Subject: 3c905-tx output error problems.
To: None <>
From: Simon Burge <>
List: current-users
Date: 08/03/1999 11:25:23

I've got a DEC Prioris 6166 running NetBSD 1.4.1(ish) with a 3Com
3c905-TX (pci) card and a Novell NE3200 (eisa) card in it.  The
network is plain old 10Mbit UTP.

There doesn't seem to be support for the 3200 at the moment:

	wincen:~ 25> dmesg | grep 'unknown device'
	unknown device NVL0701 at eisa0 slot 2 not configured

The 3c905 is more-or-less ok except it gets a lot of output errors.
Here's some config info:

	wincen:~ 26> dmesg | egrep 'ex0|nsphy0'
	ex0 at pci0 dev 6 function 0: 3Com 3c905-TX 10/100 Ethernet
	ex0: interrupting at irq 14
	ex0: MAC address 00:60:97:9c:69:b0
	nsphy0 at ex0 phy 24: DP83840 10/100 media interface, rev. 1
	nsphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
	wincen:~ 27> ifconfig ex0
		media: Ethernet 10baseT
		status: active
		inet netmask 0xffffff00 broadcast
	wincen:~ 28> netstat -in
	Name  Mtu   Network       Address              Ipkts Ierrs    Opkts Oerrs Colls
	ex0   1500  <Link>        00:60:97:9c:69:b0   244292     0    29462  2483  3186
	ex0   1500  172.73.42/24       244292     0    29462  2483  3186
	lo0   32976 <Link>                                31     0       31     0     0
	lo0   32976 127                31     0       31     0     0

Copying /netbsd (about 1.4MB) to an NFS mounted filesystem adds
about 1100 output packets, 80 output errors and 120 collisions every
time.  I've swapped cables, hub ports, etc with no improvement.
Using autoselect or manually selecting 10baseT makes no difference.
Unfortunately I can't pull the box apart, and I don't think I've got
any spare e/net cards anyways.  I've got another box with a 3c905B-TX,
and it works fine.

Two questions then:

 + Has anyone seen any behaviour like this before?  Could it be
   a faulty card?

 + Is support for the NE3200 forthcoming?

an do it's job even if you're working > with the RAID set. This isn't mentioned in the man page for raidctl. In the description of the -i option, it may pay to mention that this task can be backgrounded (with the appropriate warnings). /etc/rc support for autodetecting non-clean parity and starting a "raidctl -i" in the background would be way-cool - I'd play with this myself, but the only box I've got with RAIDframe is a production box, so this isn't an option :-( Simon.