Tags: ,


Notable Replies

  1. Ok, Hi everyone just joined the site.

    I want to make sure that my brain understands this. It’s like you slice and dice and hopefully everything comes out right. So say I have this already as my prefix list:

    10.0.12.0/24 le 32

    Now say I want to include another network. So I want the following two networks in one prefix-list.

    10.0.12.0/24 and 10.0.13.0/24, I want these covered by one prefix list.

    So I figure that this will fit within the scope of these two networks.

    2 will be size of subnets:

    0, 2, 4, 6, 8, 10
    12 and 13 — this fits just right.
    14, 16, 18 etc…

    So I delete the old prefix and add this:
    10.0.12.0/23 le 32

    I’m not worried about the “le 32” as that basically means I’m accepting all addresses in the 10.0.12.x and 10.0.13.0 scope.
    So do I understand this correctly?

    Thank you!

  2. I was so confused by this at first. I know its simple but its also a brain teaser for some reason.
    first I pulled up my boson subnet tool its free tool by the way on the Boson.com (just need to create an account) website. At first I was thinking that first two bits mean the first two spots _ _ thinking it could be anything from 0 to 192 that it did not matter if it was a zero or a one. However after putting into the subnet calculator it helped me to see.

    The next really helpful thing for me was when I went ahead and enabled the command with distribute-list prefix CLASSB in all of a sudden all my 10.x.x.x networks disappeared which shot my first theory to crap along with supporting the boson subnet calculator.

    So seeing it in play in a lab really made sense. I think the confusion goes back to the rule on how the classes are setup. When we was learning sub-netting classes that you read and say oh ok but as time goes on you just get use to seeing the numbers themselves and that they are a certain class.

    I am betting everyone that had a problem with this does not use the Class A, B, C rule anymore but instead over time have subconsciously just memorized the 1-127 is A, 128-191 is B, 192-223 is C and does not really think of the rule about class A the first bit always being 0, and class b the first two bits being set to 10, and class C having its first three bits set to 110… (hoping me explaining this in writing will actually help me remember it! lol)

    So its like trying to do a math problem and finally that silly rule in math never used much is key to how the expression functions.

    anyway below is some more information reinforcing Rene info.

    Capture

    posted the rule below in greater detail that Rene implicitly mentioned briefly in his post I am one of those type that can sometimes be slow seeing something the way it should be seen until I experience it for myself.
    https://www.tutorialspoint.com/ipv4/ipv4_address_classes.htm

    Capture

  3. Rene,

    I currently have set up R2 connected to R4.
    R4 is advertising in EIGRP:
    192.168.0.1/24
    192.168.1.1/30 (255.255.255.252)
    192.168.2.1/29 (255.255.255.248
    192.168.3.1/28 (255.255.255.240)

    I’m simply just practicing with prefix-list and wanted to filter out the /30 /29 /28 routes, and only advertise the /24

    On R4 I have done:

    R4(config#) ip prefix-list test deny 192.168.0.0/16 ge 28 le 30
    R4(config#) ip prefix-list test permit 0.0.0.0/0 le 32
    
    R4(config-router#) distribute-list test out serial0/0/0
    

    However, R2 is still showing all of the above mentioned routes in its routing table. I also tried filtering the same routes IN on R2 but to no effect. Where am I going wrong?

    Thank you so much!

  4. I have been going crazy trying to figure this out the past 24 hours. Finally figured out that I was leaving out the keyword ‘prefix’ in my “distribute-list” command syntax. Basically the distribute-list was looking for an ACL (that never existed) because I didn’t specify ‘prefix’ in the command. A little more tricky since leaving out ‘prefix’ is an acceptable command. It’s working as it should now :grinning:

Continue the discussion forum.networklessons.com

54 more replies!

Participants