Multiprotocol Configuration
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
07-19-2007 04:28 PM - edited 03-03-2019 05:57 PM
Hi Sir,
I'm configuring a router which is in the boundary of EIGRP and OSPF. Interface Fas0/0 should be EIGRP network and Se0/0 should be OSPF network.
!
int fas 0/0
desc *** EIGRP Network ***
ip address 150.50.1.1 255.255.255.0
!
int s 0/0
desc *** OSPF Network ***
ip address 150.50.2.1 255.255.255.0
!
I'll be performing mutual redistribution between these two protocols. Is it a good practice to cover one interface with overlapping routing protocols? E.g.
!
router eigrp 100
network 150.50.0.0
passive-interface se0/0
!
router ospf 100
network 150.50.2.1 0.0.0.0 area 0
!
Or should I be more explicit? E.g.
!
router eigrp 100
network 150.50.1.1 0.0.0.0
!
router ospf 100
network 150.50.2.1 0.0.0.0 area 0
!
Please advise.
Thank you.
B.Rgds,
Lim TS
- Labels:
-
Other Routing
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
07-19-2007 06:20 PM
The first option will enable EIGRP on both interfaces therefore the second option is the correct one. Don't forget to disable the classful default behavior within EIGRP and change it to classless with the 'no auto-summary' command.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
07-19-2007 10:18 PM
Hi Edison,
Thanks for your response.
Sure, will add the "no auto-summary" command under EIGRP. I just didn't show it in my post.
I'm preparing for the CCIE lab exam this year. I just wanna know the best practices, whether an interface should be covered by multiple routing protocols or not.
If my scenario uses RIP instead of EIGRP, then we can't avoid overlapping both RIP and OSPF on interface Se0/0 due to the classful nature of RIP's network command.
Please advise further.
Thank you.
B.Rgds,
Lim TS
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
07-19-2007 10:41 PM
Hi Lim,
No problem should arise, except that you should care for 2 things, first the overhead RIP traffic which will be sent over the unwanted interface, and this can be avoided by making this interface passive interface, the second thing is that if you don't want the address of the unwanted interface to be advertised and this can be avoided via route filtering.
Good luck in your CCIE preparations :)
HTH,
Mohammed Mahmoud.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
07-20-2007 05:29 AM
First, best of luck obtaining your CCIE.
Second, the best practice when enabling a routing protocol is trying to specify the intended interface as close as possible.
Personally, I always use a host network entry under the routing protocol. I've seen other CCIE candidates using the same subnet mask as the one used in the interface. Either way should work.
For instance,
router ospf 1
network 192.168.1.0 0.0.0.255 area 0
or
router ospf 1
network 192.168.1.1 0.0.0.0 area 0
To answer your question, an interface should be running only one routing protocol. In case of RIP, you can disable the interface from running RIP, if not intended, with the passive interface command.
