optimizing systemd, was: f15 no longer starts

Tue Mar 22 15:08:03 UTC 2011

On Tue, 2011-03-22 at 14:44 +0200, cornel panceac wrote:
> alright adam but, what do you think can be done to prevent such a
> think happen again?

A good one would be 'don't break rsyslog'. =) But really, this failure
doesn't worry me much. It's a pre-release, things break in pre-release
development. Our update testing system worked exactly as it's supposed
to: the broken update never made it into the release stream. AFAICS, all
QA goals were met here.

>  it is possible to change/reconfigure systemd in such a way that a
> failing service will not just keep the system from booting?

Is it possible to change our initscripts such that this particular
failure has less of an impact? Probably. Is that a good change that we
ought to do? Possibly, possibly not; it looks like it'd have significant
consequences. It's something to talk to Bill and Lennart about, I'd say.
I'm not sure they follow this list, so either bring it up on devel list,
or in the bug report.
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net