Puppet users - Mar 2011 - Changing NFS mount options to read only (ro) does not affect puppet change

I have some NFS directories that get changed to a read-only status
after a period of time.   I noticed in puppet, if I change these
values and restart puppet and puppetmaster respectively, it does not
pick up the change in the mount {} directive.

Is there a way around this, or a better way to affect the change.


Thanks.

-- 
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.

On Tue, Mar 29, 2011 at 11:03:33AM -0700, Forrie wrote:
> I have some NFS directories that get changed to a read-only status
> after a period of time.   I noticed in puppet, if I change these
> values and restart puppet and puppetmaster respectively, it does not
> pick up the change in the mount {} directive.
> 
> Is there a way around this, or a better way to affect the change.
> 
> 
> Thanks.
> 
The current mount type of puppet just checks if you have a correct entry
in fstab (or vfstab) and if the device is mounted or not.

This means that if your fstab is insync and the device is mounted, puppet
will not change anything. So if you want options to be rw and you have
rw in fstab, puppet will not complain even when the actual mount is
mounted ro.

Puppet will only trigger a remount if fstab is out of sync.

This is a known issue: http://projects.puppetlabs.com/issues/6309

-Stefan

Thanks for this information; this a somewhat a frustrating issue.

IMHO, as per configuration management, if the mount options change,
then Puppet should "do the right thing" and umount and remount with
the proper options.  One problem there is what if someone is in the
directory mount and puppet can''t unmount?  Does it retry later or log
an error; also, any issues that might affect running dependencies.
I still think it should change it.

How does puppet handle a Stale NFS Mount? (if at all)   That occurred
recently and we have to manually intervene.


Thanks!

-- 
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.

On Mar 29, 8:00 pm, Forrie <for...@gmail.com>
wrote:
> Thanks for this information; this a somewhat a frustrating issue.
>
> IMHO, as per configuration management, if the mount options change,
> then Puppet should "do the right thing" and umount and remount
with
> the proper options.  One problem there is what if someone is in the
> directory mount and puppet can''t unmount?  Does it retry later or
log
> an error; also, any issues that might affect running dependencies.
> I still think it should change it.
There is a recent thread here that is related to this topic.  See
http://groups.google.com/group/puppet-users/browse_thread/thread/2d30b8e48d49f8a6/7452f78e33910ffc?lnk=gst&q=mount+fstab#7452f78e33910ffc

> How does puppet handle a Stale NFS Mount? (if at all)   That occurred
> recently and we have to manually intervene.
I can''t think of a mechanism by which Puppet might check whether a
filesystem is mounted that will notice stale mounts, so my expectation
would be that Puppet doesn''t notice them.  Puppetlabs would probably
be willing to entertain a feature request, however, especially if you
can suggest a suitable implementation strategy.


John

-- 
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.