How do executable maps work? For now I'm using a perl script that does more
or less what you described (but it also makes symbolic links.) This kind of
hack (even if it's not pretty) would be nice placeholder for real direct map
support in autofs.
Question to the developers: why is it such a hack? Is it a limitation of
how Linux handles file systems, or is it the code itself? (Or a combination
of both?) I'm just curious.
A feature that might be more useful is the ability to see the contents of
the automounts. I occasionally get asked why auto mounting isn't working
and I have to explain that it's working fine, just invisibly.
Anyways, thank you everyone for your feedback!
Geoff
-----Original Message-----
From: Jochen Reinwand [mailto:[EMAIL PROTECTED]]
Sent: 21 June 2001 6:15
To: Dirk Wetter; Stephen Walton
Cc: Oakham, Geoff; '[EMAIL PROTECTED]'
Subject: Re: Direct Mounts
On Wednesday, 20. June 2001 23:36, Dirk Wetter wrote:
> hi,
>
> sure, that works. but if you're living in a non-100% linux
> environment you cannot use the direct maps distributed via
> NIS(+)/ldap, which would be great.
A work around would be an Executable Map. Make a bash script that converts
your NIS(+)/ldap entries to something Linux can understands.
Also the syntax for a submount differs, e.g. between Solaris and Linux. This
Problem can also be solved by an Executable Map.
> i read in the TODO for v5 that there might be - there was
> a question mark - direct maps? when would that be ?
Since autofs 4 still isn't in a stable state, I think you'll have to wait a
long time for version 5. And then there's still the question mark...
BTW. Does anybody know when autofs 4 will get to a stable state?
Jochen
