On Tue, 2015-02-03 at 15:47 +0100, Jakub Hrozek wrote:
On Tue, Feb 03, 2015 at 09:21:10AM -0500, Brendan Kearney wrote:
On Mon, 2015-02-02 at 14:08 +0100, Jakub Hrozek wrote:
On Mon, Feb 02, 2015 at 06:50:12AM -0500, Dmitri Pal wrote:
On 02/02/2015 01:52 AM, Jakub Hrozek wrote:
On Sun, Feb 01, 2015 at 11:17:02AM -0500, Brendan Kearney wrote:
On Sat, 2015-01-31 at 19:01 -0500, Dmitri Pal wrote: >On 01/31/2015 12:23 PM, Brendan Kearney wrote: >>i want to have an automount configured so that >>nas.bpk2.com:/export/music is mounted on /home/<username>/Music. i want >>to use the HOME variable, but several attempts have not gotten the job >>done. i have tried $HOME/Music, ${HOME}/Music, and {$HOME}/Music, and >>none have worked. >> >>what is the correct way to use a variable with AutoFS and SSSD? i have >>used the hard values to get things working for my id, but want to have >>this work for all users. >> >>dn: >>automountKey=/home/brendan/Music,automountMapName=auto.shares,cn=autofs,ou=Daemons,dc=bpk2,dc=com >>automountinformation: nas.bpk2.com:/export/music >>automountkey: /home/brendan/Music >>description: music >>objectclass: automount >>objectclass: top
This is a direct mount map entry. You can't use the wildcard key in a direct mount key.
Oddly enough I suspect you could use the & substitution in the mount location but that's generally not useful because it would replace the whole path (the entire key) at it's occurrence in the mount location.
You can't use macros in lookup keys with them (or indirect mounts) either because the path is used as the lookup key.
I know you think that using the information about the requesting could be used but it's not as simple as that.
The problem is that multiple processes could request a mount at the same time, resulting in a queue waiting for mount completion, and the user information that is returned will be that of the process that originally triggered the mount and so might not be what you'd expect. There's no straight forward way to fix that.
That's why macro substitution isn't done on lookup keys and can't really be done safely.
You can do tricks with a combination of indirect mounts, sub-mounts and the wildcard map entry (ie. & substitution).
I'd need some time to come up with an example of how to implement something like what your after. Would you like me to do that?
The amd map entry format, while somewhat harder to use, does allow macro substitution in keys, including in the implementation I've recently added to autofs. In fact, if you want hard to maintain, complicated maps, that do nifty things it might be worth looking at that map format.
>> >>thanks, >> >>brendan >> >>_______________________________________________ >>sssd-users mailing list >>sssd-users@lists.fedorahosted.org >>https://lists.fedorahosted.org/mailman/listinfo/sssd-users >I am not sure it is possible. Please file an RFE. > umm... how? i dont see a sign-in link in the trac page. do you have a link?
This would be an autofs RFE right? :-)
No it is more IPA and SSSD one(s). IPA to store the keys with macro and SSSD to expand the macro
No?
My understanding is that the expansion is done by the automounter deamon. IIRC there is something like "&" for homedirs. I would advise to ask on autofs lists first.
Please note I'm not saying we don't have any bugs in sssd, chances are we do, but so far I think we sholdn't touch the variables.. _______________________________________________ sssd-users mailing list sssd-users@lists.fedorahosted.org https://lists.fedorahosted.org/mailman/listinfo/sssd-users
so where does this leave me? i have a SSSD with AutoFS implementation that does not provide all the functionality i want, and indecision around who should be doing what leaves me no option to request the features
I've CC-ed the RHEL/Fedora autofs maintainer..
Ian, could you help us out with this thread? I'm afraid our autofs-fu is not strong enough. Here is a link to the archives so you don't have to hunt the thread in the archive: https://lists.fedorahosted.org/pipermail/sssd-users/2015-January/002611.html
Thank you very much for your time.