Hi Geoff,
Thanks for the reply, no I don't think it's the box, DMESG below.
Ok some test output where the IP pinged is the far end of a /30 subnet
on a dedicated 1G line rate router port of a 7609 cisco, sup 720 etc..
If I do a flood PING
# time ping -c 1000 -f 80.65.xxx.xxx
PING 80.65.xxx.xxx (80.65.xxx.xxx): 56 data bytes
--- 80.65.xxx.xxx ping statistics ---
1000 packets transmitted, 1000 packets received, 0.0% packet loss
round-trip min/avg/max/std-dev = 0.241/0.279/10.349/0.344 ms
0m0.32s real 0m0.00s user 0m0.06s system
#
# time ping -f 80.65.xxx.xxx
PING 80.65.xxx.xxx (80.65.xxx.xxx): 56 data bytes
--- 80.65.xxx.xxx ping statistics ---
26221 packets transmitted, 26218 packets received, 0.0% packet loss
round-trip min/avg/max/std-dev = 0.236/0.389/23.944/0.465 ms
0m5.89s real 0m0.04s user 0m1.59s system
#
3 users Load 0.56 0.46 0.46 Wed Oct 1
20:14:27 2008
Iface State Ibytes Ipkts Ierrs Obytes Opkts
Oerrs Colls
re0 up:U 0 5585 0 798202 5670
0 0
re1 up:U 0 0 0 0 0
0 0
nfe0 dn 0 0 0 0 0
0 0
enc0 dn 0 0 0 0 0
0 0
lo0 up 0 0 0 0 0
0 0
pflog0 up 0 0 0 0 0
0 0
trunk0 up:U 544226 5585 0 877582 5670
0 0
trunk1 up:U 0 0 0 0 0
0 0
Totals 544226 11170 0 1675784 11340
0 0
Packets are going out through trunk0 (1 member re0) i.e 5k+ pps
Doing a HPING to the same host
# time hping -c 1000 -i u100 -1 80.65.xxx.xxx
len=46 ip=80.65.xxx.xxx ttl=255 id=34206 icmp_seq=0 rtt=0.3 ms
len=46 ip=80.65.xxx.xxx ...