Thanks so much. Happy Holidays to you and everyone on the mailing list.
Shane McKinley Habersham EMC -----Original Message----- From: Robyn Orosz [mailto:[EMAIL PROTECTED] Sent: Tuesday, December 18, 2007 9:55 AM To: Shane McKinley Cc: vyatta-users@mailman.vyatta.com Subject: Re: [Vyatta-users] Routing Policy Confusion Hi Shane, First off, in your BGP peer configuration, you're telling your peer that the next-hop to reach your prefixes is itself. So, anything the peer receives from you will have a next-hop of ZZZ.ZZZ.ZZZ.ZZZ (which would be the peer's IP address). Under most circumstances, you would not want to do this. You'd want to change the next-hop value to an address on your own device (most likely this would be the same IP as your local-ip). So based on your configuration below, you'd set the next-hop to XXX.XXX.XXX.XXX. The policy is exporting all routes learned from BGP to your BGP peers and marking the next-hops of these routes as ZZZ.ZZZ.ZZZ.ZZZ (based on the config below). So, if you're receiving routes from your peer on ZZZ.ZZZ.ZZZ.ZZZ, your policy is sending them back to your peer and telling the peer that in order to reach the routes it should use itself as the next-hop. If you need to advertise a prefix that is originating from your AS, you are not going to be able to do this with your existing policy. You need to advertise routes from an IGP (such as connected or static) on your router. If you have the prefix that you want to advertise configured on eth1, for example, you can configure a policy that exports your connected route. It would look something like: policy-statement "BGP_Export" { term 1 { from { protocol: "connected" network4: 192.168.0.0/24 <---------This is representing the prefix you wish to advertise } then { action: "accept" } } } If your prefix is originating from a different IGP such as ospf or static, you'd configure the policy as 'from protocol ospf4' or 'from protocol static' etc. I hope this clarifies things a bit. Thank you, Robyn Shane McKinley wrote: > I would like to be able to specify which BGP routes get exported, > specifically. From what I understand, policies are the way to go about > this. > > If someone could take the liberty to explain exactly what the > statement below achieves I would be greatful: > > policy-statement "Next_Hop_Self" { > term 1 { > from { > protocol: "bgp" > } > then { > nexthop4: ZZZ.ZZZ.ZZZ.ZZZ > } > } > } > > bgp-id: XXX.XXX.XXX.XXX > local-as: 1234 > export: "Next_Hop_Self" > peer "ZZZ.ZZZ.ZZZ.ZZZ" { > local-ip: XXX.XXX.XXX.XXX > as: 4321 > next-hop: ZZZ.ZZZ.ZZZ.ZZZ > } > > What exactly is this configuration achieving and how would I throw in > this mix to specify exactly what routes to export to the peer > ZZZ.ZZZ.ZZZ.ZZZ? I wouldn't want to export my default gateway; I could > only imagine what kind of havoc that would create. > > I have read the configuration guide on policies, but I am still > confused. > > Please help build my confidence in Vyatta routing policies, > > Shane McKinley > Habersham EMC > _______________________________________________ > Vyatta-users mailing list > Vyatta-users@mailman.vyatta.com > http://mailman.vyatta.com/mailman/listinfo/vyatta-users > _______________________________________________ Vyatta-users mailing list Vyatta-users@mailman.vyatta.com http://mailman.vyatta.com/mailman/listinfo/vyatta-users
