MAE-East snafu redux

Mailman List Mirror (Read Only)
1

Has everyone just become so used to terrible performace at MAE-East, or am
I the only one seeing problems? I'm seeing perfect performance to
networks connected to Gigaswitch 1 (excluding overloaded ports), but
15-20% packet loss to anyone connected to any other Gigaswitch. In the
past MFS would jump on these tickets, find a frozen card on Gigaswitch 6,
reset it, and clear the trouble. It has been over 24 hours since I opened
a ticket, I've spoken with MFS Datanet Engineering, and I still don't
think I've spoken to anyone at MFS who is aware that all traffic between
Gigaswitches goes through Giga6.
If anyone has time tonight, could they check performance to other
providers on the same Gigaswitch and on other Gigas. If you need to know
which provider is on which switch, go to
http://www.mfsdatanet.com:80/MAE/east.map.html
Thanks, and hopefully this is on topic, even if it doesn't contain actual
router commands. :slight_smile:

Jeremiah Kristal
Senior Network Engineer
ICon CMT Corporation
jeremiah@iconnet.net
201-319-5764
x284 internal

2

Well, from our router (1292.41.177.103, catalyst 6) it looks like this (note
the packet loss!!!)

Jan

There the pings:

core0.mae-east#ping 192.41.177.181

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.41.177.181, timeout is 2 seconds:
.!.!!
Success rate is 60 percent (3/5), round-trip min/avg/max = 100/189/268 ms
core0.mae-east#ping 192.41.177.181

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.41.177.181, timeout is 2 seconds:
!!!.!
Success rate is 80 percent (4/5), round-trip min/avg/max = 4/13/20 ms
core0.mae-east#ping 192.41.177.113

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.41.177.113, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 8/11/16 ms
core0.mae-east#ping 192.41.177.101

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.41.177.101, timeout is 2 seconds:
.!.!!
Success rate is 60 percent (3/5), round-trip min/avg/max = 4/8/12 ms
core0.mae-east#ping 192.41.177.101

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.41.177.101, timeout is 2 seconds:
.!!!.
Success rate is 60 percent (3/5), round-trip min/avg/max = 1/28/44 ms
core0.mae-east#ping 192.41.177.73

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.41.177.73, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 12/17/24 ms
core0.mae-east#ping 192.41.177.119

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.41.177.119, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 16/19/24 ms
core0.mae-east#ping 192.41.177.69

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.41.177.69, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/3/4 ms

3

http://208.234.102.97/MAE/east.aggr.overlay.html

The past two years it has been a pretty much smoothe curve in the traffic
graph, but the past months (as soon as it started hitting 1500mbit/s) the
graph started to get rugged the way it is now. I interpret that as
trouble... It used to spike a few times a week max, nowadays it seems to
do that hourly.

I interpret that as overload.