[Adduser-devel] (fwd) Re: Bug#330929: how can adduser reliably find out wheter nscd is running? [madcoder@debian.org]

Marc Haber mh+adduser-devel at zugschlus.de
Sun Apr 15 11:06:09 UTC 2007


Hi,

this is from #330929 and might be helpful for adduser.

Greetings
Marc

----- Forwarded message from Pierre HABOUZIT <madcoder at debian.org> -----

> From: Pierre HABOUZIT <madcoder at debian.org>
> Subject: Re: Bug#330929: how can adduser reliably find out wheter nscd is running?
> To: Marc Haber <mh+debian-bugs at zugschlus.de>, 330929-done at bugs.debian.org
> Date: Sun, 15 Apr 2007 13:01:12 +0200
> User-Agent: Madmutt/devel (Linux)
> X-Spam-Score: (--) -2.4
> X-Spam-Report: torres.zugschlus.de
> 	Content analysis details:   (-2.4 points, 5.0 required)
> 	pts  rule name              description
> 	---- ---------------------- -------------------------------------------
> 	-2.6 BAYES_00               BODY: Bayesian spam probability is 0 to 1%
> 	[score: 0.0000]
> 	0.2 AWL                    AWL: From: address is in the auto white-list
> 
> On Fri, Sep 30, 2005 at 04:05:40PM +0200, Marc Haber wrote:
> > Package: nscd
> > Severity: wishlist
> > 
> > adduser needs to invalidate certain nscd tables after changes were
> > done. Today, it looks for the pid file of nscd to find out whether
> > nscd is running, and then calls nscd via system() with -i and the
> > appropriate arguments.
> > 
> > However, this heuristics has recently begun to fail because the
> > pidfile has moved from /var/run to /var/run/nscd. Instead of changing
> > adduser to check for both files (and probably fail on the next
> > location change), I'd like to have a clearly drawn interface to find
> > out whether it is necessary to invalidate nscd or not.
> > 
> > What do you, as nscd maintainer(s) recommend to have adduser do?
> 
>   well, Why don't you try to do "pidof nscd" ? that would give you a
> good idea. okay, it does not work on the hurd, but some alike method
> would do no ? (like ps ax | grep -q 'sbin/nsc[d]'
> 
> 
>   Else, I'd go for the "looking for the two possible pid paths", which
> is reasonable and harmless IMHO.
> 
>   Btw (I know the bug is very old) please rather come on
> debian-glibc at l.d.o for such questions in the future :)
> 
> Cheers,
> -- 
> ·O·  Pierre Habouzit
> ··O                                                madcoder at debian.org
> OOO                                                http://www.madism.org



----- End forwarded message -----

-- 
-----------------------------------------------------------------------------
Marc Haber         | "I don't trust Computers. They | Mailadresse im Header
Mannheim, Germany  |  lose things."    Winona Ryder | Fon: *49 621 72739834
Nordisch by Nature |  How to make an American Quilt | Fax: *49 3221 2323190



More information about the Adduser-devel mailing list