In 20 years, you will be more dissapointed by what you didn't do than by what you did.

BGP redistribution rules

Redistributing BGP into IGP and vice versa we need to remember about few rules that can be tricky.


IGP into BGP :
1) BGP automatically copies IGP metric to BGP
2) Origin attribute (LINK) will change to "?" - incomplete
3) Will not redistribute OSPF external by default.

BGP is not redistributing OSPF ext because of loop prevention. If some prefix has been  redistributed before from BGP int OSPF they will appear as OSPF ext, redistributing them back again into BGP will cause loops. This is default set up. To change that you need to configure

router bgp X
redistribute ospf Z match internal external


4) Sets weight attribute  to 32768

BGP  into IGP 
To be honest I have never seen this kind of redistribution  in practice. And that logical because there is no IGP who can handle (at this moment 16.09.2016) 655782 prefixes!

BTW here we can check what is current status of BGP table :
http://bgp.potaroo.net/index-bgp.html


and history of BGP table- this one is very impressive!



source : 
http://bgp.potaroo.net/

Back to the point:)
If you will ever do redistribution from BGP into IGP (e.g OSPF, EIGRP) you will do this in laboratory environment.
You need to remember about one big rules :
1) By default only eBGP routes are redistribute into IGP



 

Comments

0 Responses to "BGP redistribution rules"

Post a Comment

Popular Posts