On Friday, March 23, 2012 19:23:11, Michael Banck wrote: > On Fri, Mar 23, 2012 at 06:59:52PM -0400, Chris Knadle wrote: > > On Friday, March 23, 2012 18:26:37, Michael Biebl wrote: > > > On 23.03.2012 20:07, Chris Knadle wrote: > > > > Right now the situation may be somewhat reversed, because in the > > > > general case, daemons need to be patched to work correctly with > > > > systemd. > > > > > > This is simply not true. > > > > > > Only if you want to use socket activation, you need to patch your > > > daemon. But socket activation support is entirely optional. > > > > Lennart Pottering during his talk said that daemons needed to be patched > > to fully work with systemd, > > Where "fully work" implies socket activation, AIUI.
Yes it looks like the normal calls to socket() listen() and bind() that would normally be used are replaced by a single call to sd_listen_fds(), and if the call to sd_listen_fds() returns < 1 then the fallback is to use the standard function calls to socket() listen() and bind() again. http://0pointer.de/blog/projects/socket-activation.html -- Chris -- Chris Knadle chris.kna...@coredump.us -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/201203231940.25912.chris.kna...@coredump.us
