Date: Tue, 22 May 2007 22:05:57 +0200 From: =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= <des@des.no> To: Joe Marcus Clarke <marcus@FreeBSD.org> Cc: pjd@FreeBSD.org, "current@freebsd.org" <current@FreeBSD.org> Subject: Re: [Fwd: Serious problem with mount(8)] Message-ID: <864pm4zkre.fsf@dwp.des.no> In-Reply-To: <46534301.10005@FreeBSD.org> (Joe Marcus Clarke's message of "Tue\, 22 May 2007 15\:22\:41 -0400") References: <46534301.10005@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Joe Marcus Clarke <marcus@FreeBSD.org> writes: > I really don't know why I didn't notice this before, but when mount was > converted to use the pidfile(3) API three months ago, the behavior of > mount(8) changed with regard to restarting mountd. A pidfile client > cannot use pidfile_open(3) as this will truncate the pidfile. The > result is that mount reads in a PID of 0, and when it tries to send a > SIGHUP to this PID, it kills itself. Consequently, this is breaking the > ports Tinderbox. > > I suppose we could revert to the previous behavior, but use flopen() to > test if the file is actually locked. Maybe something like this: > > http://www.marcuscom.com/downloads/mount.c.diff Better yet, extend the pidfile API with a function which reads the contents of a PID file and also checks whether it's locked. DES --=20 Dag-Erling Sm=C3=B8rgrav - des@des.no
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?864pm4zkre.fsf>