F18 users unable to log in due to cached nsswitch.conf

[Simo Sorce]

On Wed, 2012-10-17 at 14:12 -0600, Jeff Law wrote:
> On 10/17/2012 11:07 AM, Simo Sorce wrote:
> >
> > Personally I do not like the nss_init() calls, it will just make it even
> > more difficult to diagnose 'heisenbugs' when some apps start doing it,
> > some don't and some other do it at the wrong time.
> >
> > I would rather have glibc do it automatically with rate limiting.
> > Like no more than once every 3 minutes do a stat on one of the getent
> > calls and reload if necessary, still this would be thousands of
> > unnecessary (vs 0 necessary) stat() calls every day, not the best
> > solution.
> Rather than polling, why not use inotify?

Because you do not want to have one inotify file descriptor open for
each and every process in your system, it consumes system resources for
no good reason.

> However, this really needs to be taken upstream (libc-alpha).  I'm not 
> in a position to champion this kind of change.

True,
but I haven;t done anything so far because I can't see a good solution.
It's the same problem we have with resolv.conf, there too I think the
solution is not res_init() but to have a default dns caching daemon, and
have resolv.conf permanently welded to 127.0.0.1, but that's just me.

Simo.

-- 
Simo Sorce * Red Hat, Inc * New York