From owner-freebsd-questions Thu Aug 13 12:59:56 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA00756 for freebsd-questions-outgoing; Thu, 13 Aug 1998 12:59:56 -0700 (PDT) (envelope-from owner-freebsd-questions@FreeBSD.ORG) Received: from resnet.uoregon.edu (resnet.uoregon.edu [128.223.144.32]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA00747 for ; Thu, 13 Aug 1998 12:59:53 -0700 (PDT) (envelope-from dwhite@resnet.uoregon.edu) Received: from localhost (dwhite@localhost) by resnet.uoregon.edu (8.8.5/8.8.8) with SMTP id MAA24653; Thu, 13 Aug 1998 12:59:15 -0700 (PDT) (envelope-from dwhite@resnet.uoregon.edu) Date: Thu, 13 Aug 1998 12:59:15 -0700 (PDT) From: Doug White To: "sysadmin@mfn.org" cc: "'dg@root.com'" , "'dhw@whistle.com'" , "'freebsd-questions@freebsd.org'" Subject: RE: FTP Symptom of Network Problem... In-Reply-To: <01BDC5E3.91EB31C0@noc.mfn.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Wed, 12 Aug 1998, sysadmin@mfn.org wrote: > (1) The terminator on the "monitor" side of the coax > was bad. That helps. :-) > (3) Having replaced the terminator, the problem did *not* > resolve, it merely got "better". I was finally able to > experience the phenomenon myself, and track it down > to the de0 driver ("server" is de0 based, all the rest are > NE2000's). When the de0 driver gets a lot of crc/framing > errors, it "gets lost", literally. In order to recover it, you > have to (a) wait for it, which could take from 2 seconds to > 2 hours, or, (b) reset it by bringing it down/up with ifconfig. > I replaced the card with other known good cards, and the > problem did not change. I can confirm this behavior somewhat. Here in the Network Research Lab we have 10 FreeBSD machines with Kingston and Adaptec cards which operate under the de driver. We used to have Kingston SOHO hubs linking them all together. However, these hubs, to put it nicely, could be greatly improved with a sledgehammer. They would constantly collide and generate bad packets, which made the client machines, who are running NIS and NFS mounted /usr/local and /home, unusable. It'd play heck with our main server as well, which would occaisionally go on vacation. Resetting the interface with ifconfig down/up would solve it for a while. We received a Kingston EtherRX Pro 24 port rackmount hub which is much better quality than the Kingston SoHo. The SOHOs were sent back and now I swear by Asante minihubs. :) Doug White | University of Oregon Internet: dwhite@resnet.uoregon.edu | Residence Networking Assistant http://gladstone.uoregon.edu/~dwhite | Computer Science Major To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message