From owner-freebsd-stable Tue Jan 25 15:45: 3 2000 Delivered-To: freebsd-stable@freebsd.org Received: from visionaire.ping.de (visionaire.ping.de [195.37.123.61]) by hub.freebsd.org (Postfix) with ESMTP id 379D81515C for ; Tue, 25 Jan 2000 15:44:49 -0800 (PST) (envelope-from thomas@visionaire.ping.de) Received: from dante.visionaire.net (mail@dante.visionaire.net [192.168.208.42]) by visionaire.ping.de (8.9.3/8.9.3) with ESMTP id AAA00517; Wed, 26 Jan 2000 00:39:49 +0100 (CET) (envelope-from thomas@dante.visionaire.net) Received: from thomas by dante.visionaire.net with local (Exim 2.05 #1 (Debian)) id 12DFYq-00007n-00; Wed, 26 Jan 2000 00:39:48 +0100 Date: Wed, 26 Jan 2000 00:39:48 +0100 From: Thomas Keusch To: US Debian User Mailing List Cc: US FreeBSD Stable Mailing List Subject: FreeBSD, Linux and NFS Message-ID: <20000126003948.A476@dante.visionaire.net> Reply-To: thomas@members.ping.de Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii User-Agent: Mutt/1.0i Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hello, does anybody happen to be running NFS between FreeBSD und Linux successfully and painless? I've got FreeBSD 3.4S cvsupped xmas or one, maybe two days after, running as NFS server, while my Linux v2.2.14 box is the client. On Linux, I mount stuff like this: $ mount -t nfs -o ro,noauto,user,tcp,rsize=8192,wsize=8192,timeo=14,intr \ darkstar:/storage /tmp/storage.darkstar On the FreeBSD side of the universe, I've got this in /etc/rc.conf: $ grep -e nfs -e mount /etc/rc.conf nfs_server_enable="YES" nfs_server_flags="-t -n 4" mountd_flags="-2 -r" Ok, so if I get things right, which I think (hopefully) I do, this means that Linux mounts NFS filesystems via TCP/IP, the FreeBSD server forces NFS v2 via the "-2" switch given to mountd and the 4 nfsd only serve stuff via TCP/IP ("-t" argument). But, stressing my home LAN a little bit the last few days, it happened to me three times already that the server simply stops serving the fs (or NFS in general, I didn't check before restarting stuff :-( ), while it stays perfectly functional in all other ways I can think of. Ok, it may be that no new/further NFS exported fs may be mounted by remote machines, I don't know yet, but I can get you current if asked to. Maybe it is just a regular permission problem that it can't mount the fs in question via loopback networking. Restarting nfsd/mountd, whichever order of starting them, doesn't seem to help, on the Linux box processes accessing the NFS fs still hang. Having restarted the daemons, I can mount another fs, though. (Don't know if this works without restarting, sorry. But I can test..) *Finally* :-) ... finally this is the error message I get on the Linux box syslog: dante kernel: nfs: server darkstar not responding, still trying Any ideas on that one? Is my assumption, that _only_ NFS v2 via TCP/IP, *NOT* UDP, will work between FreeBSD and Linux, correct? At all? If it ever was? (I just seem to have remembered this sort of stuff popping up on the lists and v2/tcp NFS being the solution) So, is there some help for me? I don't really know where to look at now.. Thanks for your valueable time, if you actually read this far :-)) HAND, thomas -- thomas. .powered.by.debian/linux. .served.by.FreeBSD. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message