Has anything changes hardware/software wise on the systems?
	Does netstat -m report anything interesting while the problem
	is being exhibited? 

		David/absolute

.---- I've been too drunk to love ----.-- I've been too drunk to remember -.
|          too drunk to care          |    the hell of the night before    |
|  looked like death, felt like hell  |   I've been drinking myself blind  |
`------ been the worse for wear ------'-- and still I'll drink some more --'


On Sun, 10 Jan 1999, Jukka Marin wrote:

> Date: Sun, 10 Jan 1999 17:52:55 +0200
> From: Jukka Marin <jmarin@pyy.jmp.fi>
> To: Port-i386 <port-i386@netbsd.org>
> Subject: SMC 21041 / NetBSD 1.3F problems
> 
> Dear All,
> 
> I'm experiencing _serious_ problems with my Ethernet network.
> 
> I have been using this SMC 21041 card for .. a long time without problems.
> Now my network is 99% unusable.  I have replaced all cabling, disconnected
> all the systems I possibly can, no help.  I also replaced an Ethernet card
> on another machine which I first thought was the source of problems.
> 
> This is what happens:
> 
> At one moment, everything works just fine.  Then, suddenly:
> 
> ping: sendto: No buffer space available
> ping: sendto: No buffer space available
> ping: sendto: No buffer space available
> ping: sendto: No buffer space available
> 64 bytes from 212.246.39.150: icmp_seq=119 ttl=255 time=34774.558 ms
> 64 bytes from 212.246.39.150: icmp_seq=120 ttl=255 time=33776.082 ms
> 64 bytes from 212.246.39.150: icmp_seq=121 ttl=255 time=32778.039 ms
> 64 bytes from 212.246.39.150: icmp_seq=122 ttl=255 time=31780.653 ms
> 64 bytes from 212.246.39.150: icmp_seq=123 ttl=255 time=30781.911 ms
> 64 bytes from 212.246.39.150: icmp_seq=124 ttl=255 time=29783.725 ms
> 64 bytes from 212.246.39.150: icmp_seq=125 ttl=255 time=28784.580 ms
> 
> If I watch the LEDs on my hub, I can see _no_ packets coming out of
> the 21041 card.  At the same time, the other two machines on the
> same net can ping each other just fine.  It's only this 21041 card
> which seems to die completely.  However, if it decides to come back
> alive in time, no ping packets get lost.  They just pile up and then
> get out in one burst, it seems.
> 
> It also seems that one specific NetBSD machine triggers the problem.
> I replaced the Ethernet card on it, I replaced the cabling, the problem
> remained the same.
> 
> Has anyone had problems like this with 21041?  Could it be a problem
> in the NetBSD driver?  Or is my 21041 just dying?
> 
> I happen to live in a town where _no_ computer store has any reasonable
> networking cards in store.  Argh.  All I could find was an ISA d-link
> card (I already have one of those and I never could make it work with
> NetBSD).
> 
> Any ideas at all?
> 
> Thanks,
> 
>   -jm (always in trouble..)
>