Difference Between RSVP-TE Hot Standby and Fast Reroute

Hello,

I am hoping someone can help me clarify the difference between RSVP-TE's Hot Standby LSP and the Fast Reroute function? They appear to be the same to me. Does one have a recovery time quicker than the other? In both cases, the secondary LSP has already been created (unlike a backup LSP in which the tunnel is not built until the primary LSP has failed). Can someone help me understand the advantages of using one over the other?

Find more posts tagged with

Comments

networker050184

A lot of it comes from understanding how LSPs are singled and set up.

Host standby LSP will be a backup tunnel built from the ingress router that is standby for failure scenarios so another LSP does not need to be signaled.

Fast Reroute is a function for intermediary routers to route traffic around a failure before a new LSP needs to be signaled by the ingress router.

alemoo

Thanks for the reply, networker050184.

So it would appear that they are both very similar, with the main difference being that for Fast Reroute, the router that detects the failure, will proactively reroute traffic.

I can't upload images to show an example.

Let's say you have A to B to C in a straight line.
Underneath B, you have a link going downwards, towards D.
D connects towards the right to E, and E goes back up to C

1) The link from B to C fails.
2) B immediately reroutes traffic via B to D to E to C, and simultaneously informs A of the failure.
3) A receives the fail notification, and builds a new tunnel: A to B to D to E to C
4) After this tunnel is built, all traffic is routed through this new path
5) FRR tunnel B to D to E to C is torn down? What about the original tunnel A to B to C?

Thanks!

powmia

alemoo wrote: »

Thanks for the reply, networker050184.

So it would appear that they are both very similar, with the main difference being that for Fast Reroute, the router that detects the failure, will proactively reroute traffic.

Thanks!

Be careful with your wording here. In both cases, the router that detects the failure will use a pro-actively established tunnel to reroute traffic onto. The difference is in which router that failover occurs.

For one, you're referring to path protection, not 'hot-standby.' In path protection, it's the headend router that has an alternate LSP setup to reroute traffic in the event that the LSP is broken (end-to-end, so anywhere along that LSP). With fast-reroute, what is typically the core of your network has every router pre-signalling an alternate nhop or nnhop for LSPs that have been signalled through that router and will reroute traffic around a specific point of failure. It may be hard to visualize the difference (especially since the end result could actually be the same path for either methods of protection), so you'll have to do so research here if you need further clarification.

alemoo

Ah you are correct powmia - I am getting the feature names confused as I am with Huawei. Thanks again, to both of you.