Cosmetic startup issue with readahead scripts

[Matthias Saou]

Miloslav Trmac wrote :

> Matthias Saou napsal(a):
> > Now to fix that... first I guess it should be asked if the 'else action
> > $"Starting $subsys: " $i start' is really useful. Aren't all current
> > init scripts expected to manage their output on their own?
> I think so - but it is breaking compatibility after all.  I guess we'll
> have to do it sooner or later.
> 
> > Should I bugzilla this against initscripts and let discussion follow
> > there?
> This is #190989 already.

Thanks, I've added a comment there.

For the readahead and i8kbuttons cases, a (very ugly) hack would be to
also grep for "echo_" in order to catch the case when the script
doesn't use daemon(), action() etc. but echo_success and/or
echo_failure directly.

Still, I'd really prefer trying to clean everything up so that we can
tear this all out, since it seems like an old leftover anyway, and
creates inconsistency with the output from the "service" command.

Maybe another option :
Could rc check if any output at all was made by a given init script
once it has finished executing (they're all sequential anyway, right?),
and echo a "Started $subsys" line if there was none?

Matthias

-- 
Clean custom Red Hat Linux rpm packages : http://freshrpms.net/
Fedora Core release 5.90 (Test) - Linux kernel 2.6.17-1.2293_FC6
Load : 0.20 0.26 0.31