From owner-freebsd-current Mon Aug 7 17:03:13 1995 Return-Path: current-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.11/8.6.6) id RAA08965 for current-outgoing; Mon, 7 Aug 1995 17:03:13 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.cdrom.com (8.6.11/8.6.6) with ESMTP id RAA08951 for ; Mon, 7 Aug 1995 17:03:10 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id RAA01756; Mon, 7 Aug 1995 17:02:53 -0700 From: "Rodney W. Grimes" Message-Id: <199508080002.RAA01756@gndrsh.aac.dev.com> Subject: Re: workaround for talk's address problem To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Mon, 7 Aug 1995 17:02:52 -0700 (PDT) Cc: freebsd-current@FreeBSD.org, joerg_wunsch@uriah.heep.sax.de In-Reply-To: <22696.807786926@time.cdrom.com> from "Jordan K. Hubbard" at Aug 7, 95 02:15:26 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2151 Sender: current-owner@FreeBSD.org Precedence: bulk > > > > > > > > talk(1) has problems with multi-homed hosts. To negotiate the > > > > connection with the remote peer, it uses the first address as returned > > > > by a call to gethostbyname(). This will cause the connection to hang > > > > > > NFS has the exact same problem, FWIW. If there's a more general > > > solution, we should go for it. > > > > NFS does not have such a problem, or at least I have never seen it, and > > I have _lots_ of networks, all but 1 serving NFS: > > I can reproduce this easily. Just take my gateway box and make both > of its addresses, the "lower numbered one for slip" and the "higher > one for local subnet", use the same name. Then try to NFS mount > something from one of the private subnet hosts - it will fail since > DNS returns the entry for the slip line, not the ethernet. I bit > myself with this and have since gone to separate hostnames for each > interface. Lets see here: gndrsh# nslookup gndrsh Server: gndrsh.aac.dev.com Address: 0.0.0.0 Name: gndrsh.aac.dev.com Addresses: 198.145.92.17, 198.145.92.33, 198.145.92.49, 198.145.92.241 Okay, I am subnetted 0xfffffff0, .17 is my 10Mb/s ether, .33 and .49 are 2 100Mb/s ethernets. I can nfs mount from any of those three networks without any problems what so ever. I do it all day long and have been for over a month with this setup. .241 is a slip line, but an NFS mount from 198.145.92.45 has no problem at all mounting disks from gndrsh. I don't know what you are doing, but it should just work. I suspect you have a routing setup problem. Or the udp hack was causing you greif. Are you sure you restarted mountd after screwing with the DNS names? mountd only does DNS lookup of host names when it reads the /etc/exports file. If the IP addresses are added to DNS after that your screwed. Infact mountd is pretty braindead in that respect, and should probably do the DNS lookup each time a request is made, or atleast not use data longer than the DNS TTL value. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD