cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1873
Views
4
Helpful
6
Replies

IS-IS Single topology vs Multitopology

nwekechampion
Level 3
Level 3

Hi Guys,

Can anyone explain to me in simple terms what the difference is between IS-IS single topology and multitopology?

The documents available online are very confusing and conflicting, even cisco's.

I have tried to simulate both on my labs and cannot still understand the difference.

  1. What is the main difference between both?
  2. When would I need to use either ST or MT? (as in use-case)
  3.  Which one is used for
    dual-stack
    ?
  4. They say in Single topology, the
    lsdb
    is shared with
    ipv4 and ipv6
    address families. Can someone show me how to figure this out in
    show isis database
    ?
  5. They say, MT has separates LSDB for both v4 and v6 --> how do I verify this on router?

 

Thanks 

1 Accepted Solution

Accepted Solutions

I do not understand what is not clear, did you look at the document ?
- it looks the same (left two images)

pieterh_0-1686725119377.png

but there are differences

 

pieterh_2-1686725204204.png

 



View solution in original post

6 Replies 6

pieterh
VIP
VIP

did you find this  page ? ISIS for IPv6 ⋆ IpCisco
it is just what is says, with single topology

IPv4 and IPv6

share the same topology
with multi topology there can be a different topology created for

IPv4

and another for

IPv6


NB! the topologies can also look the same (same components, same links between components)
but still it is

multi-topology

because they are separately administrated

Hi Peter,

Could you explain this a bit further please "NB! the topologies can also look the same (same components, same links between components)
but still it is

multi-topology

because they are separately administrated"

Thanks for your response

Joseph W. Doherty
Hall of Fame
Hall of Fame

"When would I need to use either ST or MT? (as in use-case)"

The only use-cases that pop to mind for using a

multi-topology (for IPv4 and IPv6)

might be the two topologies are very disjoint, and SPF might have somewhat less work computing each, separately (along with possibly doing such computation faster).  Second, possibly you might cost the same link differently when using two topologies.  Another reason would be to very much control what's exchanged between the two topologies (which I've done using multiple OSPF processes on the same devices, but without the issues attended with routing between

IPv4 and iPv6

).

I do not understand what is not clear, did you look at the document ?
- it looks the same (left two images)

pieterh_0-1686725119377.png

but there are differences

 

pieterh_2-1686725204204.png

 



Thanks Guys.

I had to lab this one to fully get it.

Regards

Champ

Yes.. I did.
I am a bit slow sometimes..
perhaps I'd prefer to see it in action to see to fully get it.

I used the correct topology this time, and it all became very clear

Thanks again

Review Cisco Networking for a $25 gift card