Date: Wed, 3 Aug 2005 16:57:15 -0300 (ADT) From: "Marc G. Fournier" <scrappy@hub.org> To: Robert Watson <rwatson@FreeBSD.org> Cc: freebsd-fs@freebsd.org, jeff@freebsd.org, freebsd-stable@freebsd.org Subject: Re: Consistent file system hang with RELENG_6 of today ... Message-ID: <20050803165145.L974@ganymede.hub.org> In-Reply-To: <20050729092647.Q74149@fledge.watson.org> References: <20050728231728.E968@ganymede.hub.org> <20050729001252.N1194@ganymede.hub.org> <20050729092647.Q74149@fledge.watson.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 29 Jul 2005, Robert Watson wrote: > > On Fri, 29 Jul 2005, Marc G. Fournier wrote: > >> Now, as most of the "old timers" here know, my jail environment makes use >> of unionfs to "mount" a template layer, to share common binaries amongst >> the VMs ... >> >> I have tried two different ways of doing this, both result in the exact >> same 'file system hang' ... and I have a core of each 'method' ... >> >> The first was to use mount_devfs to, of course, mount the /dev directory >> within the jail itself ... >> >> The second, I used the same /dev that existed within the jail, based on >> bulding a jail using a 4.x system ... >> >> In both cases, the hang appears to be at the same spot, where 'sendmail' >> (in this case, the postfix port) starts up ... > > It would also be interesting to know: if you take unionfs out of the picture > (i.e., you use a regular file system for testing purposes), does the problem > go away? Just checked this, and yes, the jail starts if unionfs is not involved ... the problem is in unionfs itself :( And, there hasn't been any changes that I can see in unionfs since May 3rd ... so it seems a fairly 'long standing' bug :( ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050803165145.L974>