Hello Stefan Schwarzer!

I'm looking over (old) util-linux bug reports to try to identify
which ones are still relevant and which has already been fixed
in newer upstream releases...

On Thu, Jul 26, 2012 at 09:59:24AM +0200, Stefan Schwarzer wrote:
> Package: mount
> Version: 2.20.1-5.1
> Severity: normal
> 
> Dear Maintainer,
> 
> I have several entries for cifs shares in /etc/fstab, each marked as 
> 'user' mountable. I expect that 'mount -at cifs' scans through fstab and
> attempts to mount each of these, either deciding to perform the mount 
> by itself or by delegating to mount.cifs.
> 
> What happens is that the mount command is rejected with the error
> message 
> 
> sts@nbofm020:~$ mount -at cifs
> mount: only root can do that

With util-linux 2.24.2 the error message is a bit more clear:

$ mount -at cifs
mount: only root can use "--all" option

Even though in your case being able to use the -a option might seem
convenient, I can't come up with a good general semantics on how
it should behave in all cases.
If you think you can come up with a way to describe how you'd
like to see it work, I'd recommend you to argue it on the
upstream mailing list.

> 
> though each single filesystem is mounted correctly
> 
> sts@nbofm020:~$ mount /ipmfas/ofm-projekte
> sts@nbofm020:~$ df
> Filesystem                                              1K-blocks       Used 
> Available Use% Mounted on
> [...]
> //ipmfas/ofm-projekte                                  2732431772 2628478316 
> 103953456  97% /ipmfas/ofm-projekte
> sts@nbofm020:~$ 
> 
> In fact, 'mount -at cifs' as user does not even reference /etc/fstab as
> can be seen by running strace, so it seems to link the -a option to 
> a mandatory uid/euid of 0 at an early stage. 
> 
> By the way, umount also does not correctly honor 'user', since from the 
> same shell a subsequent 
> 
> sts@nbofm020:~$ umount /ipmfas/ofm-projekte 
> umount: only root can unmount //ipmfas/ofm-projekte from /ipmfas/ofm-projekte
> 
> fails. 

I guess this is because of bug #702935 and should hopefully be resolved
in newer upstream releases.

Do you agree with my assessment that this bug can be closed once we get
the new upstream release in the archive?

Regards,
Andreas Henriksson


-- 
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]

Reply via email to