Hello all,

i just discovered a nice thing about this old problem with our dying xmit logic
on the SGEC adapters. I just installed my RS/6000 and configured NIS on it, as
my VAX also acts as NIS server. And it was about 10 minutes later... my VAX
crashed again due to a big lack of network capabilities :)

After booting, i got lots of these messages on my console:
Jul 17 23:08:09 nereus /netbsd: ze0: xmit logic died, resetting...

After logging in as root, the things got a bit more interesting, as this
appears:
Jul 17 23:31:21 nereus rpcbind: connect from 192.168.177.68 to getport/addr(ypserv)
Jul 17 23:31:26 nereus /netbsd: ze0: xmit logic died, resetting...
Jul 17 23:31:26 nereus rpcbind: connect from 192.168.177.68 to getport/addr(ypserv)
Jul 17 23:31:32 nereus /netbsd: ze0: xmit logic died, resetting...

You feel what i want to say? Every NIS request (don't know why it asks in 5
second intervals...) results in one of those evil messages, which lead to a
complete SGEC crash after a while. And now we continue:
./ypserv stop
Stopping ypserv.
nereus# ./rpcbind stop
Stopping rpcbind.
nereus#

This can keep this state for a very long time... no more xmit logic illness.

Could this help someone to get things a bit clearer in this problem? Or is it
even solved yet? (My kernel is some days older :)

Ok, the kernel version:
NetBSD nereus.unixiron.org 1.5W NetBSD 1.5W (GENERIC) #0: Tue Jun  5 11:01:54 CEST 2001     michael@nereus.unixiron.org:/usr/src/syssrc/sys/arch/vax/compile/GENERIC vax

If nobody knows a solution, how about a PR about this? I think, this really
_IS_ a problem, as my VAX has uptimes between some hours and 40 days, depending
on network traffic... I never really wanted to reboot it :(

...Michael

-- 
visit http://www.bsdfans.org/   Home network powered by: NetBSD OpenBSD FreeBSD
Solaris HP-UX IRIX AIX MUNIX Tru64 Ultrix VMS SINIX Dolphin_Unix OpenStep MacOS