Catalyst 6509 Routing Table Issue

[ckeithjones]

After a recent upgrade to IOS s72033-adventerprisek9_wan-mz.122-33.shx8a.bin we are seeing the following routing messages in our logs. This message occurs continuously for various periods of time then disappears randomly on various switches and for multiple different subnets. Any ideas on what could be causing this?

Keith

Jun  2 09:42:30: RT: SET_LAST_RDB for 137.244.235.0/24  

OLD rdb: via 137.244.128.182, Vlan853  

NEW rdb: via 137.244.128.174, Vlan851

Jun  2 09:42:30: RT: SET_LAST_RDB for 137.244.235.0/24  

OLD rdb: via 137.244.128.174, Vlan851  

NEW rdb: via 137.244.128.182, Vlan853

Jun  2 09:42:30: RT: SET_LAST_RDB for 137.244.235.0/24  

OLD rdb: via 137.244.128.182, Vlan853  

NEW rdb: via 137.244.128.174, Vlan851

Jun  2 09:42:30: RT: SET_LAST_RDB for 137.244.235.0/24  

OLD rdb: via 137.244.128.174, Vlan851  

NEW rdb: via 137.244.128.182, Vlan853

Jun  2 10:50:33: RT: SET_LAST_RDB for 137.244.235.0/24 

OLD rdb: via 137.244.128.182, Vlan853  

NEW rdb: via 137.244.128.174, Vlan851

Jun  2 10:50:33: RT: SET_LAST_RDB for 137.244.235.0/24  

OLD rdb: via 137.244.128.174, Vlan851  

NEW rdb: via 137.244.128.182, Vlan853

Routing table portion:

sh ip route 137.244.235.0

Routing entry for 137.244.235.0/24  

  Known via "ospf 100", distance 110, metric 30, type intra area  

  Last update from 137.244.128.174 on Vlan851, 1w1d ago 

   Routing Descriptor Blocks:  

* 137.244.128.182, from 137.244.128.250, 1w1d ago, via Vlan853      

   Route metric is 30, traffic share count is 1    

  137.244.128.174, from 137.244.128.250, 1w1d ago, via Vlan851       

   Route metric is 30, traffic share count is 1

[dominic.caron]

In an ospf network, it's a bit fast for normal route refresh. You got some route instability.

[rsimoni · Accepted solution]

Hi Keith,

you see the message because 'debug ip routing' is enabled.

It does not mean that the route is instable (show ip route shows it is stabled since 1w1d) but that you have traffic hitting a prefix which has multiple paths available.

Basically it is just giving details about the actual forwarding current path used for load balancing between the two available paths.  So, the router will switch back and forth between the two paths to distribute the load.

This also explains why you randomly see those messages on few switches every now and then as you will see them only when you have traffic hitting the prefix.

With SXI release 'debug ip routing' also monitors for these types of RDB change on top of RIB related events.

Bug CSCtj34656 documents it.

Bottom line you should not use debugs as generic monitoring tools as they are conceived for other purposes.

You'd better off disable it.

regards,

Riccardo