On Wed, Jun 27, 2018 at 12:53:05PM +0200, Guillem Jover wrote: > > Most likely you meant to break timidity-daemon (<= 2.14.0-1~) rather > > than (>= 2.14.0-1~). > > I'm not sure that'd be correct either. From the bug that this was > supposed to fix (because the changelog description is not clear on > what was broken and what was the fix), this seems to be a new behavior > from timidity-daemon starting in that version, which does not play nice > when pulseaudio is also in the system, and both fight for exclusive > access to the sound devices. > > But this seems to be both a (perhaps temporary) workaround (why not > instead try to diagnose why timidity-daemon now requires exclusive > access to the sound devices?), and in the wrong place (if the problem > is with pulseaudio, then either of the packages need to possibly Break > the other, but not between timidity and timidity-daemon, I for one do > not have pulseaudio on this particular system for example).
I also don't think that a Break is required here at all. If it needs to start after pulseaudio because otherwise the sound device is blocked, maybe the Required-Start field in the init script needs to be changed (to $all). (Maybe that was a regression from the last upgrade that the init script is started earlier.) Regards, Reiner
signature.asc
Description: PGP signature
