LARSON@CRVAX.SRI.COM (Alan Larson) (09/03/89)
IP routing has become more and more of a concern to many of us,
as routes become stranger and stranger. Earlier this morning, we
were unable to find any routes to milnet. Just now, I tried again,
and found that the route is, to say the least, a bit strange.
It looks like the mailbridges are doing quite a bit of passing
the buck between each other. I have attached some traceroute
outputs to the bottom of this message. They were all taken in
a one minute period.
Alan
$ multi trace nic.ddn.mil
traceroute to NIC.DDN.MIL (26.0.0.73), 30 hops max, 38 byte packet
1: 128.18.10.254 (128.18.10.254) 10 ms 0 ms 0 ms
2: SRI-GW.ARPA (128.18.1.1) 10 ms 0 ms 10 ms
3: MARINA-DEL-REY-MB.DDN.MIL (10.6.0.22) 200 ms MCLEAN-MB.DDN.MIL
(10.3.0.111) 350 ms MARINA-DEL-REY-MB.DDN.MIL (10.6.0.22) 180 ms
4: NIC.DDN.MIL (26.0.0.73) 290 ms 240 ms 210 ms
$
$ multi trace nic.ddn.mil
traceroute to NIC.DDN.MIL (26.0.0.73), 30 hops max, 38 byte packet
1: 128.18.10.254 (128.18.10.254) 10 ms 10 ms 0 ms
2: SRI-GW.ARPA (128.18.1.1) 0 ms 10 ms 0 ms
3: RESTON-DCEC-MB.DDN.MIL (10.6.0.20) 170 ms MOFFETT-FLD-MB.DDN.MIL
(10.4.0.51) 250 ms 50 ms
4: NIC.DDN.MIL (26.0.0.73) 170 ms 190 ms 150 ms
$
$ multi trace nic.ddn.mil
traceroute to NIC.DDN.MIL (26.0.0.73), 30 hops max, 38 byte packet
1: 128.18.10.254 (128.18.10.254) 0 ms 10 ms 10 ms
2: SRI-GW.ARPA (128.18.1.1) 0 ms 0 ms 10 ms
3: MCLEAN-MB.DDN.MIL (10.3.0.111) 310 ms MOFFETT-FLD-MB.DDN.MIL
(10.4.0.51) 120 ms MARINA-DEL-REY-MB.DDN.MIL (10.6.0.22) 140 ms
4: NIC.DDN.MIL (26.0.0.73) 170 ms 290 ms 190 ms
$
$ multi trace nic.ddn.mil
traceroute to NIC.DDN.MIL (26.0.0.73), 30 hops max, 38 byte packet
1: 128.18.10.254 (128.18.10.254) 10 ms 10 ms 0 ms
2: SRI-GW.ARPA (128.18.1.1) 10 ms 10 ms 10 ms
3: RESTON-DCEC-MB.DDN.MIL (10.6.0.20) 150 ms MARINA-DEL-REY-MB.DDN.MIL
(10.6.0.22) 130 ms MCLEAN-MB.DDN.MIL (10.3.0.111) 310 ms
4: NIC.DDN.MIL (26.0.0.73) 260 ms 440 ms 250 ms
$
$ multi trace nic.ddn.mil
traceroute to NIC.DDN.MIL (26.0.0.73), 30 hops max, 38 byte packet
1: 128.18.10.254 (128.18.10.254) 10 ms 0 ms 0 ms
2: SRI-GW.ARPA (128.18.1.1) 10 ms 0 ms 0 ms
3: MOFFETT-FLD-MB.DDN.MIL (10.4.0.51) 90 ms RESTON-DCEC-MB.DDN.MIL
(10.6.0.20) 160 ms MCLEAN-MB.DDN.MIL (10.3.0.111) 270 ms
4: NIC.DDN.MIL (26.0.0.73) 160 ms 140 ms 310 ms
-------Mills@UDEL.EDU (09/03/89)
Alan, You are seeing much the same thing as described in my Internet monthly report for July. While I have no hard evidence and have lost direct ARPANET connectivity formerly used to diagnose such zoo events as EGP route churn, the characteristics of the problem appear very much like multiple EGP routes with the same "metric," causing route changes on every EGP update. We all know that EGP is being used as a routing algorithm in many places in the Internet, although such was not the intent in the design. However, considering that pragmatic fact, there are a number of things that could be done to EGP in order to improve its "routing" performance, some already done to at least some implementations of EGP. I have in mind split-horizon, hold-down, hysterisis and other tricks of the trade. The question in most minds, it seems, is whether we should go to the trouble of doing these gruesome things as against waiting for something better to come along. Unfortunately, this question has been in our communal minds since 1983 and only recently has led to what appears to be real progress in various working groups of the IETF. Dave
satlas@BBN.COM (09/06/89)
Alan,
Can you tell me why SRI-GW.ARPA is forwarding the echo requests to
three different Mailbridges within a very short period of time? Is
SRI-GW.ARPA running some form of loadsharing? The Mailbridges do not
control the choice of the next hop for SRI-GW.ARPA, but can just offer
reachability information via EGP. From the traceroute information it
appears that the Mailbridges are receiving the ICMP packets and
forwarding them to the proper destination.
Regards,
Stephen
Unrelated note: From the traceroute information Alan forwarded the
round trips from the west coast to the east coast appear very good.
Truncated version of previous message.
-----------------------------------------------------------------
> IP routing has become more and more of a concern to many of us,
>as routes become stranger and stranger. Earlier this morning, we
>were unable to find any routes to milnet. Just now, I tried again,
>and found that the route is, to say the least, a bit strange.
>traceroute to NIC.DDN.MIL (26.0.0.73), 30 hops max, 38 byte packet
> 1: 128.18.10.254 (128.18.10.254) 10 ms 10 ms 0 ms
> 2: SRI-GW.ARPA (128.18.1.1) 10 ms 10 ms 10 ms
> 3: RESTON-DCEC-MB.DDN.MIL (10.6.0.20) 150 ms MARINA-DEL-REY-MB.DDN.MIL
> (10.6.0.22) 130 ms MCLEAN-MB.DDN.MIL (10.3.0.111) 310 ms
> 4: NIC.DDN.MIL (26.0.0.73) 260 ms 440 ms 250 ms