cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1037
Views
5
Helpful
13
Replies
Highlighted
Beginner

IPV6 BGP routes not installed in Routing Table


Hello,

I am peering with various IPV6 peers and i am receiving routes from my peers but I am not installing routes in my routing table.


Below are various outputs (output values are truncated)

1. shows the ipv6 bgp output table, showing a neighbor and amounf of the prefixes received.
2. shows the ipv6 routes received from the neighbor BBBB:BBB:1::a500:BBB:B, (other routes have been removed)
3. shows that the next hops are reachable

I am using an ASR 9001 - S, IOS XR version 5.1.2


RP/0/RSP0/CPU0:#sh bgp ipv6 unicast summary
Mon Feb  1 13:50:59.399 WAT
BGP router identifier XX.XX.XX.XX, local AS number XXXXX
BGP generic scan interval 60 secs
BGP table state: Active
Table ID: 0xe0800000   RD version: 3007
BGP main routing table version 3007
BGP scan interval 60 secs

BGP is operating in STANDALONE mode.


Process       RcvTblVer   bRIB/RIB   LabelVer  ImportVer  SendTblVer  StandbyVer
Speaker            3007          1          0       3007           1           0

Neighbor        Spk    AS MsgRcvd MsgSent   TblVer  InQ OutQ  Up/Down  St/PfxRcd
BBBB:BBB:1::a500:BBB:B
                  0   XXX    3917    3858        0    0    0    1d07h          5

.................................................................

RP/0/RSP0/CPU0:#sh bgp ipv6 unicast neighbors BBBB:BBB:1::a500:BBB:B routes
Mon Feb  1 13:53:23.092 WAT
BGP router identifier XX.XX.XX.XXX, local AS number XXXX
BGP generic scan interval 60 secs
BGP table state: Active
Table ID: 0xe0800000   RD version: 3007
BGP main routing table version 3007
BGP scan interval 60 secs

Status codes: s suppressed, d damped, h history, * valid, > best
              i - internal, r RIB-failure, S stale, N Nexthop-discard
Origin codes: i - IGP, e - EGP, ? - incomplete
   Network            Next Hop            Metric LocPrf Weight Path
*  CCCC:CCC:CCC::/48  BBBB:BBB:1::a500:BBB:B
                                               0             0 XXX i

................................................................


RP/0/RSP0/CPU0:#ping BBBB:BBB:1::a500:BBB:B
Mon Feb  1 13:54:12.683 WAT
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to BBBB:BBB:1::a500:BBB:B, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 123/124/125 ms

13 REPLIES 13
Highlighted
Advisor

What is the output of:

show bgp ipv6 unicast CCCC:CCC:CCC::/48

Could you maybe post the output of:

debug bgp vpnv6 unicast 

and then reset the BGP IPv6 peer.

Highlighted

Hello Philip,

the second suggestion is out of context the original poster is not running 6VPE, it is simply a question of multiple eBGP peerings in IPv6. The path from this neighbor is not the best so it is not installed in the IPv6 routing table.

Hope to help

Giuseppe

Highlighted

I just posted the prefix above as an example.

None of the ipv6 prefixes i am receiving are installed in the routing table, even best

RP/0/RSP0/CPU0:#sh bgp ipv6 unicast
Tue Feb  2 20:32:43.057 WAT
BGP router identifier 44.44.44.444, local AS number 333XX
BGP generic scan interval 60 secs
BGP table state: Active
Table ID: 0xe0800000   RD version: 3531
BGP main routing table version 3531
BGP scan interval 60 secs

Status codes: s suppressed, d damped, h history, * valid, > best
              i - internal, r RIB-failure, S stale, N Nexthop-discard
Origin codes: i - IGP, e - EGP, ? - incomplete
   Network            Next Hop            Metric LocPrf Weight Path
*  2001::/32          2001:FFF:F::a500:FFFF:1
                                             100             0 2222 1111 0000 i
*                     2001:FFF:F::a500:FFFF:1
                                              30             0 6666 29900 i
*                     2001:FFF:F::FFFF:1
                                              40             0 6666 23232 i
*>                    2001:FFF:F::FFFF:1
                                               0             0 9999 i


output truncated

ALl i get in my table are directly connected routes


RP/0/RSP0/CPU0:INT-IXP-A9K-PR-01#sh route ipv6
Tue Feb  2 20:35:19.856 WAT

Codes: C - connected, S - static, R - RIP, B - BGP, (>) - Diversion path
       D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
       N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
       E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP
       i - ISIS, L1 - IS-IS le
vel-1, L2 - IS-IS level-2
       ia - IS-IS inter area, su - IS-IS summary null, * - candidate default
       U - per-user static route, o - ODR, L - local, G  - DAGR
       A - access/subscriber, a - Application route, (!) - FRR Backup path

Gateway of last resort is not set

C    2001:FFF:X::/64 is directly connected,
      2d18h, GigabitEthernet0/0/1/0
L    2001:FFF:X::a503:FFFF:1/128 is directly connected,
      2d18h, GigabitEthernet0/0/1/0
C    2001:FFF:X::/64 is directly connected,
      2d14h, TenGigE0/0/2/1
L    2001:FFF:X::FFFF:1/128 is directly connected,
      2d14h, TenGigE0/0/2/1

Highlighted

Output

RP/0/RSP0/CPU0:#sh bgp ipv6 unicast CCCC:C:CCC::/48
Tue Feb 2 20:08:12.390 WAT
BGP routing table entry for CCCC:C:CCC::/48
Versions:
Process bRIB/RIB SendTblVer
Speaker 2304 2304
Last Modified: Jan 31 06:28:09.353 for 2d13h
Paths: (1 available, best #1)
Not advertised to any peer
Path #1: Received by speaker 0
Not advertised to any peer
ASAS
XXXX:XXX:X::XXX:1 from XXXX:XXX:X::XXX:1 (22.22.222.2)
Origin IGP, metric 0, localpref 100, valid, external, best, group-best, import-candidate
Received Path ID 0, Local Path ID 1, version 2304
Community: 55555:ASAS
Origin-AS validity: not-found

Re;

debug bgp vpnv6 unicast 

I am guessing the syntax for the command in ios xr is different, so i ran the debug command below

"debug bgp address-family vpnv6 unicast" this however did not produce any output and i think that is fine cause i am not receiving any vpnv6 routes, so i ran the debug for ipv6 unicast i.e.
debug bgp address-family ipv6 unicast,

Below is the output


RP/0/RSP0/CPU0:Feb 2 20:14:36.458 WAT: bgp[1054]: %ROUTING-BGP-5-ADJCHANGE_DETAIL : neighbor XXXX:XXX:X::XXX:1 Up (VRF: default; AFI/SAFI: 2/1) (AS: ASAS)

RP/0/RSP0/CPU0:Feb 2 20:14:37.608 WAT: bgp[1054]: [default-rtr] (ip6u): ===bgp_add_path===: net=3333:0:333::/48, nver=3527, nfl=0x04000001: newpath=0x51097a10(XXXX:XXX:X::XXX:1/128,XXXX:XXX:X::XXX:1,0,0x4000000000020001): link=1, pcount=1

RP/0/RSP0/CPU0:Feb 2 20:14:37.608 WAT: bgp[1054]: [default-rtr] (ip6u): Not an import multipath -> import multipath for 3333:0:333::/48 from XXXX:XXX:X::XXX:1 (path: ASAS)

RP/0/RSP0/CPU0:Feb 2 20:14:37.608 WAT: bgp[1054]: [default-rtr] (ip6u): ===bgp_bestpath===: net=2620:0:870::/48, nver=3527, nfl=0x04000001: lastpath=0x51097a10(XXXX:XXX:X::XXX:1/128,XXXX:XXX:X::XXX:1,0,0x4000000000020001): PELEM=0x0(lpathid=0, ver=0, fl=0x00000000, path=0x0)

RP/0/RSP0/CPU0:Feb 2 20:14:37.608 WAT: bgp[1054]: [default-rtr] (ip6u): ===bgp_calculate_bestpath===: net=3333:0:333::/48, nver=3527, nfl=0x04040001, pelemver=3527: oldbest=0x0(NULL,::,0), newbest=0x51097a10(XXXX:XXX:X::XXX:1/128,2001:XXXX:XXX:X::XXX:10), bumpv=1

RP/0/RSP0/CPU0:Feb 2 20:14:37.608 WAT: bgp[1054]: [default-rtr] (ip6u): ===bgp_change_bestpath===: net=3333:0:333::/48, nver=3527, nfl=0x04040001: oldbest=0x0: PELEM=0x510b8e04(lpathid=1, ver=3527, fl=0x00001101, path=0x51097a10)

RP/0/RSP0/CPU0:Feb 2 20:14:37.608 WAT: bgp[1054]: [default-rtr] (ip6u): ===bnet_chgflags_update===: net=3333:0:333::/48, nver=3527, nfl=0x04040001: oldtblattr=0x0, oldnh=0x0: PELEM=0x510b8e04(lpathid=1, ver=3527, fl=0x00001101, path=0x51097a10)

RP/0/RSP0/CPU0:Feb 2 20:14:37.608 WAT: bgp[1054]: [default-rtr] (ip6u): ===bgp_bestpath===: COMPLETE: net=3333:0:333::/48, nver=3527, nfl=0x04044001: pselect=0, bumpv=1(1:1,0,0,0)

Highlighted

Could you post the output of:

show ipv6 route CCCC:C:CCC::/48
Highlighted

RP/0/RSP0/CPU0:#sh route ipv6 CCCC:C:CCC::/48
Tue Feb  2 20:40:16.039 WAT

% Network not in table

Highlighted

Hello,

Had to contact Cisco TAC, turned out to be a software bug.

"CSCuq79148 - BGP label thread is stuck in CDM init:"

This bug is fixed starting with IOS XR 5.1.3 so an IOS upgrade is required.

The workaround is to reset the BGP process which i did and i can install IPV6 routes now.

Thank you all

Highlighted

Hello M-oyegunle,

you have been kind to provide feedback on this issue. Now this thread has an happy end.

So it was a SW bug.

Best Regards

Giuseppe

Highlighted

Hello, m-o

there is a single BGP advertisement for the prefix and it is considered best. it should be installed.

is the prefix learned in some other way a static route?

RP/0/RSP0/CPU0:#sh bgp ipv6 unicast CCCC:C:CCC::/48
Tue Feb 2 20:08:12.390 WAT
BGP routing table entry for CCCC:C:CCC::/48
Versions:
Process bRIB/RIB SendTblVer
Speaker 2304 2304
Last Modified: Jan 31 06:28:09.353 for 2d13h
Paths: (1 available, best #1)
Not advertised to any peer
Path #1: Received by speaker 0
Not advertised to any peer
ASAS
XXXX:XXX:X::XXX:1 from XXXX:XXX:X::XXX:1 (22.22.222.2)
Origin IGP, metric 0, localpref 100, valid, external, <<<<best<<<, group-best, import-candidate
Received Path ID 0, Local Path ID 1, version 2304
Community: 55555:ASAS
Origin-AS validity: not-found

Do you know that you need a policy written in RPL to accept the prefix from an eBGP peer in IOS XR?

Have you applied a policy like 

route-policy  PASS-ALL

pass

end-policy

see

http://www.cisco.com/c/en/us/td/docs/routers/asr9000/software/asr9k_r5-3/routing/configuration/guide/b_routing_cg53xasr9k/b_routing_cg53xasr9k_chapter_010.html#con_1197962

Hope to help

Giuseppe

Highlighted

No the prefix is not learned via any other way.

and yes i have route policies applied in both inbound and outbound directions

Highlighted
Advisor

You definately have ipv6 unicast routing enabled?

ipv6 unicast-routing
Highlighted
Beginner

ipv6 unicast-routing is enabled by default in IOS_XR yeah?

Highlighted

Hello, m-oyegunle yes ipv6 unicast routing is enabled by default in IOS XR.

I don't see the '>' near the CCCC:CCCC:CCC::/48 prefix, you may be learning the prefix in some other way?

the missing '>' means the BGP is not considered best by BGP.

Your router might have installed the prefix from another eBGP session.

Hope to help

Giuseppe