I’m in final preparation for my second attempt and I have been doing a lot of troubleshooting scenarios lately. I created a MPLS topology in GNS3 and sent it to my friend Darren for testing. He is taking his lab very soon and he performed well on this lab. The lab contains multiple faults but I won’t say how many since that would spoil some of the surprise.

The assignment is to make sure CE1 can ping CE2 loopback 6.6.6.6.

Post in comments what you did to make it work or if you need a hint to get you going in the right direction. You need to edit the .net file to use your own working dir and IOS image. You need IOS images for 3725 and 7200. Start with the configurations provided by importing the configs or simply pasting them in whatever you prefer but you should not look at the startup config before starting.

Download the .net and config files here.

This is what the topology looks like.

MPLS troubleshooting scenario
Tagged on:                     

19 thoughts on “MPLS troubleshooting scenario

  • April 23, 2012 at 5:49 pm
    Permalink

    We’ve got…
    – CEF turned off on P1
    – P2 not advertising labels for 4.4.4.4
    – PE2 not redistributing OSPF into BGP

    Think that’s all, I should have written it all down though.

    Reply
    • April 23, 2012 at 7:05 pm
      Permalink

      Good start but that is not all 🙂 Did you find it interesting?

      Reply
  • April 23, 2012 at 8:59 pm
    Permalink

    Hey Daniel, that’s a nice one. Got ping working. I guess the constant LDP neighbor flapping on PE1, P1, P2, PE2 isn’t part of TS since it looks more like a Dynamips issue (CPU is under 10%). I’ve increased the LDP discovery timers and got a somewhat stable topology.

    Reply
    • April 23, 2012 at 9:17 pm
      Permalink

      Hi,

      I don’t think there should be flapping but there is one parameter that might cause that. I’ll talk to you on chat and we can discuss what you found 🙂

      Reply
    • April 27, 2012 at 9:25 am
      Permalink

      hi
      I had the same dynamips problem as you with LDP sessions always flapping
      I removed the “idlemax” command on .net file which resolved issue

      Reply
  • April 24, 2012 at 4:51 am
    Permalink

    hi,
    i still can’t ping to 6.6.6.6 but CE1 have recieved advertise from CE2

    CE1
    R1#sh ip route
    Codes: C – connected, S – static, R – RIP, M – mobile, B – BGP
    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
    i – IS-IS, su – IS-IS summary, L1 – IS-IS level-1, L2 – IS-IS level-2
    ia – IS-IS inter area, * – candidate default, U – per-user static route
    o – ODR, P – periodic downloaded static route

    Gateway of last resort is not set

    35.0.0.0/24 is subnetted, 1 subnets
    C 35.35.35.0 is directly connected, FastEthernet1/0
    5.0.0.0/32 is subnetted, 1 subnets
    C 5.5.5.5 is directly connected, Loopback0
    6.0.0.0/32 is subnetted, 1 subnets
    O IA 6.6.6.6 [110/3] via 35.35.35.3, 01:04:22, FastEthernet1/0
    11.0.0.0/32 is subnetted, 1 subnets
    O E2 11.11.11.11 [110/1] via 35.35.35.3, 01:04:17, FastEthernet1/0
    12.0.0.0/32 is subnetted, 1 subnets
    O E2 12.12.12.12 [110/1] via 35.35.35.3, 01:04:17, FastEthernet1/0
    46.0.0.0/24 is subnetted, 1 subnets
    O IA 46.46.46.0 [110/2] via 35.35.35.3, 01:04:22, FastEthernet1/0

    CE2
    R6#sh ip route
    Codes: C – connected, S – static, R – RIP, M – mobile, B – BGP
    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
    i – IS-IS, su – IS-IS summary, L1 – IS-IS level-1, L2 – IS-IS level-2
    ia – IS-IS inter area, * – candidate default, U – per-user static route
    o – ODR, P – periodic downloaded static route

    Gateway of last resort is not set

    35.0.0.0/24 is subnetted, 1 subnets
    O IA 35.35.35.0 [110/2] via 46.46.46.4, 01:29:27, FastEthernet1/0
    5.0.0.0/32 is subnetted, 1 subnets
    O IA 5.5.5.5 [110/3] via 46.46.46.4, 01:27:18, FastEthernet1/0
    66.0.0.0/32 is subnetted, 1 subnets
    C 66.66.66.66 is directly connected, Loopback1
    6.0.0.0/32 is subnetted, 1 subnets
    C 6.6.6.6 is directly connected, Loopback0
    11.0.0.0/32 is subnetted, 1 subnets
    O E2 11.11.11.11 [110/1] via 46.46.46.4, 01:27:13, FastEthernet1/0
    12.0.0.0/32 is subnetted, 1 subnets
    O E2 12.12.12.12 [110/1] via 46.46.46.4, 01:27:13, FastEthernet1/0
    46.0.0.0/24 is subnetted, 1 subnets
    C 46.46.46.0 is directly connected, FastEthernet1/0

    maybe flapping on my GNS3

    Reply
    • April 24, 2012 at 6:30 am
      Permalink

      Hi,

      Good job but if you can’t ping you still have some errors to find 🙂 Dig deep and think what is needed for MPLS to work.

      Reply
  • April 24, 2012 at 10:09 am
    Permalink

    Yeah it was good, I’ll have to go over it again as I definitely got it all working, I probably just fixed something and forgot about 😛

    Reply
  • April 24, 2012 at 11:05 am
    Permalink

    Ah yes, forgot that the LDP router-id needed to be reconfigured on P1 to not use the highest loopback but loopback0 instead. I had done this yesterday but forgot all about it 😛

    Reply
    • April 24, 2012 at 11:36 am
      Permalink

      Messing with router-IDs is a really good way of testing how well someone knows a protocol and you have plenty to choose from, OSPF, LDP, MSDP etc 🙂

      Reply
  • April 27, 2012 at 9:38 am
    Permalink

    is there something with virtual link md5 key ?
    there was a space on PE2 so I add to type it again to have area 0 routes added into OSPF.

    Reply
    • April 27, 2012 at 9:44 am
      Permalink

      Indeed 🙂

      I put that error there to demonstrate misconfiguration of virtual link and that in some cases it might stay up even though authentication is mismatching.

      Reply
      • April 27, 2012 at 4:33 pm
        Permalink

        yes, very useful
        I was surprised to see the status staying up while the session was down

        thanks for sharing this, your forum is great !
        best wishes for the upcoming attempt

        Reply
  • May 4, 2012 at 9:12 pm
    Permalink

    This was great post. BTW, when will you take your second attempt?. I’m sure you will pass this time.

    Reply
    • May 5, 2012 at 7:30 am
      Permalink

      Hi,

      Unfortunately I did not pass again. I will post briefly about it now. Good that you liked my post 🙂

      Reply
  • October 30, 2012 at 3:44 pm
    Permalink

    Another thing that we can do is to remove the virtual link and do “capability vrf-lite” under the ospf 2 process on PE2…this will get the 66.66.66.66 loopback installed as a O IA route instead of just an O though. Anyway, I did not find this step to be necessary in order to get reachability.

    Enable cef, change the router-id coming off the wrong loopback on the P router, advertise the labels for the 4.4.4.4 loopback on the other P router, and redistribute ospf on PE2 into bgp.

    Nice lab Daniel! had fun!

    Reply
  • September 13, 2015 at 7:20 pm
    Permalink

    I know this is an old post, but thanks for the lab!

    Reply
  • February 1, 2016 at 5:29 pm
    Permalink

    Same here, just went through it and it took me 45 minutes to resolve. I’m curious how long it took others? I think 45 min. was way too long for this, so definitely helped evaluate where i’m at in my studies. Thank you!

    My train of thought was this:
    1) Okay, just like i’d do on the job, how should I approach this? Typically a user issue, so I’ll start from the PECE connection.
    2) Okay, that looks good, so must be something in the MPLS network. Is basic MPLS functioning properly? No! Okay lemme fix that…
    3) Now that basic MPLS is good now (including the IGP), is BGP working between the PE routers? Er…kinda…need to fix that!
    4) Okay, the MPLS network (including IGP), and BGP are functioning correctly now. I should be able to re-test the ping…..YES! WORKS!

    I wonder now if I can cut down my my troubleshooting process. I got stuck in particular with CEF. It’s always the easy stuff that gets me, hah.

    Nice lab, well done.

    Reply
  • February 17, 2019 at 7:17 pm
    Permalink

    1. The need of virtual link not required at all , but put it intentionally.
    2. Cef was not enable on one.
    3. Some routes did not advertise.
    4. No mpls advertise-label was configured on P2.
    I set up the entire new range on every router and force all the router to use loop 0 but did not do it on the P1 😔

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *