cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
393
Views
4
Helpful
5
Replies

Removing default route on a Distro causing downstream LAN issues

snickernet
Level 1
Level 1

@snickernet 
Has anyone run into this issues with Cisco 9500 Distro on a Campus LAN environment where I keep having to add the all zeros default route (static route)  back into the distro switch.

If static route is removed....Downstream issues with other building in a campus LAN

  1. losing tacacs login (has to resort to offline password to access distro)
  2. losing switches  - access switches not distro
  3. causing tunnel 0 flapping on distros but not configured
  4. all zeros route clears problem only on building we're making changes but not needed on the other distros in the campus
  5. dropping data, phones and wireless connection in some buildings
    Any ideas is appreciated. Thanks! 

5 Replies 5

Can you share topolgy 

MHM

Distro connected to Core A and Core B and both cores has its own distro independently, connected to the rest of the building dowstream.

We recently migrate some of the buildings from EIGRP to OSPF.
One of the building have issues with OSPF where by we standardize to have this config on all distros:

Router ospf 100
default-information originate
router-id 1xx.1xx.1xx.xx

no ip route 0.0.0.0 0.0.0.0 1xx.1xx.1xx.xx 
when this IP route is removed problem occurs with tunnel 0 flapping with no config on them, causing some building to lose network connectivity. We migrate other bldg from EIGRP to OSPF with no issues except for one building having problem with Static Route.

 

Hello @snickernet ,

the default-information originate creates an LSA type 5 for network 0.0.0.0/0 only if there is a default route in the IP routing table of the node.

By the way you should inject a defaul route in OSPF routing domain at the core level and not on each distro switch.

For the building that have issues check if you have a correct OSPF neighborship with core switches using

show ip ospf neighbors

check also if there is a default route in OSPF database with

show ip ospf database external 0.0.0.0

when the static route is present the distro switch injects a default route that can be used by downstream devices, but to be noted the LSA would be sent also to the core level.

This is the reason why I have written above that the default route should be injected at core level

Hope to help

Giuseppe

 

Tunnel flapping meaning there is recursive issue

I.e. the tunnel destination is learn from tunnel itself.

Static route have lowest AD that why always device use defualt route for tunnel destination 

When you change to ospf which have higher AD than IGP run in tunnel (i think it eigrp) the device start confuse abd issue happened 

Confirm above 

Thanks 

MHM

show ip route | include 00 

Use this command many times to see if defualt route add to rib many time and next hop or type of route 

MHM