From owner-freebsd-isp Mon Aug 14 23: 5:45 2000 Delivered-To: freebsd-isp@freebsd.org Received: from misery.sdf.com (misery.sdf.com [204.244.213.49]) by hub.freebsd.org (Postfix) with ESMTP id 5BBC137B97C for ; Mon, 14 Aug 2000 23:05:42 -0700 (PDT) (envelope-from tom@sdf.com) Received: from tom (helo=localhost) by misery.sdf.com with local-esmtp (Exim 2.12 #1) id 13OZWB-0006Xi-00; Mon, 14 Aug 2000 22:44:07 -0700 Date: Mon, 14 Aug 2000 22:43:46 -0700 (PDT) From: Tom Samplonius To: ARCHIVE Cc: freebsd-isp@FreeBSD.ORG Subject: Re: NFS wows In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-isp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, 15 Aug 2000, ARCHIVE wrote: > Hello all; > > We have an install where there are multiple fbsd boxes. Now to make life > simple for kernel rebuilds and the like I nfs mount from one machine. This > machine has the ports and the src and for added comfort the user > directories. At first this seamed like a great idea but it is turning out to > be a night mare. When the nfs server "disappears" for what ever reason, all > the client services seam to hang. That includes login, web, ssh, named... > nothing works. Is there a way to make it so that if that partition > disappears the machines don't become hosed? That shouldn't happen, unless the NFS mounted filesystems contain files that the hung processes need. I do this all the time actually. As long as I don't enter /usr/src while the NFS server is down, everything is ok. If I do, my process will hang waiting for the server to come back up. You should take a look at what you are mounting exactly. If sounds like you are mounting all of /usr or something on your clients, rather than just /usr/src and /usr/obj You can also make the NFS mount soft (as opposed to hard), so they will time out after a while. However, there will still be a delay. Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-isp" in the body of the message