Training and Certification

 View Only
last person joined: 2 days ago 

How to get the most from Juniper's education services and get advice on your certification journey.
  • 1.  OSPF Multi-area Adjacency

    Posted 05-21-2024 22:48

    Hi,

    I started learny JNCIS-SP certification. Has I remember,  last time I did my time I did my CCNP, an interface can't belong more than one interface, it can be a feature added after 2016 maybe but doesn't matters.  I seen now, it somethings been standardized after initial release of OSPF.

    My question is what would be the benefit of choosing an area as primary over the other for a connection between 2 ABRs, I notice that Cisco is doing the same using different command than Juniper but getting a similar result at the end, for compatibility purpose I suppose.  There only one physical connection between the 2 routers, so how to properly choose the Primary area and Secondary area(s) ?

    Thank you



    ------------------------------
    Xine
    ------------------------------


  • 2.  RE: OSPF Multi-area Adjacency

    Posted 03-04-2025 10:28

    Hi @xine32

    The benefit of choosing an area link as primary over the other for a connection between 2 ABRs is that one of your links can be selected as the preferred path for the area where it was configured as secondary if the main path in that specific area is not available.

    Let's see the below example:

    If R1 wants to reach R3's loopback IP 3.3.3.3, the the shortest path in area 100 is to reach it is through interface lt-0/0/0.1, the "show route" command confirms this.

    labroot@R1:R1> show route 3.3.3.3 

    inet.0: 13 destinations, 13 routes (13 active, 0 holddown, 0 hidden)
    + = Active Route, - = Last Active, * = Both

    3.3.3.3/32         *[OSPF/10] 00:01:09, metric 1
                        >  to 192.168.1.2 via lt-0/0/0.1

    The traceroute confirms that the destination IP 3.3.3.3 is one hop away

    labroot@R1:R1> traceroute 3.3.3.3    
    traceroute to 3.3.3.3 (3.3.3.3), 30 hops max, 40 byte packets
     1  3.3.3.3 (3.3.3.3)  3.495 ms  0.805 ms  0.567 ms

     

    Default behavior


    What happens if interface lt-0/0/0.1 goes down for any reason?

    The the traffic will look for the shortest intra-area path in area 100, that path will be R1-R4-R2-R3.

    The show route confirms this 

    labroot@R1:R1> show route 3.3.3.3 

    inet.0: 12 destinations, 12 routes (12 active, 0 holddown, 0 hidden)
    + = Active Route, - = Last Active, * = Both

    3.3.3.3/32         *[OSPF/10] 00:10:35, metric 3
                        >  to 192.168.1.10 via lt-0/0/0.5

    The traceroute confirms that the destination IP 3.3.3.3 is 3 hops away

    labroot@R1:R1> traceroute 3.3.3.3 
    traceroute to 3.3.3.3 (3.3.3.3), 30 hops max, 40 byte packets
     1  192.168.1.10 (192.168.1.10)  1.194 ms  0.577 ms  0.428 ms
     2  192.168.1.17 (192.168.1.17)  0.480 ms  0.597 ms  0.447 ms
     3  3.3.3.3 (3.3.3.3)  0.640 ms  0.752 ms  0.589 ms

    Default behavior without multiarea adjacency

    We know there is a shortest path through area 0 (R1-R2-R3) according to the diagram but that path is not in used because interface lt-0/0/0.3 does not participating in OSPF area 100 by default. Furthermore, the link through area 0 in an Inter-Area link from the perspective of area 100, so that connection is less preferred.

    Now what happens if we ask interface lt-0/0/0.3 to participate in OSPF area 100 as well?

    labroot@R1:R1# set protocols ospf area 0.0.0.100 interface lt-0/0/0.3 secondary             

    [edit]
    labroot@R1:R1# show | compare 
    [edit logical-systems R1 protocols ospf area 0.0.0.100]
          interface lt-0/0/0.1 { ... }
    +     interface lt-0/0/0.3 {
    +         secondary;
    +     }

    [edit]
    labroot@R1:R1# commit

    (The above configuration needs to be completed on R2 lt-0/0/0.4 interface as well) 

    If we configure lt-0/0/0.3 to participate in area 100 it will become part of the best path to reach the destination IP 3.3.3.3, this because the interface lt-0/0/0.3 will generate type 1 LSAs for area 100 as well.
    The traceroute confirms that the traffic is now going through interface lt-0/0/0.3

    labroot@R1:R1> show route 3.3.3.3 

    inet.0: 11 destinations, 11 routes (11 active, 0 holddown, 0 hidden)
    + = Active Route, - = Last Active, * = Both

    3.3.3.3/32         *[OSPF/10] 00:10:47, metric 2
                        >  to 192.168.1.6 via lt-0/0/0.3

    The traceroute confirm that the destination IP is two hops away 

    labroot@R1:R1> traceroute 3.3.3.3 
    traceroute to 3.3.3.3 (3.3.3.3), 30 hops max, 40 byte packets
     1  192.168.1.6 (192.168.1.6)  0.688 ms  0.624 ms  0.435 ms
     2  3.3.3.3 (3.3.3.3)  0.627 ms  0.638 ms  0.562 ms

    Default behavior without multiarea adjacency

    As you can see in this example, multiarea adjacency helped us to choose the shores path for both areas ( area 0 and area 100 ) and not just the shortest path for area 100. 

    We hope this information helps.

    Best Regards

    David Conejo