cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
3167
Views
7
Helpful
7
Replies

MPLS Config Help

xcz504d1114
Level 4
Level 4

This is driving me insane, it's not a difficult problem, I have a loopback in the VRF on both cores, configurations were copy and pasted to ensure they were identical, BGP peer's are up, redistribution is working fine, but I cannot ping between the loopbacks!

I have 2 6509's, connected with a 802.1q trunk

Configuration:

ip vrf Testing
rd 111:1
route-target both 111:1

int vlan 400

ip address 10.65.65.2 255.255.255.0

mpls ip

int loopback 0

ip address 10.65.64.255

router eigrp 64
no auto-summary
network 10.0.0.0 0.31.255.255
network 10.32.0.0 0.15.255.255
network 10.48.0.0 0.7.255.255
network 10.64.0.0 0.63.255.255
network 10.128.0.0 0.127.255.255

address-family ipv4 vrf Testing
no auto-summary
network 10.0.0.0 0.31.255.255
network 10.32.0.0 0.15.255.255
network 10.48.0.0 0.7.255.255
network 10.64.0.0 0.63.255.255
network 10.128.0.0 0.127.255.255
default-metric 10000 100 255 1 1500
autonomous 111
redistribute bgp 65064

router bgp 65064
no auto-summ
no synch
network 0.0.0.0
neighbor R peer-group
neighbor R remote-as 65064
neighbor R update-source loop 0
neighbor 10.65.64.254 peer-group R

address-family vpnv4
neighbor 10.65.64.254 peer-group R
neighbor R send-community both

address-family ipv4 vrf Testing
no auto-summ
no synch
redistribute eigrp 111

int loopback 99

ip vrf forward Testing

ip address 10.111.1.1 255.255.255.0

Router 1:

show ip bgp neighbor:

BGP neighbor is 10.65.64.254,  remote AS 65064, internal link
Member of peer-group R for session parameters
  BGP version 4, remote router ID 10.65.64.254
  BGP state = Established, up for 03:36:33

For address family: VPNv4 Unicast
  BGP table version 10, neighbor version 10/0
  Output queue size : 0
  Index 1, Offset 0, Mask 0x2
  1 update-group member
  R peer-group member
  Community attribute sent to this neighbor
                                 Sent       Rcvd
  Prefix activity:               ----       ----
    Prefixes Current:               2          1 (Consumes 68 bytes)

show ip route vrf Testing:

Gateway of last resort is not set

     10.0.0.0/24 is subnetted, 3 subnets
C       10.111.2.0 is directly connected, Loopback99
C       10.111.22.0 is directly connected, Loopback98
B       10.111.1.0 [200/0] via 10.65.64.254, 03:38:30

show mpls ldp neigh:

Peer LDP Ident: 10.65.64.254:0; Local LDP Ident 10.65.64.255:0
        TCP connection: 10.65.64.254.646 - 10.65.64.255.36970
        State: Oper; Msgs sent/rcvd: 793/795; Downstream
        Up time: 02:12:39
        LDP discovery sources:
          Vlan400, Src IP addr: 10.65.65.3

Router 2:

show ip bgp neighbor:

BGP neighbor is 10.65.64.255,  remote AS 65064, internal link
Member of peer-group R for session parameters
  BGP version 4, remote router ID 10.65.64.255
  BGP state = Established, up for 03:39:34

For address family: VPNv4 Unicast
  BGP table version 10, neighbor version 10/0
  Output queue size : 0
  Index 1, Offset 0, Mask 0x2
  1 update-group member
  R peer-group member
  Community attribute sent to this neighbor
                                 Sent       Rcvd
  Prefix activity:               ----       ----
    Prefixes Current:               1          2 (Consumes 136 bytes)
    Prefixes Total:                 1          3
    Implicit Withdraw:              0          1
    Explicit Withdraw:              0          0
    Used as bestpath:             n/a          2
    Used as multipath:            n/a          0

show ip route vrf Testing:

Gateway of last resort is not set

     10.0.0.0/24 is subnetted, 3 subnets
B       10.111.2.0 [200/0] via 10.65.64.255, 03:41:22
B       10.111.22.0 [200/0] via 10.65.64.255, 02:35:31
C       10.111.1.0 is directly connected, Loopback99

From router 2:

R2#ping vrf Testing 10.111.1.1

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 10.111.1.1, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/1 ms
R2#ping vrf Testing 10.111.2.1

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 10.111.2.1, timeout is 2 seconds:
.....
Success rate is 0 percent (0/5)

1 Accepted Solution

Accepted Solutions

Hello again,

R1#show ip cef vrf Testing 10.111.1.1 detail
10.111.1.0/24, epoch 3, flags rib defined all labels
   NetFlow: Origin AS 0, Peer AS 0, Mask Bits 24
   recursive via 10.65.64.254 label 26
     nexthop 10.64.1.254 Port-channel1 unusable: no label   <--- !!


So, you have the vpnv4 route and the prefix. But, the recursion to the IGP global route

and the associated label failed in producing an MPLS label. This usually means that

MPLS is broken in the core/global context across the outgoing interface.

What does the following give you?

show ip route 10.65.64.254

show ip cef 10.65.64.254

show running interf  Port-channel1

show mpls ldp disc

show mpls ldp neighbor

Thanks,

Luc

View solution in original post

7 Replies 7

Giuseppe Larosa
Hall of Fame
Hall of Fame

Hello,

post  a show module

be aware that not all hardware combinations on C6500 support MPLS full features: for example, you need SUP720BXL or better the sup720B or Base is not eligible for MPLS services.

This can result in lack of connectivity in the forwarding plane

Hope to help

Giuseppe

Thanks for the reply! Here is the output of the show mod, looks like I am running the VS720 3C model, which does support MPLS.

Mod Ports Card Type                              Model              Serial No.
--- ----- -------------------------------------- ------------------ -----------
  1   16  16 port 1000mb GBIC ethernet           WS-X6416-GBIC      SAL0603GPJZ
  2   16  16 port 1000mb GBIC ethernet           WS-X6416-GBIC      SAL0603GQ4T
  4   48  48-port 10/100/1000 RJ45 EtherModule   WS-X6148A-GE-45AF  SAL1432QAFS
  5    5  Supervisor Engine 720 10GE (Active)    VS-S720-10G        SAL1317NVA2
  6    0  2 port adapter Enhanced FlexWAN        WS-X6582-2PA       JAE1353T7OV
  7    0  2 port adapter Enhanced FlexWAN        WS-X6582-2PA       JAE115182E9
  9   48  48 port 10/100/1000mb EtherModule      WS-X6148-GE-45AF   SAL1306JB8F

Mod MAC addresses                       Hw    Fw           Sw           Status
--- ---------------------------------- ------ ------------ ------------ -------
  1  0008.207a.0728 to 0008.207a.0737   2.0   5.4(2)       8.7(0.22)BUB Ok
  2  0008.207a.0c68 to 0008.207a.0c77   2.0   5.4(2)       8.7(0.22)BUB Ok
  4  f866.f2b2.5f20 to f866.f2b2.5f4f   4.1   8.4(1)       8.7(0.22)BUB Ok
  5  0023.33ab.2608 to 0023.33ab.260f   3.1   8.5(3)       12.2(33)SXH7 Ok
  6  8843.e129.3080 to 8843.e129.30bf   2.4   12.2(33)SXH7 12.2(33)SXH7 Ok
  7  001e.7a5b.b400 to 001e.7a5b.b43f   2.3   12.2(33)SXH7 12.2(33)SXH7 Ok
  9  0024.c40f.7908 to 0024.c40f.7937   8.2   7.2(1)       8.7(0.22)BUB Ok

Mod  Sub-Module                  Model              Serial       Hw     Status
---- --------------------------- ------------------ ----------- ------- -------
  4  IEEE Voice Daughter Card    WS-F6K-48-AF       SAL1432Q5J7  2.5    Ok
  5  Policy Feature Card 3       VS-F6K-PFC3C       SAL1321QE3K  1.0    Ok
  5  MSFC3 Daughterboard         VS-F6K-MSFC3       SAL1319PRPG  1.1    Ok
  9  IEEE Voice Daughter Card    WS-F6K-48-AF       SAL1306J1QX  2.3    Ok

Mod  Online Diag Status
---- -------------------
  1  Pass
  2  Pass
  4  Pass
  5  Pass
  6  Pass
  7  Pass
  9  Pass

Hi,

Can you ping with specifying the source IP address of the loopback in the VRF?

Let's have a look at this output:

show ip bgp vpnv4 all labels

show mpls forwarding detail

show mpls interface detail

show ip cef vrf Testing 10.111.1.1 detail

on both routers (reversing the IP address for that last command)

Thanks,

Luc

Thanks for the reply, even with specifying a source address within the VRF I am unable to successfully ping.

R1#show ip bgp vpnv4 all labels
   Network          Next Hop      In label/Out label
Route Distinguisher: 111:1 (Testing)
   10.111.1.0/24    10.65.64.254    nolabel/26
   10.111.2.0/24    0.0.0.0         IPv4 VRF Aggr:26/nolabel(Testing)
   10.111.22.0/24   0.0.0.0         IPv4 VRF Aggr:26/nolabel(Testing)

The forwarding detail is actually a large output (several hundred interfaces active on this router), so I grabbed the Testing VRF and a random label:

26         Pop Label  IPv4 VRF[V]      0             aggregate/Testing
        MAC/Encaps=0/0, MRU=0, Label Stack{}
        VPN route: Testing
        No output feature configured

31         No Label   10.6.16.0/24     0             Po1        10.64.1.254
        MAC/Encaps=14/14, MRU=1504, Label Stack{}
        0024509DE8000023EA356C000800
        No output feature configured
    Per-destination load-sharing, slots: 0 4 8 12
           No Label   10.6.16.0/24     0             Vl488      10.66.80.3
        MAC/Encaps=14/14, MRU=1504, Label Stack{}
        0024509DE8000023EA356C000800
        No output feature configured
    Per-destination load-sharing, slots: 1 5 9 13
           No Label   10.6.16.0/24     0             Vl493      10.66.85.3
        MAC/Encaps=14/14, MRU=1504, Label Stack{}
        0024509DE8000023EA356C000800
        No output feature configured
    Per-destination load-sharing, slots: 2 6 10 14
           No Label   10.6.16.0/24     0             Vl505      10.66.97.3
        MAC/Encaps=14/14, MRU=1504, Label Stack{}
        0024509DE8000023EA356C000800
        No output feature configured
    Per-destination load-sharing, slots: 3 7 11 15

R1#show mpls int detail
Interface Vlan400:
        IP labeling enabled (ldp)
        LSP Tunnel labeling not enabled
        BGP labeling not enabled
        MPLS operational
        MTU = 1500

R1#show ip cef vrf Testing 10.111.1.1 detail
10.111.1.0/24, epoch 3, flags rib defined all labels
  NetFlow: Origin AS 0, Peer AS 0, Mask Bits 24
  recursive via 10.65.64.254 label 26
    nexthop 10.64.1.254 Port-channel1 unusable: no label

R2#show ip bgp vpnv4 all labels
   Network          Next Hop      In label/Out label
Route Distinguisher: 111:1 (Testing)
   10.111.1.0/24    0.0.0.0         IPv4 VRF Aggr:26/nolabel(Testing)
   10.111.2.0/24    10.65.64.255    nolabel/26
   10.111.22.0/24   10.65.64.255    nolabel/26

26         Pop Label  IPv4 VRF[V]      0             aggregate/Testing
        MAC/Encaps=0/0, MRU=0, Label Stack{}
        VPN route: Testing
        No output feature configured

37         No Label   10.6.124.0/24    0             Se7/1/1    point2point
        MAC/Encaps=4/4, MRU=4474, Label Stack{}
        0F000800
        No output feature configured

R2#show mpls int detail
Interface Vlan400:
        IP labeling enabled (ldp)
        LSP Tunnel labeling not enabled
        BGP labeling not enabled
        MPLS operational
        MTU = 1500

R2#show ip cef vrf Testing 10.111.2.1 detail
10.111.2.0/24, epoch 5, flags rib defined all labels
  NetFlow: Origin AS 0, Peer AS 0, Mask Bits 24
  recursive via 10.65.64.255 label 26
    nexthop 10.64.1.253 Port-channel1 unusable: no label

Hello again,

R1#show ip cef vrf Testing 10.111.1.1 detail
10.111.1.0/24, epoch 3, flags rib defined all labels
   NetFlow: Origin AS 0, Peer AS 0, Mask Bits 24
   recursive via 10.65.64.254 label 26
     nexthop 10.64.1.254 Port-channel1 unusable: no label   <--- !!


So, you have the vpnv4 route and the prefix. But, the recursion to the IGP global route

and the associated label failed in producing an MPLS label. This usually means that

MPLS is broken in the core/global context across the outgoing interface.

What does the following give you?

show ip route 10.65.64.254

show ip cef 10.65.64.254

show running interf  Port-channel1

show mpls ldp disc

show mpls ldp neighbor

Thanks,

Luc

You are right on the money, right after I posted that, I went to figure out why port-channel 1 was showing no label, and I wasn't running a port-channel between my cores! I added mpls ip to the port-channel interface going into my datacenter and it worked like a charm! Thanks for the help in getting me there! That show ip cef vrf command definitely did it!

Craig

After I added the mpls ip to the port-channel interface and did my testing, I thought about why my data path wasn't the path I expected it to be. We upgraded our backbone to ten gig two weeks ago, the port channel properly scaled, but the trunk port between the cores did not scale to the VLAN's, the VLAN's still have a default speed of 1gig, so the EIGRP believes the optimal path to the other core is through my datacenter instead of the directly connected link.

Thanks again for your help!

Craig

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: