[c-nsp] EoMPLS/Sup32/xconnect - missing something obvious

Paolo Lucente pl+list at pmacct.net
Sun May 24 07:22:18 EDT 2009


Hi,

make sure OSPF is configured properly and you can see router
loopbacks advertised in the IGP: LDP establishing adjacencies
is not enough to generate labels. 

Also, as the pseudowire is simply Eth-to-Eth, i guess you
don't want the interworking feature (interworking ip) to
lie within the pseudowire-class.

As Gert suggested, try a simplified configuration first.

Cheers,
Paolo



On Fri, May 22, 2009 at 08:14:50PM -0400, L'argent wrote:
>
>
> Setup: Two directly connected Sup32's directly connected one running  
> 12.2-33.SXH5, the other 12.2-33.SXI1 otherwise identical configurations.
> Goal: Test (new) MPLS configuration using EoMPLS (xconnect) between G4/2  
> on each. I want to transport 1G over a 10G link between two sites.
>
> Problem: The VC won't come up. Labels don't appear to be generated. Not  
> sure what knob to twiddle to make it happy. LDP appears to see its  
> neighbor, but no MPLS adjacency appears. I am guessing I'm missing  
> something dead obvious.
>
> Router 1:
>
> mls ip slb purge global
> mls netflow interface
> no mls flow ip
> no mls flow ipv6
> mls cef error action reset
> mpls traffic-eng tunnels
> mpls ldp graceful-restart
> mpls ldp discovery targeted-hello accept
> mpls ip default-route
> mpls label range 50 524000
> mpls label protocol ldp
>
> pseudowire-class EtherEncap
>  encapsulation mpls
>  interworking ip
>
> interface Loopback0
>   ip address 10.0.0.113 255.255.255.255
>
> interface GigabitEthernet4/2
>  mtu 1526
>  no ip address
>  speed nonegotiate
>  mpls ldp discovery transport-address interface
>  mpls bgp forwarding
>  mpls label protocol ldp
>  mpls ip
>  xconnect 10.0.0.114 100 encapsulation mpls
>
> router ospf 1
>  log-adjacency-changes
>  network 10.0.0.113 0.0.0.0 area 0
>  bfd all-interfaces
>  mpls traffic-eng router-id Loopback0
> !
> router bgp 65517
>  bgp router-id 10.0.0.113
>  bgp log-neighbor-changes
>  bgp confederation identifier 100
>  bgp confederation peers 65516 65518
>  neighbor 10.0.0.114 remote-as 65518
>  neighbor 10.0.0.114 ebgp-multihop 255
>  neighbor 10.0.0.114 update-source Loopback0
>  neighbor 10.0.0.114 version 4
>  !
>  address-family ipv4
>   neighbor 10.0.0.114 activate
>   neighbor 10.0.0.114 send-label
>   no auto-summary
>   no synchronization
>   network 10.0.0.113 mask 255.255.255.255
>  exit-address-family
> !
> ip classless
>
> #sh mpls l2 binding
>   Destination Address: 10.0.0.114,  VC ID: 100
>     Local Label:  unassigned.
>     Remote Label: unassigned
>
>
> #sh mpls l2 vc
>
> Local intf     Local circuit              Dest address    VC ID      Status
> -------------  -------------------------- --------------- ----------  
> ----------
> Gi4/2          Ethernet                   10.0.0.114 100        DOWN
>
> #sh mpls l2 vc detail
> Local interface: Gi4/2 down, line protocol down, Ethernet down
>   Destination address: 10.0.0.114, VC ID: 100, VC status: down
>     Output interface: none, imposed label stack {}
>     Preferred path: not configured
>     Default path: no route
>     No adjacency
>   Create time: 00:53:56, last status change time: 00:53:56
>   Signaling protocol: LDP, peer 10.0.0.114:0 up
>     Targeted Hello: 10.0.0.113(LDP Id) -> 10.0.0.114
>     MPLS VC labels: local unassigned, remote unassigned
>     Group ID: local unknown, remote unknown
>     MTU: local unknown, remote unknown
>     Remote interface description:
>   Sequencing: receive disabled, send disabled
>   VC statistics:
>     packet totals: receive 0, send 0
>     byte totals:   receive 0, send 0
>     packet drops:  receive 0, send 0
>
> #sh ip route 10.0.0.114
> Routing entry for 10.0.0.114/32
>   Known via "ospf 1", distance 110, metric 2, type intra area
>   Last update from 10.1.1.38 on TenGigabitEthernet5/1, 02:35:37 ago
>   Routing Descriptor Blocks:
>   * 10.1.1.38, from 10.0.0.114, 02:35:37 ago, via TenGigabitEthernet5/
> 1
>       Route metric is 2, traffic share count is 1
>
> #sh mpls interfaces
> Interface              IP            Tunnel   BGP Static Operational
> GigabitEthernet4/2     Yes           No       Yes No     No
>
> #sh mpls ldp neigh
>     Peer LDP Ident: 10.0.0.114:0; Local LDP Ident 10.0.0.113:0
>         TCP connection: 10.0.0.114.1026 - 10.0.0.113.646
>         State: Oper; Msgs sent/rcvd: 174/172; Downstream
>         Up time: 02:20:12
>         LDP discovery sources:
>           Targeted Hello 10.0.0.113 -> 10.0.0.114, active, passive
>         Addresses bound to peer LDP Ident:
>           10.0.0.114 10.1.1.38
>
>
> #sh xconnect all detail             <--- command doesn't exist on SXH5
> Legend: XC ST=Xconnect State, S1=Segment1 State, S2=Segment2 State
> UP=Up, DN=Down, AD=Admin Down, IA=Inactive, NH=No Hardware
> XC ST  Segment 1                         S1 Segment 2       S2
> ------+---------------------------------+--+---------------------------------+--
> DN     ac   Gi4/2(Ethernet)              DN mpls 10.0.0.114:100       DN
>             Interworking: ip                     Local VC label unassigned
>                                                  Remote VC label unassigned
>                                                  pw-class: EtherEncap
>
> Router 2:
>
> mls ip slb purge global
> mls netflow interface
> no mls flow ip
> no mls flow ipv6
> mls cef error action reset
> mpls traffic-eng tunnels
> mpls ldp graceful-restart
> mpls ldp discovery targeted-hello accept
> mpls ip default-route
> mpls label range 50 524000
> mpls label protocol ldp
>
> pseudowire-class EtherEncap
>  encapsulation mpls
>  interworking ip
>
> interface Loopback0
>   ip address 10.0.0.114 255.255.255.255
>
> interface GigabitEthernet4/2
>  mtu 1526
>  no ip address
>  speed nonegotiate
>  mpls ldp discovery transport-address interface
>  mpls bgp forwarding
>  mpls label protocol ldp
>  mpls ip
>  bfd interval 50 min_rx 100 multiplier 3
>  xconnect 10.0.0.113 100 encapsulation mpls
>
> router ospf 1
>  log-adjacency-changes
>  network 10.0.0.114 0.0.0.0 area 0
>  bfd all-interfaces
>  mpls traffic-eng router-id Loopback0
> !
> router bgp 65518
>  bgp router-id 10.0.0.114
>  bgp log-neighbor-changes
>  bgp confederation identifier 100
>  bgp confederation peers 65517
>  neighbor 10.0.0.113 remote-as 65517
>  neighbor 10.0.0.113 ebgp-multihop 255
>  neighbor 10.0.0.113 update-source Loopback0
>  neighbor 10.0.0.113 version 4
>  !
>  address-family ipv4
>   neighbor 10.0.0.113 activate
>   neighbor 10.0.0.113 send-label
>   no auto-summary
>   no synchronization
>   network 10.0.0.114 mask 255.255.255.255
>  exit-address-family
> !
> ip classless
>
> #sh mpls l2 binding
>   Destination Address: 10.0.0.113,  VC ID: 100
>     Local Label:  unassigned.
>     Remote Label: unassigned
>
> #sh mpls l2 vc detail
> Local interface: Gi4/2 down, line protocol down, Ethernet down
>   MPLS VC type is Ethernet, interworking type is IP
>   Destination address: 10.0.0.113, VC ID: 100, VC status: down
>     Output interface: if-?(0), imposed label stack {}
>     Preferred path: not configured
>     Default path: no route
>     No adjacency
>   Create time: 02:11:47, last status change time: 02:11:30
>   Signaling protocol: LDP, peer 10.0.0.113:0 up
>     MPLS VC labels: local unassigned, remote unassigned
>     Group ID: local unknown, remote unknown
>     MTU: local unknown, remote unknown
>     Remote interface description:
>   Sequencing: receive disabled, send disabled
>   VC statistics:
>     packet totals: receive 0, send 0
>     byte totals:   receive 0, send 0
>     packet drops:  receive 0, send 0
>
> #sh ip route 10.0.0.113
> Routing entry for 10.0.0.113/32
>   Known via "ospf 1", distance 110, metric 2, type intra area
>   Last update from 10.1.1.37 on TenGigabitEthernet5/1, 02:34:18 ago
>   Routing Descriptor Blocks:
>   * 10.1.1.37, from 10.0.0.113, 02:34:18 ago, via TenGigabitEthernet5/
> 1
>       Route metric is 2, traffic share count is 1
>
>
> #sh mpls interfaces
> Interface              IP            Tunnel   BGP Static Operational
> GigabitEthernet4/2     Yes           No       Yes No     No
>
> #sh mpls ldp neigh
>     Peer LDP Ident: 10.0.0.113:0; Local LDP Ident 10.0.0.114:0
>         TCP connection: 10.0.0.113.646 - 10.0.0.114.1026
>         State: Oper; Msgs sent/rcvd: 173/175; Downstream
>         Up time: 02:21:08
>         LDP discovery sources:
>           Targeted Hello 10.0.0.114 -> 10.0.0.113, active, passive
>         Addresses bound to peer LDP Ident:
>           10.0.0.113 10.1.1.34  10.1.1.37
>
>
> *Feb  2 06:05:54.223: AToM MGR [10.0.0.114, 100]: Event provision, state ch
> anged from idle to provisioned
> *Feb  2 06:05:54.223: AToM MGR [10.0.0.114, 100]: Provision vc
> *Feb  2 06:05:54.223: AToM LDP [10.0.0.114]: Opening session, 1 clients
> *Feb  2 06:05:54.223: AToM LDP [10.0.0.114]: Session is up
> *Feb  2 06:05:54.227: AToM MGR [10.0.0.114, 100]: Signaling peer-id of VC c
> hanged to 10.0.0.114
> *Feb  2 06:05:54.227: AToM MGR [10.0.0.114, 100]: Event ldp up, state chang
> ed from provisioned to ldp ready
> *Feb  2 06:05:54.227: AToM MGR [10.0.0.114, 100]: Take no action
> *Feb  2 06:05:54.231: AToM MGR [10.0.0.114, 100]: Event SSS service found,
> state changed from ldp ready to ldp ready
> *Feb  2 06:05:54.231: AToM MGR [10.0.0.114, 100]: Take no action
>
> --------------------------------------------
>
> I've spent a few hours googling, searching and other things. It *seems*  
> like this should be super-simple and I'm left thinking I've missed  
> something obvious.
>
> No labels are being assigned even though I believe I've correctly  
> assigned VC ID:100 and agreed to IBGP internal exchange of routing  
> information and labels.
>
> I see 113(R1) sending a targeted hello to 114 (R2) and 114 seeing 113  
> up. The VC is down, even though L3 connectivity exists between the two  
> loopbacks, the mpls config shows no adjacency, no default path, no  
> output interface.
>
> Every sample config I've seen online seems to agree with my config (that  
> didn't still use terms like tagswitching or mpls l2connect), so I am  
> pretty sure I am being completely bone-headed about this. I've matched  
> mtu sizes on both interfaces, so that should be okay as well.
>
> Please tell me where I've taken a wrong turn in MPLS land.
>
> Thanks so much!
>
> LA
> _______________________________________________
> 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