[252] in linux-net channel archive
Re: 3C503 quits responding to network
daemon@ATHENA.MIT.EDU (Gary Anderson)
Sat Apr 29 02:19:01 1995
Date: Fri, 28 Apr 1995 09:49:15 -0400 (EDT)
From: Gary Anderson <ganderson@clark.net>
To: Paul Gortmaker <paul@rasty.anu.edu.au>
cc: linux-net@vger.rutgers.edu
In-Reply-To: <199504280408.AA00534@rasty.anu.edu.au>
On Fri, 28 Apr 1995, Paul Gortmaker wrote:
> >
> > Trying to troubleshoot a recently set up a 386/33 with Linux, kernel
> > 1.2.4 (and 1.2.6), along with Paul Gortmaker's NE2000/8390 patches against
> > 1.2.6, using a 3C503 for the interface. Problem is, and has been, that
> > periodically the machine stops responding to the net. The machine itself
> > doesn't lock up, it just goes deaf and won't even respond to pings. However,
>
> The fact that you recently set the machine up makes me think you are
> running routed or something equally evil.
>
Nope. Not running routed, gated, or bcastd.
> > if we walk over to the console, do something on it _outbound_ to the net
> > (e.g. - ping another machine) it begins responding to the net again. Have
> > tried three different kernels, 1.2.1, 1.2.4, and now 1.2.6, as I said. The
>
Sorry. Guess I didn't make that terribly clear. The problem has existed
regardless of what kernel we have run - 1.2.1, 1.2.4, and 1.2.6. I tried
to acknowledge that there had been different versions of your patches,
Paul, that were out there, and that we had tried your latest version
against the latest version of the kernel, 1.2.6. I believe you had said
your latest patches were for kernel 1.2.6, and that your patches would
break against any earlier version of the kernel. We had not tried any
earlier versions of your patches.
> I don't think I had patches against 1.2.1. Is this a problem you
had with
> unpatched kernels? If so, then it isn't anything I've broke. If you look
> at my patches to 8390.[c,h] you will find that they are mostly cosmetic.
> The real changes are in ne.c --- so please make it clear if the machine
> was broken before you applied my patches.
Yes, it was. Guess I should make clear here that I was in no way
pointing a finger at your patches. Just letting readers know what we had
tried so far, and that we were applying the right version of your patches
against the proper level of the kernel.
>
> > temporary solution we've adopted is to have cron ping another machine every 5
> > Here is what shows up in the syslog:
> >
> > Apr 27 13:21:31 garc kernel: eth0: transmit timed out, TX status 0x0, ISR 0x0.
> > Apr 27 13:21:31 garc kernel: eth0: Possible network cable problem?
>
> Your packet grew old on the card without ever being sent. The Tx and
> interrupt status bytes don't indicate anything other than nothing
> has happened recently. (ISR=0 -> all 8390 interrupts have been ack'ed
> and TSR=0 -> Tx packet status has already been read/ack'ed)
>
> Paul.
>
Ummm....okay. What direction should we look in now ?
Gary Anderson
ganderson@clark.net
