From owner-freebsd-hackers Fri Mar 26 3:35:44 1999 Delivered-To: freebsd-hackers@freebsd.org Received: from octopus.originative (originat.demon.co.uk [158.152.220.9]) by hub.freebsd.org (Postfix) with ESMTP id B607B14D5B for ; Fri, 26 Mar 1999 03:35:41 -0800 (PST) (envelope-from paul@originative.co.uk) Received: by octopus with Internet Mail Service (5.5.2232.9) id ; Fri, 26 Mar 1999 11:33:40 -0000 Message-ID: From: paul@originative.co.uk To: toasty@home.dragondata.com, lem@cantv.net Cc: eischen@vigrid.com, mrcpu@internetcds.com, wilko@yedi.iaf.nl, FreeBSD-hackers@FreeBSD.ORG Subject: RE: comments on de driver error message? Date: Fri, 26 Mar 1999 11:33:32 -0000 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2232.9) Content-Type: text/plain; charset="iso-8859-1" Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > -----Original Message----- > From: Kevin Day [mailto:toasty@home.dragondata.com] > Sent: 30 December 1998 19:02 > To: lem@cantv.net > Cc: eischen@vigrid.com; mrcpu@internetcds.com; wilko@yedi.iaf.nl; > FreeBSD-hackers@FreeBSD.ORG > Subject: Re: comments on de driver error message? > > > > > > FWIW, I had the same configuration (with switches instead of hubs) > > and were getting the same errors in a couple of boxes. > > > > When I certified the patch cords of the top two machines appearing > > in the logs with a WireScope 155 I found one of them bad. After > > replacing both of them, I've seen no more messages like those. > > > > At first I was bitching about the cheap cards or the driver but hey, > > it was the cable. What stroke me is that even with an expensive > > cable-testing device one of the cables was 'good'! > > > > Regards. > > > > -lem > > One other thing to consider though. Most other ethernet > drivers in FreeBSD > don't report the errors at all. de is just talkative. So, > just because the > driver complains of a crc error once or twice a day, doesn't > mean you really > need to go out and change cables. If you had picked another card, you > wouldn't even know the problem existed. The lnc driver is "talkative" as well and I always pipe up at this point and argue it's (and de's) case. These are real errors, they may be hard to find and irritating to net admins but that doesn't change the fact that when the drivers report errors (assuming no bug in the code) that they exist. I've lost count of the number of times that clients have eventually found problems that were the cause of these errors even though they couldn't find anything obviously wrong initially. I find it irritating that cards just don't report these errors by default. I'm happy for there to be a mechanism to switch off errors that admins can't resolve and would rather not see persistently but not being aware of them in the first place is not a useful feature. My experience has been that more often than not these errors are caused by very real problems that need fixing. Paul. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message