Another loopback question....

[louis0001]

I understand the advantages of loopbacks but have never implemented them yet. As we want to use OSPF, the loopbacks become more important to us.

I read in an article that:

It is very common to assign all the IP addresses used for loopback interfaces from one address block. For example, an ISP with around 200 routers in a network might assign a /24 network (253 usable addresses) for addressing the loopback interface on each router. If this is done, all dependent systems can be configured to permit this address range to access the particular function concerned, whether it is security, unnumbered WAN links, or the iBGP mesh. 

Now, assigning all the loopbacks from a /24 would keep it nice and tidy. However I can't get my head around doing this on routers on a totally different subnet? 

eg
ROUTER A = 10.1.1.1/24
ROUTER B = 10.1.2.1/24
ROUTER C = 10.1.3.1/24

Is it ok to do this?

[chrihussey]

I'm having trouble understanding your question, could you clarify?

Aside from that, a loopback is usually given a /32 mask, assigning a /24 to each loopback would be a huge waste of address space. There may be instances where it could be used, but not in this case.

Thanks

[Diana Karolina Rojas]

Hello Louis!

First of all you have to know this: "OSPF treats Loopback interfaces as STUB NETWORKS and advertise them as HOST ROUTES (with mask /32) regardless of their configured/native mask." So the idea to have a /24 as native mask is to better management, that way you can know that all the OSPF loopbacks are the ones inside the network X.X.X.X/YY and also you don't have to gaste a subnet by each router that is unnecesary and impractical (because as Chri said is a gaste of addresing)... but can do that if you want.

Please do not forget to rate usefull post.

Best Regards,

[Richard Burts]

I understand the question from the original poster differently from Chris or Diana. The original poster refers to an article suggesting that some providers use a /24 subnet to provision management addresses on their equipment. The original poster focuses on the idea of a management address with a /24 mask as seen in this part of the post

ROUTER A = 10.1.1.1/24
ROUTER B = 10.1.2.1/24
ROUTER C = 10.1.3.1/24

what the original poster misses is that the provider was using a single subnet and all of the devices would have management addresses from that subnet. So for his environment it would look more like

ROUTER A = 10.1.1.1/24
ROUTER B = 10.1.1.2/24
ROUTER C = 10.1.1.3/24

I have worked with customers who use this practice and it works well, especially for relatively large networks. It might not have as much benefit for small or medium networks. It makes it easy to distinguish management traffic from other data traffic and can be especially helpful if you have security policies or QOS requirements that would want to treat management traffic differently.

HTH

Rick

[Heath Deschenes]

I believe the recommendation was just to use all of the loopbacks from a /24 network space, such as 10.1.1.0/24 In that case each loopback would be an individual /32.

It would look more like this:

ROUTER A = 10.1.1.1/32
ROUTER B = 10.1.1.2/32
ROUTER C = 10.1.1.3/32