On Thu, Oct 31, 2013 at 09:21:09PM +0100, Jason A. Donenfeld wrote: > On Thu, Oct 31, 2013 at 1:28 PM, Gilles Chehade <[email protected]> wrote: > > > > As for smtpd, it's simpler: it shouldn't terminate unless it explicitely > > is terminated by root somehow. If it has terminated for any reason other > > than root killing it through killall or smtpctl stop (which is the same) > > then we have a bug. > > > > > So this is what I'm wondering about. When, say for hypothetical example, > the lookup process hits a bug and quits. The master process is smart and > realizes this, and then terminates itself. When the master process > terminates due to an error like this, can there be a specific error code? >
Yup, we can return 1 in the parent process, I'll do that tomorrow > Of course I should report bugs. But when they do happen, I want to be > prepared. Also, sometimes a subprocess will be terminated by the kernel due > to OOM. In this case, too, I want to know when it happens. -- Gilles Chehade https://www.poolp.org @poolpOrg -- You received this mail because you are subscribed to [email protected] To unsubscribe, send a mail to: [email protected]
