[c-nsp] SUP 720 3BXL + 2 BGP Full View + MPLS VPN

Pavel Kuzin pk at nodex.ru
Fri Feb 29 02:59:54 EST 2008


Hi!

No. I do not redistribute bgp into igp.
I was send my BGP config in previons letter.
I do not have any vfr`s configured.. :-(

--
Pavel D.Kuzin
Technical Director.
Nodex LTD.
Saint-Petersburg, Russia
pk at nodex.ru
http://nodex.ru
tel. +7 812 3800033 ext. 311
fax. +7 812 2304573

----- Original Message ----- 
From: "Ian Cox" <icox at cisco.com>
To: "Pavel Kuzin" <pk at nodex.ru>
Sent: Friday, February 29, 2008 2:37 AM
Subject: Re: [c-nsp] SUP 720 3BXL + 2 BGP Full View + MPLS VPN


> 
> Your not redistributing bgp into your igp? Normally the only way the 
> prefix table in bgp causes a label per prefix is if you place it into 
> a vrf, or you redistribute bgp into your igp. The normal case is you 
> only have labels created for routes in your igp. Since you don't have 
> routes in the vrf, is your bgp tabe somehow ending up in your igp?
> 
> 
> Ian
> 
> 
> At 12:58 AM 2/29/2008 +0300, you wrote:
>>Hi
>>
>>BGP currently only in global. No one VRF configured in system.
>>
>>This is my BGP config:
>>
>>router bgp XXXXX
>>   bgp router-id XX.XX.XX.15
>>   bgp always-compare-med
>>   no bgp default ipv4-unicast
>>   bgp cluster-id XX.XX.XX.0
>>   bgp log-neighbor-changes
>>   neighbor XX.XX.XX.10 remote-as XXXXX
>>   neighbor XX.XX.XX.10 description XXXXX-GW
>>   neighbor XX.XX.XX.10 update-source Loopback0
>>   neighbor XX.XX.XX.14 remote-as XXXXX
>>   neighbor XX.XX.XX.14 description XXX-GW
>>   neighbor XX.XX.XX.14 update-source Loopback0
>>   neighbor XX.XX.225.153 remote-as 3216
>>   neighbor XX.XX.225.153 description GOLDEN-TELECOM
>>   neighbor XX.XX.67.33 remote-as 1299
>>   neighbor XX.XX.67.33 description TELIA-SONERA
>>   !
>>   address-family ipv4
>>     neighbor XX.XX.XX.10 activate
>>     neighbor XX.XX.XX.10 send-community extended
>>     neighbor XX.XX.XX.10 route-reflector-client
>>     neighbor XX.XX.XX.14 activate
>>     neighbor XX.XX.XX.14 send-community both
>>     neighbor XX.XX.XX.14 route-reflector-client
>>     neighbor XX.XX.225.153 activate
>>     neighbor XX.XX.225.153 next-hop-self
>>     neighbor XX.XX.225.153 route-map from_golden in
>>     neighbor XX.XX.225.153 route-map GOLDEN-OUT out
>>     neighbor XX.XX.67.33 activate
>>     neighbor XX.XX.67.33 next-hop-self
>>     neighbor XX.XX.67.33 route-map from_telia in
>>     neighbor XX.XX.67.33 route-map TELIA-OUT out
>>     no auto-summary
>>     no synchronization
>>
>>   exit-address-family
>>   !
>>   address-family vpnv4
>>     neighbor XX.XX.XX.10 activate
>>     neighbor XX.XX.XX.10 send-community extended
>>     neighbor XX.XX.XX.10 route-reflector-client
>>     neighbor XX.XX.XX.14 activate
>>     neighbor XX.XX.XX.14 send-community extended
>>     neighbor XX.XX.XX.14 route-reflector-client
>>   exit-address-family
>>
>> >
>>
>> > Are you placing the BGP routes into a VRF rather than the global
>> > table? In the global case you will end up with labels just for IGP routes,
>> > with the VRF case you will end up with a label per prefix.
>> >
>> >
>> > Ian
>> >
>> >
>> > At 11:35 PM 2/28/2008 +0300, Pavel Kuzin wrote:
>> >
>> >> Hi!
>> >>
>> >>
>> >> Thanks for your reply.
>> >>
>> >>
>> >> I have just simple MPLS and BGP config too.
>> >>
>> >>
>> >> #sh run | in mpls
>> >> mpls ldp loop-detection mpls label protocol ldp mpls mtu 1540 mpls ip mpls
>> >> mtu 1540 mpls ip mpls mtu 1540 mpls ip mpls ldp router-id Loopback0 force
>> >>
>> >> #sh mpls interfaces
>> >> Interface              IP            Tunnel   BGP Static Operational
>> >> Vlan100                Yes (ldp)     No       No  No     Yes
>> >> Vlan410                Yes (ldp)     No       No  No     Yes
>> >> Vlan301                Yes (ldp)     No       No  No     Yes
>> >>
>> >>
>> >>
>> >>> hi pavel,
>> >>>
>> >>> i must admit, i'm not an mpls guru, but im working on it ;) whilst i
>> >>> can't offer you any direct solution, i can say i'm a bit fascinated
>> >>> by this because i have for example a sup720 3bxl taking 3 x full views
>> >>> 243k
>> >>> routes per peer, have mpls tag switching enabled on 5 SVI's and i dont
>> >>> see the behaviour you see.
>> >>>
>> >>> router#sh mpls int | i Vl Vlan2                  Yes (ldp)     No Yes
>> >>> Vlan3                  Yes (ldp)     No       Yes
>> >>> Vlan4                  Yes (ldp)     No       Yes
>> >>> Vlan12                 Yes (ldp)     No       Yes
>> >>> Vlan14                 Yes (ldp)     No       Yes
>> >>>
>> >>>
>> >>>
>> >>> router# sh ip bgp sum Neighbor        V    AS MsgRcvd MsgSent
>> >>> TblVer
>> >>> InQ OutQ Up/Down
>> >>> State/PfxRcd
>> >>> 61.88.xxx.xx    4  7474 19165667  424902 69092680    0    0 21w2d
>> >>> 243009
>> >>> 61.88.xxx.xx    4  7474 20943846  426386 69092680    0    0 5w2d
>> >>> 243007
>> >>> 220.157.xx.x    4  9723 12753587 15032863 69092733    0    0 2w0d
>> >>> 243093
>> >>>
>> >>>
>> >>>
>> >>> router#sh platform hardware cap | b L3 Forward L3 Forwarding
>> >>> Resources
>> >>> FIB TCAM usage:                     Total        Used
>> >>> %Used
>> >>> 72 bits (IPv4, MPLS, EoM)     524288      246423
>> >>> 47%
>> >>> 144 bits (IP mcast, IPv6)      262144           5
>> >>> 1%
>> >>>
>> >>>
>> >>>
>> >>> router#sh mpls ip bind sum Total number of prefixes: 1453
>> >>>
>> >>>
>> >>> Generic label bindings
>> >>> assigned        learned prefixes      in labels     out labels 1453
>> >>> 1446          13042
>> >>>
>> >>>
>> >>>
>> >>> my mpls config is pretty straight forward
>> >>>
>> >>> mpls label range 16 256000 mpls label protocol ldp no tag-switching
>> >>> ip propagate-ttl tag-switching tdp router-id Loopback0 force
>> >>>
>> >>> ... very curious !!
>> >>>
>> >>>
>> >>>
>> >>> hardware is sup720 pfc3bxl with ios 12.2(18)SXF8
>> >>>
>> >>> kind regards,
>> >>>
>> >>> --matt
>> >>>
>> >>>
>> >>>
>> >>>
>> >>>
>> >>>
>> >>
>> >> _______________________________________________
>> >> cisco-nsp mailing list  cisco-nsp at puck.nether.net
>> >> https://puck.nether.net/mailman/listinfo/cisco-nsp
>> >> archive at http://puck.nether.net/pipermail/cisco-nsp/
>> >
>> >
>>
>>_______________________________________________
>>cisco-nsp mailing list  cisco-nsp at puck.nether.net
>>https://puck.nether.net/mailman/listinfo/cisco-nsp
>>archive at http://puck.nether.net/pipermail/cisco-nsp/
> 
>


More information about the cisco-nsp mailing list