MPLS LDP not UP

[vin.marco]

Hello experts
I have a problem with an MPLS adjacency.
I have an ASR900 connected to a 10Gb C9300, MPLS and OSPF are enabled in this interface, adjacencies are created and everything works.
When the C9300 is restarted, the MPLS and OSPF adjacencies are not recreated, they are only created if I enter ASR900 and delete and re-enter the MPLS IP command.
Can you tell me why?

[David Ruess]

Hello,

So when your 9300 comes up your OSPF Adjacencies do NOT come up? You need to fix that problem first as the MPLS rides over the routing protocols to for adjacencies. Make sure your loopbacks are also advertised into OSPF as that's what MPLS uses for neighbors unless otherwise configured. Based on the synopsis it seems as if the configuration isnt saved before rebooting.

Can you provide the running config of both devices?

-David

[MHM Cisco World]

I think the opsf is UP but the mpls not re-established after SW reboot.

This ldp-igp sync 

https://www.cisco.com/c/en/us/td/docs/ios-xml/ios/mp_ldp/configuration/15-s/mp-ldp-15-s-book/mp-ldp-igp-synch.html

MHM

[vin.marco]

The ospf process does not go up because there is no mpls adjacency.

The loopback was inserted into the ospf process, as I said, before the reboot of the c9300 all ospf and mpls worked.

After the reboot nothing goes up anymore.

[MHM Cisco World]

Do 

Show ip ospf neighbor 

Show ip ospf interface breif 

Are you see neighbor? Are the interface are UP DR/BDR/DOTHER?

MHM

[MHM Cisco World]

Also share 

Show ip ospf mpls ldp interface <ospf interface>

MHM

[David Ruess]

MPLS is independent of OSPF so the fact the adjacency does not come up is not a problem with MPLS its a problem with OSPF.

[Harold Ritter]

Hi @vin.marco ,

Do you use the MPLS LDP IGP synchronization feature by any chance ("mpls ldp sync" under the ospf process)?

If so, could please provide the output for the following command for both devices:

show mpls ldp igp sync

Regards,

[vin.marco]

Thanks all for support, below the output first the C9300 reload:

OSPF:

interface TenGigabitEthernet0/1/7
 description Link to TEST  
 mtu 9198
 ip address 10.86.186.253 255.255.255.252
 ip ospf network point-to-point
 ip ospf 1000 area 0
 cdp enable
 mpls ip
 bfd template RID
!
interface Loopback86
 description Loopback vrf Global 
 ip address 10.86.1.23 255.255.255.255
end

ASR900#sh ip ospf neighbor 

Neighbor ID     Pri   State           Dead Time   Address         Interface
10.86.1.105       0   FULL/  -        00:00:34    10.86.186.254   TenGigabitEthernet0/1/7

ASR900#sh ip ospf interface l86
Loopback86 is up, line protocol is up 
  Internet Address 10.86.1.23/32, Interface ID 25, Area 0
  Attached via Network Statement
  Process ID 1000, Router ID 10.86.1.23, Network Type LOOPBACK, Cost: 1
  Topology-MTID    Cost    Disabled    Shutdown      Topology Name
        0           1         no          no            Base
  Loopback interface is treated as a stub Host

RMC4DFDI12002#sh ip ospf interface t0/1/7
TenGigabitEthernet0/1/7 is up, line protocol is up 
  Internet Address 10.86.186.253/30, Interface ID 23, Area 0
  Attached via Interface Enable
  Process ID 1000, Router ID 10.86.1.23, Network Type POINT_TO_POINT, Cost: 1
  Topology-MTID    Cost    Disabled    Shutdown      Topology Name
        0           1         no          no            Base
  Enabled by interface config, including secondary ip addresses
  Transmit Delay is 1 sec, State POINT_TO_POINT, BFD enabled
  Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5
    oob-resync timeout 40
    Hello due in 00:00:05
  Supports Link-local Signaling (LLS)
  Cisco NSF helper support enabled
  IETF NSF helper support enabled
  Can be protected by per-prefix Loop-Free FastReroute
  Can be used for per-prefix Loop-Free FastReroute repair paths
  Not Protected by per-prefix TI-LFA
  Index 1/9/9, flood queue length 0
  Next 0x0(0)/0x0(0)/0x0(0)
  Last flood scan length is 1, maximum is 20
  Last flood scan time is 0 msec, maximum is 4 msec
  Neighbor Count is 1, Adjacent neighbor count is 1 
    Adjacent with neighbor 10.86.1.105
  Suppress hello for 0 neighbor(s)

MPLS

ASR900#sh mpls ldp neighbor

Peer LDP Ident: 10.86.1.105:0; Local LDP Ident 10.86.1.23:0
        TCP connection: 10.86.1.105.38041 - 10.86.1.23.646
        State: Oper; Msgs sent/rcvd: 4682/4896; Downstream
        Up time: 2d17h
        LDP discovery sources:
          TenGigabitEthernet0/1/7, Src IP addr: 10.86.186.254
          Targeted Hello 10.86.1.23 -> 10.86.1.105, active, passive
        Addresses bound to peer LDP Ident:
          10.86.1.105     10.86.187.85    10.86.186.254   


ASR900#sh ip ospf mpls ldp interface tenGigabitEthernet 0/1/7
TenGigabitEthernet0/1/7
  Process ID 1000, Area 0
  LDP is not configured through LDP autoconfig
  LDP-IGP Synchronization : Required
  Holddown timer is not configured
  Interface is up 
ASR900#

RMC4DFDI12002#sh mpls ldp igp sync 
    TenGigabitEthernet0/1/7:
        LDP configured; LDP-IGP Synchronization enabled.
        Sync status: sync achieved; peer reachable.
        Sync delay time: 0 seconds (0 seconds left)
        IGP holddown time: infinite.
        Peer LDP Ident: 10.86.1.105:0
        IGP enabled: OSPF 1000
		
	

Configuration

mpls ldp nsr
mpls ldp graceful-restart
mpls ldp session protection
mpls ldp label
mpls ldp router-id Loopback86 force
!
router ospf 1000
 router-id 10.86.1.23
 nsr
 nsf ietf
 fast-reroute per-prefix enable area 0 prefix-priority high
 fast-reroute per-prefix enable prefix-priority high
 fast-reroute keep-all-paths
 redistribute static
 network 10.86.1.23 0.0.0.0 area 0
 network 10.86.186.8 0.0.0.3 area 0
 network 10.86.186.16 0.0.0.3 area 0
 network 10.86.186.24 0.0.0.3 area 0
 network 10.86.186.28 0.0.0.3 area 0
 network 10.86.186.60 0.0.0.3 area 0
 network 10.86.248.23 0.0.0.0 area 0
 bfd all-interfaces
 mpls ldp sync
!

today i tried the reload C9300 but i had no problems 

[MHM Cisco World]

As I and @Harold Ritter suspect 
there is sync 
you use LO as MPLS router-id and MPLS ldp need reachability of router-id to work, and so it need ospf which is not work since you run sync.
so fast solution is use static route for LO user as MPLS LDP router-id and this make MPLS work and follow by OSPF 

try this way 

MHM

[vin.marco]

in fact I configured a static route only for the LO and everything went back up, but what I don't understand is why it happens.

[MHM Cisco World]

It like egg or chicken

The ospf depends on mpls to be up

The mpls depend of ospf to advertise LO to be UP

And this loop end in both not work. 

So as workaround we use static route for LO use as mpls ldp router-id to make mpls up then ospg automatically follow mpls and be up too.

Goodluck 

MHM

[MHM Cisco World]

Indeed' holdtime can solution' I already try this in lab today' but the max metric is stop me.

I need to know the effect of max metric on both ospf and ldp.

Update you soon 

Thanks 

MHM

[Harold Ritter]

Hi @MHM Cisco World ,

Configuring "mpls ldp igp sync holddown" definitely solves this issue. 

I had a large service provider customer that run into this issue a long time ago. They had an aggregate static route configured on the edge devices and had "mpls ldp igp sync holddown" set to its default value (infinite). This caused an outage in their network. They set the "mpls ldp igp sync holddown" short value and the issue was fixed.

Regards, 

[MHM Cisco World]

@vin.marco 

Hi Mr @Harold Ritter I know holdtime but I need to confirm that LO is learn when the one side send it with max metric 
so that why I postponed my suggestion BUT, since I have little time NOW I run lab and confirm that the LO that MPLS LDP use as router-ID is advertise by OSPF after holdtime is end, and the Peer can use it if there is no other path to this LO 
I make holdtime 30 ms (it small I know)
the R1 advertise the 1.1.1.1 to R2, LO 1.1.1.1 is MPLS LDP router-id and you can see it metric is MAX still R2 use it since it dont have anyother path to LO. 
HOPE this solve your issue

MHM