cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1834
Views
0
Helpful
10
Replies

is-is fast re-route topology

junaide45
Level 1
Level 1

what is the concept behind fast reroute in IS-IS

as it is not working for some of my devices in the network

Can someone explain what parameters of is-is have to met in order to make complete working fast re-route topology??

3 Accepted Solutions

Accepted Solutions

Ok in your topology IPFRR will partially work

Example let's say P3 is rotuing via P1 to reach PE2

In the case of P1-PE1 link faulre P1 can not use p3 as a loop free frr because igp still not converged and p3 still using p1 to reach pe2

What you can do is multi home PE3 and PE1 to both P1 and P3 to get around this limitation

Other option is if your network is mpls enabled you may read about remote LFA with LDP which can help you here

Hope this help

View solution in original post

ok as per my example above

if P1 pre compute a back up path via P3 to reach PE2 prefixes and in a normal state P3 uses P1 to reach PE2 prefixes

then at the moment of failure if P1 uses P3 before ISIS converge P3 still uses P1 to reach these prefixes and this will form a loop

however if P1 uses P3 as a back path to reach P4-PE4 subnet then P1 can route via P3 as P3 dose not use P1 to reach this subbent and will not form a loop

hope this help

View solution in original post

10 Replies 10

Marwan ALshawi
VIP Alumni
VIP Alumni

I think your referring here to lfa with isis

Can you share your topology and how are doing it

LFA/ipfrr works with 70 to 90% of topologies that's why we need to look at your topology and the path/prefix to be protected

thanks for reply

topology is something like this

      P1 ------ PE1 ------- P2 ---- PE2

      /                             |  

     /                              |   

   P3----PE3----PE4-------P4 ----- PE5

isis level-2 is configured

with fast reroute enabled in the whole ring

Ok in your topology IPFRR will partially work

Example let's say P3 is rotuing via P1 to reach PE2

In the case of P1-PE1 link faulre P1 can not use p3 as a loop free frr because igp still not converged and p3 still using p1 to reach pe2

What you can do is multi home PE3 and PE1 to both P1 and P3 to get around this limitation

Other option is if your network is mpls enabled you may read about remote LFA with LDP which can help you here

Hope this help

This is exactly the same problem i m facing

whenever P1--PE1 or PE1--P2 link goes down we have service impact

The multi-home solution is not practical right now

yes our network is MPLS enabled, but unfortunately i am not aware of remote LFA

how can I know more about this solution

Thanks again

again thank you ... i will read and try to look for solution

but i have some questions ...

why igp will not converge? is there any issue in is-is or is it the normal behaviour?

and why you said for this topology FRR will work partially; frankly how we can know this?

ok as per my example above

if P1 pre compute a back up path via P3 to reach PE2 prefixes and in a normal state P3 uses P1 to reach PE2 prefixes

then at the moment of failure if P1 uses P3 before ISIS converge P3 still uses P1 to reach these prefixes and this will form a loop

however if P1 uses P3 as a back path to reach P4-PE4 subnet then P1 can route via P3 as P3 dose not use P1 to reach this subbent and will not form a loop

hope this help

thank you very much yet again

It was really helpful ... will try to implement this in our network

as Marwan said 70 to 80% can be covered with LFA, i think redesign some metrice plus using Remote LFA (P1 to PE4)can raise your percentage up to ~90%.

however achiving more than 98% can be done only with the help of the TE tunnels.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: