Date: Mon, 26 Dec 2022 15:20:56 -0700 From: Nathan Huff <nhuff@acm.org> To: freebsd-hackers@freebsd.org Subject: Re: daemon(8) exit behavior Message-ID: <86o7rp7qnr.fsf@enyo.nrhuff.com> In-Reply-To: <23fba5e2-0f7d-b609-9ec0-009d6f6bd085@souji-thenria.net> References: <86tu1i6q2s.fsf@enyo.nrhuff.com> <23fba5e2-0f7d-b609-9ec0-009d6f6bd085@souji-thenria.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Souji Thenria <mail@souji-thenria.net> writes: > On 12/26/22 18:18, Nathan Huff wrote: >> >> 1. Is there any interest in upstreaming this? >> > > Because I run in to this problem as well in the past, I would be most > interested in this. > I have submitted https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=268580 with the patch. > > I think using the timeout limit which is used by FreeBSD when shutting > down the system, makes the most sens and would have some consistency > throughout the system, at leas in my opinion. I don't want to tie daemon too closely to the init system since I use it in other contexts at times so that is why I have the delay configurable. > Moreover, I think you should clean up the pidfile, if the daemon process > created it. IIRC that's also the current behavior. > The pidfiles do get cleaned up in my version. They just don't get cleaned up until the supervised process actually exits. With the current behavior the pidfiles can get removed even if the supervised process has not exited. -- Nathan Huff
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?86o7rp7qnr.fsf>