Topology Independent LFA (TI-LFA) for Segment Routing

Problem with LFA : The most common citied topology is the ring topology when it comes to illustrating LFA coverage.

If a failure happens at the link between R1-R2 then it cannot sent it to R3 as a FRR backup for R1-R2 as R3 will send the traffic back to R2. Basically R3 doesn’t meet inequality

1 (3 > 1 + 2). R3 best path is through the failed link. So plain LFA won’t work and there were the remote LFA comes into the picture.

In case of rLFA we need to search for the PQ node i.e node with overlap between Extended P-Space and Q-Space.
Extended P-Space : The extended P space of the protecting router with respect to the protected link is the union of the P-Space of the neighbors in that set of neighbors with respect to the protected link. Extended P-Space contains the routers that are R2’s direct neighbor. R3 can reach without using the R2’s, R1 link which R4 and R5 node.

Q-Space : Q-Space of a router with respect to protected link is the set of routers from which that specific router that can reached without any path, transiting that protected link. Q space contains the router that normally reach R6(Destination) without using the R2 (Source) -R1 link which is R1, R5 and R4 nodes.  

PQ Node : A router that is both extended P Space and Q Space is a PQ Node. Any router that is PQ Node is a rLFA candidate. The candidate router to whom R2 sends the packet, it will forwards the packet to the destination without traversing through R2(Src) – R1 link. In this case R4 and R5 are PQ-Node and are considered remoted LFA candidates for R2 (Source)  

In case of rLFA, R2 (Source) will use nearest PQ Node as its backup path in the event of R1-R2 link failure. The way R2 will achieve by stacking two LDP labels. Outer LDP label is the label to reach R4 and inner LDP label Y to reach R6(Dest) from R4. With the help of trageted LDP session, R2 (Source) can know what LDP label (Y) is used by R4 for sending traffic towards R6(Dest) Which brings up the point that in order for the PLR to know what label is being used by PQ Node for Dest (D) it has established a Targeted LDP session with PQ node to get the FEC to label mappings.  

Now if cost between R5 and R6 increases then, we don’t have any nodes in PQ Space which means rLFA cannot provide coverage in this situation.  

In case of TI-LFA’s :  

  • No Need to have Targeted LDP sessions  :

So the first improvement is SR brings no need of LDP sessions. In SR will stack two labels first is the Node-SID label of R6 (Dest) and top label as Node SID of R4.  

  • Double Segment Protection : 

As we saw earlier that in the case where there is no overlap between P and Q space then rLFA doesn’t provide coverage. TI-LFA can provide coverage in these kind of situations by steering the packets from P space to Q Space node.  

Assume that R5, R6 link SID is 24065. Then R2 sends three label stack with first two top labels are used to steer the packets from P to Q space nodes and third labels the destination prefix-sid label.  

  • Path optimality :

Another thing which TI-LFA brings is the path optimality from PLR perspective . TI-LFA always prefers the post convergence path from a PLR point of view.  

R8 and R7 link failed.  

In case of RSVP TE FRR (Link protection and facility backup) traffic will circle all the way back to the next-hop (R7), obviously its sub-optimal, but is a temporarily conditional till the headend R1 realizes the failure.  

In case of LFA, The path goes via R8 to R3, which is a higher metric. Considering that metric represents bandwidth of a link in a network, in the event of failure, traffic will be going through low bandwidth links which could cause congestion and packet drops on the low capacity link. In this case it would be better if PLR, R8 can use the post convergence path from its perspective. I.e. via R2, R3 and R4 as  a backup path, but in the case of LFA, R8 cant use R2 as  a backup path as R2 best via R8 (Loop will occur). 

In case of TI-LFA it is possible for the PLR, R8 use the post convergence path which is via R2, R3 and R4 to R5.  

R8 send the packet with two stack labels with the top label to get the R3 and inner prefix-sid for the destination.  

Path to R3 from R8 perspective is Via R2.  

When R2 sees the top label for R3, its pop the label and send it towards R3 and from their it will follow the destination.


Categories: MPLS, Nokia, Routing

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s