On Sun, Jun 26, 2005 at 10:35:46PM +0100, Steve Kemp wrote: > > Not sure how hard would be to integrate this with Debian (had to tune some > > files by hand on test box), but it's definitely worth trying. > > > > On my test box it started system almost *three* times faster. > > I guess the obvious question would be .. does it require any > changes to the init.d scripts themselves to gain this speedup? Yes and no ;) By default it's total replacement of old init system. It comes with its own starting scripts by default. And they're very "gentooish", but hey. all this packaging is about to integrate it with Debian. > Or is the asynchronous nature sufficient to boost startup > performance? I'm going to try to use our existing scripts although it needs some tweaking cause we don't have any dependency information in them. > If that is the case that'd be great - but I'd be initially > suspicious unless it could ensure the correct ordering such > that network access is available before servers start, etc. Its starting scripts keep information about ordering. I suppose that for now it will be packaged to be used as a total replacement of our init system. But maybe some minor changes in our init.d scripts policy would allow us to use it with Debian scripts. We'll see. I'm quite busy now with my exams so I didn't look much into the code. regards fEnIo -- ,''`. Bartosz Fenski | mailto:fenio@debian.org | pgp:0x13fefc40 | irc:fEnIo : :' : 32-050 Skawina - Glowackiego 3/15 - w. malopolskie - Poland `. `' phone:+48602383548 | proud Debian maintainer and user `- http://skawina.eu.org | jid:fenio@jabber.org | rlu:172001
Attachment:
signature.asc
Description: Digital signature