[Sussex] Mounting remote disks

[Jan Henkins]

Hello Brendan,

I'm not sure what the issue is with the NIS error messages, although I 
doubt that it is a fatal error. From a superficial look, I think that 
autofs is trying to get a gid/uid mapping from NIS (which is controlled 
via the /etc/nsswitch.conf file if memory serves), and upon not finding 
that resource it logs the failure. For NFS it would be a good idea to 
have a centralised UID/GID mapping service via something like LDAP 
(preferable over NIS), but in your case the mount is done via CIFS, 
which does not need such granular controls.

I will have a deeper look at the issue tonight.

Regards,
Jan Henkins

On 2013-10-14 11:04, Brendan Whelan wrote:
> Jan, Many thanks for your detailed response. I had started looking at
> the autofs method then hit problems so asked about fstab.
> Below are copies of the files that I am using for autofs and the
> results of trying to mount the remote drive. ( I can mount the remote
> drive from a shell script using "mount" which confirms the address,
> username, password, etc. ) As can be seen the problem that I have hit
> relates to "nisplus" and "nis+ tables". Searching on the Internet
> gives some confusing information. Any suggestions as to what I need 
> to
> set/change? Thanks, Brendan
>
> cat /etc/auto.master
> #
> # Sample auto.master file
> # This is an automounter map and it has the following format
> # key [ -mount-options-separated-by-comma ] location
> # For details of the format look at autofs(5).
> #
> /misc    /etc/auto.misc
> #
> # NOTE: mounts done from a hosts map will be mounted with the
> #    "nosuid" and "nodev" options unless the "suid" and "dev"
> #    options are explicitly given.
> #
> /net    -hosts
> #
> # 10 Sep 13 BW Added mount for shared area
> /mnt/bu_server /etc/auto.bu_server
> #
> # Include central master map if it can be found using
> # nsswitch sources.
> #
> # Note that if there are entries for /net or /misc (as
> # above) in the included master map any keys that are the
> # same will not be seen as the first read key seen takes
> # precedence.
> #
> +auto.master
>
> ----------------------------------------------------------
> cat /etc/auto.bu_server
> bu_server -fstype=cifs,rw,user=my_user,pass=my_password 
> ://10.1.101.104/bloodbm
>
> ----------------------------------------------------------
> chmod 755 /etc/auto.bu_server
>
> ----------------------------------------------------------
> service autofs restart
>
> ---------------------------------------------------------
> ls -ls /mnt/bu_server     - no files listed
>
> ---------------------------------------------------------
> /var/log/messages contains:
> automount[25594]: lookup_read_master: lookup(nisplus): couldn't
> locate nis+ table auto.master
>
> 
> --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> On 13/10/2013 22:03, Jan Henkins wrote:
>> Hello Brendan,
>>
>> On 13/10/13 16:40, Brendan Whelan wrote:
>>> I use cifs to mount a remote drive at a client's site. Currently, 
>>> the mount is part of a cron job and I would like to reduce the system 
>>> overhead of repeated unmount/mount commands. Could I simply add a 
>>> line in fstab?
>>
>> Yes, indeed you can.
>>
>>> Periodically, the remote drive will probably become inaccessible 
>>> because of network glitches or reboots of that system. Will fstab 
>>> take care of the remount when the remote drive becomes accessible 
>>> again? Do I need to include "auto"?
>>
>> All the keyword "auto" means that it will automatically mount the 
>> volume during bootup. This unfortunately does not mean that the volume 
>> will be remounted automatically if there was a problem. So instead of 
>> using an entry in fstab, a possible alternative solution would be to 
>> use autofs. Here are two relevant sites with info:
>>
>> 
>> http://www.howtoforge.com/accessing_windows_or_samba_shares_using_autofs
>> https://help.ubuntu.com/community/Autofs
>>
>> So instead of mounting things automatically during bootup, autofs 
>> only mounts the remote share as soon as it's accessed. That way if the 
>> mount-point disappears, it should connect properly the next time it's 
>> needed provided the remove share has been restored. This works well 
>> with NFS (which we use extensively at work), but since I haven't used 
>> autofs with CIFS I cannot vouch for it's suitability in practise. The 
>> documentation suggests that it's solid, so please test this properly 
>> before you trust it implicitly.
>>
>>> If I don't get the new line right first time in fstab then 
>>> presumable the current lines will continue to work?
>>
>> All the other entries in /etc/fstab will still work if one line is 
>> wrong.
>>
>> Hope this helps!
>>

-- 
Regards,
Jan Henkins