[Soekris] Re: NET4801: Network performance problems with OpenBSD 3.4

Peter Fors pikofarad at bredband.net
Tue Nov 18 18:47:42 UTC 2003


Nobody runs OpenBSD 3.4 on their Net4801's and can confirm or deny my
findings?


"Peter Fors" <pikofarad at bredband.net> wrote in message
news:000d01c3abb1$4871bb40$fe00a8c0 at peter...
Hi!

Can anyone explain this:

# ifconfig sis0 192.168.0.100 netmask 255.255.255.0
# ping 192.168.0.1
PING 192.168.0.1 (192.168.0.1): 56 data bytes
64 bytes from 192.168.0.1: icmp_seq=0 ttl=64 time=9.909 ms
64 bytes from 192.168.0.1: icmp_seq=1 ttl=64 time=4.942 ms
64 bytes from 192.168.0.1: icmp_seq=2 ttl=64 time=4.933 ms
--- 192.168.0.1 ping statistics ---
3 packets transmitted, 3 packets received, 0.0% packet loss
round-trip min/avg/max/std-dev = 4.933/6.594/9.909/2.345 ms
# ifconfig sis1 up
# ping 192.168.0.1
PING 192.168.0.1 (192.168.0.1): 56 data bytes
64 bytes from 192.168.0.1: icmp_seq=0 ttl=64 time=2.757 ms
64 bytes from 192.168.0.1: icmp_seq=1 ttl=64 time=2.621 ms
64 bytes from 192.168.0.1: icmp_seq=2 ttl=64 time=2.627 ms
--- 192.168.0.1 ping statistics ---
3 packets transmitted, 3 packets received, 0.0% packet loss
round-trip min/avg/max/std-dev = 2.621/2.668/2.757/0.075 ms
# ifconfig sis2 up
# ping 192.168.0.1
PING 192.168.0.1 (192.168.0.1): 56 data bytes
64 bytes from 192.168.0.1: icmp_seq=0 ttl=64 time=0.588 ms
64 bytes from 192.168.0.1: icmp_seq=1 ttl=64 time=0.376 ms
64 bytes from 192.168.0.1: icmp_seq=2 ttl=64 time=0.376 ms
64 bytes from 192.168.0.1: icmp_seq=3 ttl=64 time=0.379 ms
--- 192.168.0.1 ping statistics ---
4 packets transmitted, 4 packets received, 0.0% packet loss
round-trip min/avg/max/std-dev = 0.376/0.429/0.588/0.094 ms
# ifconfig sis1 down
# ping 192.168.0.1
PING 192.168.0.1 (192.168.0.1): 56 data bytes
64 bytes from 192.168.0.1: icmp_seq=0 ttl=64 time=4.416 ms
64 bytes from 192.168.0.1: icmp_seq=1 ttl=64 time=2.614 ms
64 bytes from 192.168.0.1: icmp_seq=2 ttl=64 time=2.621 ms
--- 192.168.0.1 ping statistics ---
3 packets transmitted, 3 packets received, 0.0% packet loss
round-trip min/avg/max/std-dev = 2.614/3.217/4.416/0.847 ms
#

In other words: If not all the network interfaces are 'up' something
seriously bad seems to happen with network traffic... The system is a
Net4801 with BIOS 1.22 and running a default installation of OpenBSD 3.4
(generic kernel) on a 10 GB notebook drive. In the above example the system
was running in single user mode (ie boot -s). Why is this happening? I have
tried compiling a custom kernel also (based on the NET4801 flashdist
prototype), but it doesn't make any difference. Can anyone please test this
and see if the same thing happens in their setup.

Regards

/ Peter Fors






More information about the Soekris-tech mailing list