Date: Mon, 2 Jun 1997 07:41:01 -0600 (MDT) From: Marc Slemko <marcs@znep.com> To: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= <ache@nagual.pp.ru> Cc: Peter Wemm <peter@spinner.dialix.com.au>, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-ports@FreeBSD.ORG Subject: Re: cvs commit: ports/net/gated/patches patch-aa Message-ID: <Pine.BSF.3.95.970602073803.19392D-100000@alive.znep.com> In-Reply-To: <Pine.BSF.3.96.970602163219.1306A-100000@nagual.pp.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 2 Jun 1997, [KOI8-R] Андрей Чернов wrote: > On Mon, 2 Jun 1997, Peter Wemm wrote: > > > Nah, this was on my todo list for some time. Somebody suggested it to me > > months ago (it might have been Bruce, or perhaps Jordan or Satoshi), back > > when /var/run was being entirely cleaned at boot - as it should IMHO. If > > something wants the file to stay around, it should't store it there. > > Hmm. As I understand you think that gated/innd/etc. must be fixed to store > pid file at /var/db/ instead, so I need to restore /var/run cleaning in > /etc/rc... This method have one disadvantage, each port/port upgrade > which produce .pid file must be explicitly checked that it test/not test > for .pid file presence at startup... > I am not shure, which solution is better variant... Do you have any > additional arguments/ideas? More than one disadvantage; it means that people will now have no one place to go to try to find the pid file. For programs with frontend interfaces like gated or innd that may not be a big issue, but it is a bad thing to have to start hunting around for PID files in several places. I think I would even prefer a "do not remove" list hack for /var/run to putting them in /var/db...
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.95.970602073803.19392D-100000>