IBM UTP NET WEIRDNESS

[John Kemp]

Here is another in a seemingly endless line of weird
network behavior with our RS/320.  Our SUN pings our HP fine.
fine.  Our SUN pings the IBM very strangely.
If we ping the SUN from the IBM, no problem.  

Can anyone explain these weird numbers.  Looks like a bug
in SUN OS 4.0 ping.  But the 74% packet loss is pretty scary.

Thanks in advance,
--------  john kemp            (  (  )_  internet - kemp at hp
  -----                       (  (   __)   decnet - uiatmb::kemp
   ---    univ of illinois   (_ (   __)    bitnet - {uunet,convex}
   --     dept of atmos sci  .(____).               !uiucuxc!hp!kemp
   -      105 s gregory ave    ...          phone - (217) 333-6881
    -     urbana, il 61801    ...             fax - (217) 444-4393

sun- ping -s ibm
PING ibm: 56 data bytes
64 bytes from ibm: icmp_seq=0. time=4. ms
64 bytes from ibm: icmp_seq=1. time=2800. ms
64 bytes from ibm: icmp_seq=2. time=5563. ms
64 bytes from ibm: icmp_seq=3. time=8318. ms
64 bytes from ibm: icmp_seq=4. time=11082. ms
64 bytes from ibm: icmp_seq=5. time=13933. ms
----ibm PING Statistics----
39 packets transmitted, 10 packets received, 74% packet loss
round-trip (ms)  min/avg/max = 4/12505/24998
36.9u 1.2s 0:38 99% 0+252k 0+0io 0pf+0w

sara- ping -s hp
PING hp: 56 data bytes
64 bytes from hp (hp): icmp_seq=0. time=7. ms
64 bytes from hp (hp): icmp_seq=1. time=6. ms
64 bytes from hp (hp): icmp_seq=2. time=4. ms
64 bytes from hp (hp): icmp_seq=3. time=6. ms
64 bytes from hp (hp): icmp_seq=4. time=4. ms
64 bytes from hp (hp): icmp_seq=5. time=6. ms
----hp PING Statistics----
10 packets transmitted, 10 packets received, 0% packet loss
round-trip (ms)  min/avg/max = 4/5/7
sara-