router-id and multiple OSPF process

<!DOCTYPE html>




If a box has more than one OSPF process created, what is the rule to automatically assign the router-id? Do the processes choose the same router-id? Is it a problem to have more than one process with the same router-id inside a router??

Paulo
 


Comments

  • JoeMJoeM ✭✭✭

    Hello Paulo,

    I don't mean to sound harsh, but you are posting this question in a CCIE forum.  This is something that is very easy to test. Do yourself a favor, and separate yourself from the rest who ask without trying to learn the answer via config/verification tests.

    You will find that there will be a leap in your knowledge-base, when you learn to test (and VERIFY) the results.  Learning this state of thinking may be what saves you in a lab exam or in practice.

    This is why you will almost always see "verification" in the INE lab workbooks.  It is a very good practice to make a habit.

    CCIE prep advice aside, it is a good question, and I needed to test my answer first. ;-)

     

    Answer:   I was not really sure about the answer, because I have made a habit of using manual router-id's in all of the protocols when possible.   Manually setting the router-id gives me peace-of-mind and they are much easier to read with show command outputs.

    So, I fired up a router and tested it.  Very simple test.

    1. created two loopbacks (for fallback to the 2nd highest loop, my guess)
    2. created a physical address on f0/0 just for the sake of testing.

    TEST Results:

    R1(config-if)#router ospf 1
    R1(config-router)#do sh ip proto | s ospf
    Routing Protocol is "ospf 1"
      Outgoing update filter list for all interfaces is not set
      Incoming update filter list for all interfaces is not set
      Router ID 11.11.11.11
      Number of areas in this router is 0. 0 normal 0 stub 0 nssa
      Maximum path: 4
      Routing for Networks:
      Routing Information Sources:
        Gateway         Distance      Last Update
      Distance: (default is 110)


    R1(config-router)#router ospf 2
    R1(config-router)#do sh ip proto | s ospf
    Routing Protocol is "ospf 1"
      Outgoing update filter list for all interfaces is not set
      Incoming update filter list for all interfaces is not set
      Router ID 11.11.11.11
      Number of areas in this router is 0. 0 normal 0 stub 0 nssa
      Maximum path: 4
      Routing for Networks:
      Routing Information Sources:
        Gateway         Distance      Last Update
      Distance: (default is 110)
    Routing Protocol is "ospf 2"
      Outgoing update filter list for all interfaces is not set
      Incoming update filter list for all interfaces is not set
      Router ID 1.1.1.1
      Number of areas in this router is 0. 0 normal 0 stub 0 nssa
      Maximum path: 4
      Routing for Networks:
      Routing Information Sources:
        Gateway         Distance      Last Update
      Distance: (default is 110)


    R1#sh ip int br
    Interface                  IP-Address      OK? Method Status         Protocol
    FastEthernet0/0            12.12.12.1      YES manual up               up
    <snip>
    Loopback0                  1.1.1.1         YES manual up               up
    Loopback1                  11.11.11.11     YES manual up               up

  • JoeMJoeM ✭✭✭

    Test 2 -- Trying to match router-ids

     

    R1(config)#router ospf 2
    R1(config-router)#router-id 11.11.11.11
    % OSPF: router-id 11.11.11.11 in use by ospf process 1

    NOTE:   Router-ID does not change. Not possible to have matching RID's


    R1(config-router)#do sh ip proto | s ospf 2
    Routing Protocol is "ospf 2"
      Outgoing update filter list for all interfaces is not set
      Incoming update filter list for all interfaces is not set
      Router ID 1.1.1.1
      Number of areas in this router is 0. 0 normal 0 stub 0 nssa
      Maximum path: 4
      Routing for Networks:
      Routing Information Sources:
        Gateway         Distance      Last Update
      Distance: (default is 110)

  • It depends on the order in which the OSPF processes are created. Each time a new process is started, it goes through the same rules in allocating an router-id. Select the highest loopback avaialble. If there is no loopback available, select the highest IP address on physical links available. Note that this is on a per VRF basis. 

    Say you have a router with the following interfaces:

    Loopback0: 1.1.1.1/32

    Gig1.100:   10.1.1.1/24

    Gig1.200:   10.1.2.1/24

     

    If you create 4 OSPF processes on this router, the first 3 will be able to dynamically select a RID, but the fourth one will not, as all 3 available IPs would have been "consumed"

     

    router ospf 1

     ! this will take 1.1.1.1 as is the only loopback. Both physical interface are left in the "pool"

    router ospf 2

     ! will take 10.1.2.1 as its the highest non loopback. Only 1 physical interface is left in the "pool".

    router ospf 3

     ! will take 10.1.1.1 as its the highest non loopback left. All avaliable interfaces are consumed.

    router ospf 4

     ! will not be able to allocate a RID.

     

    If you reload the router, it then depends on the order in which the processes are loaded.

    Try it out =) 

     

     

     

  • <!DOCTYPE html>




    Joe,

     

    Thank you for being so polite.

     

    This is the kind of answer I would expect from someone  at CCIE level. At least you are good at this technical stuff. However I can not say the same about personal relationship. On personal stuff you should be at pre-school level.

     

    By the way, I am in transit and I will only be back to my lab by the end of the week.

     

    Thank you again.

     

    Paulo 

     

     


     

     

    Em Dom 28 jun. 2015, às 16:33, JoeM escreveu:

    Hello Paulo,

    I don't mean to sound harsh, but you are posting this question in a CCIE forum.  This is something that is very easy to test. Do yourself a favor, and separate yourself from the rest who ask without trying to learn the answer via config/verification tests.

    You will find that there will be a leap in your knowledge-base, when you learn to test (and VERIFY) the results.  Learning this state of thinking may be what saves you in a lab exam or in practice.

    This is why you will almost always see "verification" in the INE lab workbooks.  It is a very good practice to make a habit.

    CCIE prep advice aside, it is a good question, and I needed to test my answer first. ;-)


    Answer:   I was not really sure about the answer, because I have made a habit of using manual router-id's in all of the protocols when possible.   Manually setting the router-id gives me peace-of-mind and they are much easier to read with show command outputs.

    So, I fired up a router and tested it.  Very simple test.

    1. created two loopbacks (thinking there may be fallback to the 2nd highest loop, my guess)
    2. created a physical address on f0/0 just for the sake of testing.

    TEST Results:

    R1(config-if)#router ospf 1

    R1(config-router)#do sh ip proto | s ospf

    Routing Protocol is "ospf 1"
      Outgoing update filter list for all interfaces is not set
      Incoming update filter list for all interfaces is not set
    Router ID 11.11.11.11
      Number of areas in this router is 0. 0 normal 0 stub 0 nssa
      Maximum path: 4
      Routing for Networks:
      Routing Information Sources:
        Gateway         Distance      Last Update
      Distance: (default is 110)

     

    R1(config-router)#router ospf 2

    R1(config-router)#do sh ip proto | s ospf

    Routing Protocol is "ospf 1"
      Outgoing update filter list for all interfaces is not set
      Incoming update filter list for all interfaces is not set
    Router ID 11.11.11.11
      Number of areas in this router is 0. 0 normal 0 stub 0 nssa
      Maximum path: 4
      Routing for Networks:
      Routing Information Sources:
        Gateway         Distance      Last Update
      Distance: (default is 110)
    Routing Protocol is "ospf 2"
      Outgoing update filter list for all interfaces is not set
      Incoming update filter list for all interfaces is not set
    Router ID 1.1.1.1
      Number of areas in this router is 0. 0 normal 0 stub 0 nssa
      Maximum path: 4
      Routing for Networks:
      Routing Information Sources:
        Gateway         Distance      Last Update
      Distance: (default is 110)

     

    R1#sh ip int br

    Interface                  IP-Address      OK? Method Status                Protocol
    FastEthernet0/0            12.12.12.1      YES manual up                    up
    <snip>
    Loopback0                  1.1.1.1         YES manual up                    up
    Loopback1                  11.11.11.11     YES manual up                    up

     

     

     



    INE - The Industry Leader in CCIE Preparation


     


    Subscription information may be found at:


     


  • JoeMJoeM ✭✭✭


    Joe,


    Thank you for being so polite.


    This is the kind of answer I would expect from someone  at CCIE level.....


    You are very welcome. I spent a little bit of my time, showing my process of verification.  I am happy that it helped you. Hopefully others will also find it helpful.  There is a process to verifying each technology.  This is part of the learning process.

     

    The study tip that I shared with you should not offend you at all -- if you are truly serious about going to the next level.

    "Do you really want to be an engineer?  Or do you just want to pass the exam?"   
                                                               - TS video series by BrianDennis 

     

     

  • Martinl, router-id is just a 32-bit notation. Does not have to be an int IP.

    Sent from my iPhone

    On 1 Jul 2015, at 6:51 pm, Martinl <[email protected]> wrote:

    Important change (recent) is that interface does not have to be in UP/UP state to be considered for router -id election.  this is despite of ccna-level books/courses.  Hee is what I wrote previously:

    It is sufficient for the interface to be in the down/down state to be considered by OSPF as a prospective interface for RID selection. 

    Note: there is a difference between down and administratively down.  Most of other books (CCNA, CCNP) still stick with up/up. So for any written exam purpose, I would too.

    This change may be due to Virtual IOS used in networking and the lab.
    The question is how can I get down/down state? Can vIOS or IOU/IOL
    detect link problems at physical layer like real IOS can? After all,
    There are no real wires between virtual devices!




    INE - The Industry Leader in CCIE Preparation

    http://www.INE.com



    Subscription information may be found at:

    http://www.ieoc.com/forums/ForumSubscriptions.aspx
Sign In or Register to comment.