Bug 459360 - Strange /etc/sysconfig/autofs style
Strange /etc/sysconfig/autofs style
Status: CLOSED INSUFFICIENT_DATA
Product: Fedora
Classification: Fedora
Component: autofs (Show other bugs)
9
All Linux
medium Severity medium
: ---
: ---
Assigned To: Jeff Moyer
Fedora Extras Quality Assurance
:
Depends On:
Blocks:
  Show dependency treegraph
 
Reported: 2008-08-17 09:52 EDT by JW
Modified: 2008-12-08 11:27 EST (History)
1 user (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2008-12-08 11:27:36 EST
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---


Attachments (Terms of Use)

  None (edit)
Description JW 2008-08-17 09:52:51 EDT
Description of problem:
Normally entries in /etc/sysconfig are parameter files which apply to init.d scripts. But for autofs the sysconfig entry is parsed twice - once by init.d/autofs (which sources it, but only for the OPTIONS parameter) and again by automount daemon which has a hard-coded reference to the path.

It would make a lot more sense for there to be an /etc/autofs.conf (or even /etc/automount.conf, but there seems to be dual personality autofs/automount problem as well) and for the /etc/sysconfig/autofs to be reserved for setting invocation parameters only (including, say, as a "-c /etc/autofs.conf" option to automount).

It is also rather bad form for automount to not have a command-line option to set the location of its config file.

The automount, autofs, auto.master documentation is also quite terrible.

Also, there shouldn't ever be any need to make a distinction between "indirect" and "direct" mounts. And a map should support wildcard entries (eg dvd* ... :/dev/dvd$1) etc.

Version-Release number of selected component (if applicable):
autofs-5.0.2-16

How reproducible:
Always

Steps to Reproduce:
1. see description
  
Actual results:


Expected results:


Additional info
Comment 1 Jeff Moyer 2008-08-26 09:15:48 EDT
(In reply to comment #0)
> Description of problem:
> Normally entries in /etc/sysconfig are parameter files which apply to init.d
> scripts. But for autofs the sysconfig entry is parsed twice - once by
> init.d/autofs (which sources it, but only for the OPTIONS parameter) and again
> by automount daemon which has a hard-coded reference to the path.

Right.  autofs version 4 implemented the master map parsing in the init script, and that's why many of the configuration variables were placed there.  Version 5 moved to parsing the master map in the daemon, but maintained the same configuration file.  I agree that this is incorrect, and I feel that fixing this is a worthwhile endeavor.

> It would make a lot more sense for there to be an /etc/autofs.conf (or even
> /etc/automount.conf, but there seems to be dual personality autofs/automount
> problem as well)

Yes, I'm sure there is confusion over the terms.  autofs is the file system, automount is the daemon.  I think that we should simply live with the init script's unfortunate name, as changing it would cause more confusion that it is worth.

> and for the /etc/sysconfig/autofs to be reserved for setting
> invocation parameters only (including, say, as a "-c /etc/autofs.conf" option
> to automount).

Agreed.

> It is also rather bad form for automount to not have a command-line option to
> set the location of its config file.

Patches are accepted.  If you are not able or willing to provide patches, we can put this on the TODO list, but we currently have more important matters on hand (dealing with stabilization).

> The automount, autofs, auto.master documentation is also quite terrible.

Specific examples would be appreciated.  We're very interested in making the documentation more user-friendly, so your feedback would be very much welcome.

> Also, there shouldn't ever be any need to make a distinction between "indirect"
> and "direct" mounts. And a map should support wildcard entries (eg dvd* ...
> :/dev/dvd$1) etc.

The automounter was introduced in SunOS and has implementations in almost all flavors of UNIX.  The configuration of the automount maps is dictated by the defacto standard implementation.  We will not change the Linux implementation to depart in any incompatible way from the other implementations as maps are often shared between clients running different operating systems.
Comment 2 Ian Kent 2008-08-26 11:55:42 EDT
(In reply to comment #0)
> Also, there shouldn't ever be any need to make a distinction between "indirect"
> and "direct" mounts.

That's not sensible at all.
Direct and indirect mounts are very different.

Not only are the entries in the master map and the maps themselves
different they are handled very differently internally.
Comment 3 Bug Zapper 2008-11-26 06:06:18 EST
This message is a reminder that Fedora 8 is nearing its end of life.
Approximately 30 (thirty) days from now Fedora will stop maintaining
and issuing updates for Fedora 8.  It is Fedora's policy to close all
bug reports from releases that are no longer maintained.  At that time
this bug will be closed as WONTFIX if it remains open with a Fedora 
'version' of '8'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version' 
to a later Fedora version prior to Fedora 8's end of life.

Bug Reporter: Thank you for reporting this issue and we are sorry that 
we may not be able to fix it before Fedora 8 is end of life.  If you 
would still like to see this bug fixed and are able to reproduce it 
against a later version of Fedora please change the 'version' of this 
bug to the applicable version.  If you are unable to change the version, 
please add a comment here and someone will do it for you.

Although we aim to fix as many bugs as possible during every release's 
lifetime, sometimes those efforts are overtaken by events.  Often a 
more recent Fedora release includes newer upstream software that fixes 
bugs or makes them obsolete.

The process we are following is described here: 
http://fedoraproject.org/wiki/BugZappers/HouseKeeping
Comment 4 JW 2008-11-26 08:22:22 EST
I rest my case
Comment 5 Ian Kent 2008-11-26 09:14:13 EST
(In reply to comment #4)
> I rest my case

You rest your case?

The robot posted that message and your welcome to change the
product version above to keep this alive.

I'm aware of the need to separate these variables and it will
happen at some future time but probably not soon enough for your
liking. We don't need to have a Fedora bug open for something
that is an upstream issue and will be resolved upstream. Fact
is there are still higher priority things to do.

As far as direct and indirect mounts go, as I said, they are
distinctly different syntactically in maps and internally the
way they are handled so I don't see any sense in pretending they
are the same, since they aren't.
Comment 6 Ian Kent 2008-11-26 09:24:56 EST
(In reply to comment #0)
> 
> Also, there shouldn't ever be any need to make a distinction between "indirect"
> and "direct" mounts. And a map should support wildcard entries (eg dvd* ...
> :/dev/dvd$1) etc.

The wildcard suggestion is quite interesting but could easily cause
a lot of compatibility problems so it's quite hard to do. So you
won't see this in the 5.0 series.
Comment 7 Jeff Moyer 2008-11-26 10:09:05 EST
(In reply to comment #4)
> I rest my case

Really?  What exactly was your case?  You had a laundry list of questions, most of which were better suited for a mailing list discussion.  When asked to elucidate, you went silent.  Honestly, what did you expect?  Now, would you care to actually engage in discussion?  You can start by re-reading comment #1 and replying.

Cheers!
Jeff
Comment 8 JW 2008-11-26 21:51:50 EST
(In reply to comment #7)
> (In reply to comment #4)
> > I rest my case
> 
> Really?  What exactly was your case?  You had a laundry list of questions, most
> of which were better suited for a mailing list discussion.  When asked to
> elucidate, you went silent.  Honestly, what did you expect?  Now, would you
> care to actually engage in discussion?  You can start by re-reading comment #1
> and replying.
> 

So you don't even understand what "I rest my case" means!

I suggest you read what I reported, and ask somebody else what "I rest my case" might possibly mean. When they have explained what it means, then you will know what to expect.

Have a nice day.

Note You need to log in before you can comment on or make changes to this bug.