Comment 5 for bug 809646

2011/7/13 Thomas Goirand <email address hidden>:
> On 07/13/2011 03:49 PM, Soren Hansen wrote:
>> As for why some of the scripts use start-stop-daemon and some don't...
>> I'm not sure. Perhaps Thomas can provide some clarity.
> As much as I can remember, so scripts were able to daemonize by
> themselves, and generate a PID file. For these, I didn't use
> start-stop-daemon. For the others, I had unfortunately no other way. I
> wish all this was written with more consistency so that we would never
> need start-stop-daemon (it's not as if writing a PID and doing a fork()
> call was hard, is it? :) ).

Everything used swift-init until I merged your changes. If they failed
to daemonise appropriately, did their init scripts simply never
return? (On a side note: If that's the case, that should have been
filed as an upstream bug. These things should behave consistently. I
want the packaging to add an additional layer of integration, not be a
collection of workarounds to get stuff to work.)

--
Soren Hansen        | http://linux2go.dk/
Ubuntu Developer    | http://www.ubuntu.com/
OpenStack Developer | http://www.openstack.org/