Cool eigrp problem - update loop

From: Nick Tucker (kitt@vnet.net)
Date: Fri Jul 16 2004 - 22:56:54 GMT-3


Well I come into work tonite and what do I see:

XXXXX_R2#sh ip eigrp ev
Event information for AS 100:
1 20:24:30.996 Poison squashed: 10.148.17.0/30 reverse
2 20:24:30.980 Poison squashed: 10.148.17.0/30 reverse
3 20:24:30.964 Poison squashed: 10.148.17.0/30 reverse
4 20:24:30.948 Poison squashed: 10.148.17.0/30 reverse
5 20:24:30.932 Poison squashed: 10.148.17.0/30 reverse
...
a few seconds later:

XXXXX_R2#sh ip eigrp ev
Event information for AS 100:
1 20:25:38.847 Poison squashed: 10.148.17.0/30 reverse
2 20:25:38.815 Poison squashed: 10.148.17.0/30 reverse
3 20:25:38.787 Poison squashed: 10.148.17.0/30 reverse
4 20:25:38.779 Poison squashed: 10.148.17.0/30 reverse
5 20:25:38.751 Poison squashed: 10.148.17.0/30 reverse
...
a few more seconds later:

XXXXX_R2#sh ip eigrp ev
Event information for AS 100:
1 20:25:50.747 Poison squashed: 10.148.17.0/30 reverse
2 20:25:50.723 Poison squashed: 10.148.17.0/30 reverse
3 20:25:50.715 Poison squashed: 10.148.17.0/30 reverse
4 20:25:50.711 Poison squashed: 10.148.17.0/30 reverse
5 20:25:50.707 Poison squashed: 10.148.17.0/30 reverse

XXXXX_R2#sh ip route 10.148.17.0
Routing entry for 10.148.17.0/30
   Known via "eigrp 100", distance 90, metric 3339776, type internal
   Redistributing via eigrp 100
   Last update from 10.153.16.69 on ATM1/0/0.90, 19:13:22 ago
   Routing Descriptor Blocks:
   * 10.153.16.69, from 10.153.16.69, 19:13:22 ago, via ATM1/0/0.90
       Route metric is 3339776, traffic share count is 1
       Total delay is 260 microseconds, minimum bandwidth is 768 Kbit
       Reliability 255/255, minimum MTU 1500 bytes
       Loading 3/255, Hops 2

XXXXX_R2#sh ip route 10.148.17.0
Routing entry for 10.148.17.0/30
   Known via "eigrp 100", distance 90, metric 3337728, type internal
   Redistributing via eigrp 100
   Last update from 10.153.18.18 on FastEthernet0/1/0, 4d08h ago
   Routing Descriptor Blocks:
   * 10.153.18.18, from 10.153.18.18, 4d08h ago, via FastEthernet0/1/0
       Route metric is 3337728, traffic share count is 1
       Total delay is 180 microseconds, minimum bandwidth is 768 Kbit
       Reliability 255/255, minimum MTU 1500 bytes
       Loading 3/255, Hops 1

I guess even as interesting as the problem itself, the 1st sh ip route
above shows that it learned the route 19:13:22 ago, even though looking in
the log there has been no activity for 24 hours.

The background on this, this route is for a point to point link for a
remote site. It was migrated from 1 center to another last week.

Of course, a solution may be to just shut the link so the route will remove
itself completely and open it again. In a political-driven business world
its not always that simple and we will try it when we have a window to do
so. :)

Figured others might be interested in some wicked problems, such as this one.



This archive was generated by hypermail 2.1.4 : Sun Aug 01 2004 - 10:11:57 GMT-3