to mount automatically from /etc/fstab.
When I rebooted after the updates, I got the following for the nfs mounts:
Mar 31 15:34:24 eden mount: mount to NFS server '192.168.2.3' failed: server is down.
Mar 31 15:34:24 eden mount: mount to NFS server '192.168.2.3' failed: server is down.
I got this both when shutting down and while booting. Attempting to mount
them manually does not work either:
[root@eden ~]# mount -t nfs 192.168.2.3:/ /gateway
mount to NFS server '192.168.2.3' failed: server is down.
So I checked to see where the problem was:
[root@eden ~]# ssh -l******** 192.168.2.3
********@192.168.2.3's password:
[********@gateway *******]$ mkdir tmp
[********@gateway *******]$ mount -t nfs 192/168.2.3:/home/dave tmp
mount: only root can do that
[*******@gateway ********]$ su
Password:
[root@gateway ********]# mount -t nfs 192.168.2.3:/home/dave tmp
[root@gateway ********]# ls tmp
[root@gateway ********]# ls tmp
ash bin ...........................<snip>
[root@gateway ********]# umount tmp
[root@gateway ********]# ls tmp
[root@gateway ********]#
[root@gateway ********]# exit
So the server is working(as expected - nothing has been changed). It must
be this mornings updates(presumably to util-linx) that have broken the
client(mount?).
What is the easiest way to roll back this package?
--
Rinso
/\
/ \
/wizz\
~~~~~~~~~~~~
| |
| Rincewind 2005-03-31, 8:45 pm |
| On Thu, 31 Mar 2005 15:29:43 +0000, Rincewind mumbled something like this:
> I updated the following packages this morning, using up2date:
>
> [Thu Mar 31 10:40:29 2005] up2date installing packages:
> ['ImageMagick-6.2.0.7-2.fc3', 'gdk-pixbuf-0.22.0-16.fc3',
> 'gdk-pixbuf-devel-0.22.0-16.fc3', 'gdk-pixbuf-gnome-0.22.0-16.fc3',
> 'gtk2-2.4.14-3.fc3', 'gtk2-devel-2.4.14-3.fc3', 'initscripts-7.93.7-1',
> 'util-linux-2.12a-21']
>
> Before updating, I had a couple of nfs filesystems mounted. They are set
> to mount automatically from /etc/fstab.
>
> When I rebooted after the updates, I got the following for the nfs mounts:
>
> Mar 31 15:34:24 eden mount: mount to NFS server '192.168.2.3' failed:
> server is down. Mar 31 15:34:24 eden mount: mount to NFS server
> '192.168.2.3' failed: server is down.
>
<snip>
>
> So the server is working(as expected - nothing has been changed). It must
> be this mornings updates(presumably to util-linx) that have broken the
> client(mount?).
It does appear to be that the default behaviour of mount has changed.
I only have nfs listening to UDP on the server. Previously the default
behaviour must have tried UDP after failing on TCP(presumably the default?).
I have cured the problem by adding udp to the options in the /etc/fstab
entries for the mounts(on the client machine).
--
Rinso
/\
/ \
/wizz\
~~~~~~~~~~~~
|
|
|
|
|