EIGRP Update Intervals

Unanswered Question

We have 2 core routers running HSRP with some static routes being advertized to the WAN. Router 1 is the active router, router 2 the standby. Show standby verifies this.

Just looking at one instance of the observed phenominon, both core routers have a static route for a subnet to another network device, the route never changes.

2 routers have interfaces on a common subnet with the core routers and are getting getting route updates for this particular route. Router 3 is getting updated every couple of days from router 2 and router 4 is getting updated every month from router 1. Both routes are pointing to the standby HSRP router.

Why is the route being updated when it isn't changing and why is is pointing to the wrong router in the HSRP?

I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Jon Marshall Tue, 09/01/2009 - 04:24

Wes

Bear in mind with HSRP that it is only really used for redundant default-gateways for end devices. When routers exchange routes they use the physical interface addresses so they don't care which one of the interface addresses belongs to the active HSRP router. However if all the 4 routers are sharing a common subnet then the question should perhaps be "why does router 3 and 4 not see 2 equal cost paths to the remote network" ie. when you do a "sh ip route " you should see 2 equal cost paths.

As for the update, not sure really, could be a number of things.

Perhaps as Davy suggested, a topology diagram together with some config and "sh ip route" from R3 & R4 would enable us to help you more.

Edit - if you wanted R3 and R4 to use the active HSRP gateway then you could add static routes on R3 & R4 pointing to the virtual HSRP IP address. However this is not a recommendation - i prefer to let the routing protocol do what it does best :-)

Jon

Attached is a simplified diagram.

Here's the output from the 'sh ip route xx.xx.xx.xx' command from one of the routers, the other routers output is the same.

Routing entry for 198.80.140.0/24

Known via "eigrp 5191", distance 170, metric 28416, type external

Redistributing via eigrp 5191

Last update from 10.163.14.3 on FastEthernet0/1, 02:16:29 ago

Routing Descriptor Blocks:

* 10.163.14.3, from 10.163.14.3, 02:16:29 ago, via FastEthernet0/1

Route metric is 28416, traffic share count is 1

Total delay is 110 microseconds, minimum bandwidth is 100000 Kbit

Reliability 255/255, minimum MTU 1500 bytes

Loading 1/255, Hops 1

10.163.14.2, from 10.163.14.2, 02:16:29 ago, via FastEthernet0/1

Route metric is 28416, traffic share count is 1

Total delay is 110 microseconds, minimum bandwidth is 100000 Kbit

Reliability 255/255, minimum MTU 1500 bytes

Loading 2/255, Hops 1

it's my understanding that the routes are listed in order of precedence, correct me if I'm wrong.

Attachment: 
Jon Marshall Wed, 09/02/2009 - 09:02

Wes

Okay, so your router is seeing 2 equal cost paths to each core switch.

"it's my understanding that the routes are listed in order of precedence, correct me if I'm wrong."

Not really. If the routes are the same metric and AD which they are here then the router will alternate between the 2 next-hops. It can do this either by per-packet or per-destination, per-destination usually being the default.

So each router will use both paths to the core and simply alternate.

Jon

Peter Paluch Thu, 09/03/2009 - 23:07

Hello Wes, Jon,

Wes is correct here and it is my thinking as well. I have not heard about EIGRP readvertising routes if they are stable, as opposed to, say, OSPF that needs to readvertise the LSAs every 30 minutes. As far as I know, there is not a max age or a similar value associated with EIGRP routes.

Wes, how exactly is that route injected into EIGRP? Is it redistributed? If so, can you post the output of the command "show ip eigrp topology 198.80.140.0" from the router where you see the flapping?

Best regards,

Peter

Actions

This Discussion