Date: Fri, 6 Feb 1998 21:24:20 -0500 (EST) From: Robert Watson <robert@cyrus.watson.org> To: Evan Champion <evanc@synapse.net> Cc: FreeBSD Hackers <hackers@FreeBSD.ORG> Subject: Re: More NFS Problems Message-ID: <Pine.BSF.3.96.980206212226.10258B-100000@trojanhorse.pr.watson.org> In-Reply-To: <025801bd336a$037b3cc0$2844c00a@cello.synapse.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 6 Feb 1998, Evan Champion wrote: > What I see is if the server goes down when only a few processes are trying > to use the mount, the client never times out, but when the server does come > back up, the client restore the mount properly and everything goes along > quite nicely. > > If, on the other hand, a huge number of processes are stuck on the mount, > the stuck processes basically have to be terminated before the mount will be > accessible again, even long after the server has come back up. I have seen this behavior on a number of our -stable clients. The clients hang trying to access the file system even after the server comes back up again. The machine eventually has to be rebooted. So much for a stateless system :). As a result, we are very cautious about rebooting our file server during active client usage. I haven't had this particular problem on our -current machines, but they are a little flaky anyway so who knows :). Robert N Watson Carnegie Mellon University http://www.cmu.edu/ SafePort Network Services http://www.safeport.com/ robert@fledge.watson.org http://www.watson.org/~robert/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.980206212226.10258B-100000>