BGP route can't redistribute in ospf router

Unanswered Question
Jul 20th, 2008
User Badges:

Hi all, I am testing the failover fo the setup network but it seems fail.


Basically, the connection is work in normal condition. But when I try to make

Router 26 (10.1.1.2 port down ) and Router 36 ( 172.20.1.1 port down ). I expect

traffic from Router 16A will go though Router 36 and then Router 26 until to 1400 (e.g. 202.122.65.1)


But the final result is Router 26 and 36 can learn 202.122.65.0/26 route but 16A can't

learn any Ex route which seems that Router 36 can't redistribute any bgp route into 16A.


Anyone know what is wrong on my configuration?


Thanks



  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
acbenny Sun, 07/20/2008 - 18:17
User Badges:

Thanks Milan,


Got it, but how can I avoid routing loop in this case As it mentioned that it is dangerous to issue this command.

Marwan ALshawi Sun, 07/20/2008 - 19:01
User Badges:
  • Purple, 4500 points or more
  • Community Spotlight Award,

    Best Publication, December 2015

ok u can achive it by doing condational redistribution with route-maps

in the case u can install that route when ur connection is lost

in oother words when a spicific route not insstaled in ur routing table


hth


Rate if helpful

milan.kulik Sun, 07/20/2008 - 23:17
User Badges:
  • Red, 2250 points or more

Hi,


another idea:

Tagging the iBGP prefixes while incoming and redistributing them to OSPF with higher metric than the eBGP ones.


BR,

Milan



acbenny Sun, 07/20/2008 - 23:59
User Badges:

Thanks your idea, But how to configure it by command line. I don't know how to do it.

milan.kulik Tue, 07/22/2008 - 01:58
User Badges:
  • Red, 2250 points or more

Hi,


IMHO, modifying your configs the way presented in the 36-new.txt file might work.


route-map incoming

would add zzz community to the prefixes received from the iBGP neighbor.


Only those prefixes with community matching would be redistributed to OSPF.


On the other hand,

redistribute ospf 123 metric 100 match internal

would redistribute only OSPF internal routes to BGP (not that sent from BGP to OSPF originally).

You'd lose the possibility to redistribute static to OSPF and then to BGP, e.g., but I hope you can live without it.


BR,

Milan





Attachment: 
acbenny Sun, 07/20/2008 - 23:59
User Badges:

Thanks your idea, But how to configure it by command line. I don't know how to do it.

Actions

This Discussion