TI-LFA FRR
Topology-Independent Loop-free Alternate Fast Re-route (TI-LFA FRR) protects links and nodes on segment routing tunnels. If a link or node fails, TE FRR rapidly switches traffic to a backup path, minimizing traffic loss.
Related Concepts
Concept |
Definition |
|---|---|
P space |
The P space contains a set of nodes reachable to the root node on links, not the protected link, along the SPF tree originating from the protected link's source node functioning as the root node. |
Extended P space |
The extended P space contains a set of nodes reachable to the root nodes on links, not the protected link, along the SPF trees originating from neighbors of protected link's source node functioning as the root nodes. |
Q space |
The Q space contains a set of nodes reachable to the root node on links, not the protected link, along the reverse SPF tree originating from the protected link's destination node functioning as the root node. |
PQ node |
A PQ node resides in both the extended P space and Q space. The PQ node functions as the destination node of a protected tunnel. |
LFA |
The loop-free alternate (LFA) algorithm computes a standby link. A root node that can provide a standby link runs SPF to compute the shortest path to a destination node. The root node then computes a loop-free standby link with the smallest cost. |
RLFA |
Remote LFA (RLFA) computes a PQ node based on a protected path and establishes a tunnel between the source and PQ nodes to provide next hop protection. If the protected link fails, traffic automatically switches to the backup path, which improves network reliability. |
TI-LFA |
In some LFA FRR or RLFA scenarios, the P space and Q space do not share nodes or have directly connected neighbors. Consequently, no backup path can be calculated, which does not meet reliability requirements. In this situation, TI-LFA can be used. The TI-LFA algorithm computes the P space and Q space based on a protected path, a shortest path tree (also called a post-convergence tree), and a repair list. The algorithm establishes a segment routing tunnel between the source node and PQ node to provide backup next hop protection. If the protected link fails, traffic automatically switches to the backup path, which improves network reliability. |
Background
Conventional LFA requires that at least one neighbor be a loop-free next hop to a destination. RLFA requires that there be at least one node that connects to the source and destination nodes along links without passing through any faulty node. However, TI-LFA uses an explicit path to represent a backup path, which poses no requirements on topology constraints and provides more reliable FRR.
Benefits
- Meets basic requirements of IP FRR rapid convergence.
- Theoretically supports all protection scenarios.
- Uses an algorithm with moderate complexity.
- Selects a backup path over a converged route and has no intermediate state, compared with the other FRR techniques.
TI-LFA FRR Fundamentals
In Figure 5-22, PE1 is a source node, P1 is a faulty node, and PE3 is a destination node. Link costs are marked.
TI-LFA traffic protection involves link and node protection.
Link protection: protects traffic passing through a specific link.
Node protection: protects traffic passing through a specific node. Node protection takes precedence over link protection.
In the following example, the process of node protection is as follows. In Figure 5-22, traffic travels along a path PE1 -> P1 -> P5 -> PE3. If P1 fails, TI-LFA computes the P space, Q space, SPF tree (also called the post-convergence tree), backup outbound interface, and repair list. Traffic is forwarded along the backup path to the destination PE3, which implements rapid protection to prevent traffic loss.
Computes the extended P space. It contains a set of nodes reachable to the root nodes on links, not the protected link, along the SPF trees originating from neighbors of protected link's source node functioning as the root nodes.
Computes the Q space. It contains a set of nodes reachable to the root node on links, not the protected link, along the reverse SPF tree originating from the protected link's destination node functioning as the root node.
Computes the post-convergence SPF tree. It excludes the primary next hop.
- Computes a backup outbound interface and a repair list.
Backup outbound interface: In some scenarios, the P space and Q space do not share nodes or have directly connected neighbors. In this case, the backup outbound interface is the next hop outbound interface after convergence.
Repair list: It is a constrained path that directs traffic to the Q node. The repair list consists of a P node label and adjacency labels of the P-to-Q path. In Figure 5-22, the repair list consists of P3's node label 100, and P3-to-P4 adjacency label 9304.
- A node SID advertised by the repair node is preferentially selected.
- A prefix SID advertised by the repair node is preferentially selected. The smaller the SID, the higher the priority.
- A node that does not support segment routing or a node that does not advertise a prefix or node SID cannot function as a repair node.
TI-LFA FRR Backup Path Forwarding
After a TI-LFA backup path is computed, if the primary path fails, traffic switches to the backup path, preventing packet loss.
In Figure 5-23, Device F is a P node, and Device H is a Q node. The primary next-hop B fails, which triggers FRR switching. Traffic switches to the backup path. Table 5-6 shows the detailed process.
Device |
Process |
|---|---|
Device A |
Device A encapsulates a label stack to a packet based on the repair list from outer to inner: Node label of the P node (Device F) = Start label in next-hop Device D's SRGB + Label offset of the P node = 720; P-to-Q adjacency labels of 130 and 240; Destination node label = Start label of the Q node's SRGB + Label offset of the destination node (Device C) = 310 |
Device D |
After receiving the packet, Device D searches the label forwarding table based on the outgoing label and finds a matching entry with the outgoing label of 120 and next hop being Device F. Device D swaps the outgoing label for 120 and forwards the packet to Device F. |
Device F |
After receiving the packet, Device F searches the label forwarding table based on the outgoing label. Device F is the egress node so that it removes the label. It finds a matching entry with a routed path label of 130, the outgoing label as empty, and the next hop being Device G. Device F removes label 130 and forwards the packet to Device G. |
Device G |
After receiving the packet, Device G searches the label forwarding table based on the outgoing label, removes label 240, and forwards the packet to Device H. |
Device H |
After receiving the packet, Device H searches the label forwarding table based on the outgoing label and finds a matching entry with the outgoing label of 510 and the next hop being Device E. Device H swaps the outer label for 510 and forwards the packet to Device E. Device E forwards the packet to Device C. The packet travels along the shortest path. |
TI-LFA FRR Protection Usage Scenarios
TI-LFA FRR Protection |
Description |
Deployment |
|---|---|---|
TI-LFA FRR protects IP forwarding. |
Traffic is transmitted over an IP routed primary path, and a TI-LFA FRR backup path is computed. |
|
TI-LFA FRR protects traffic on a segment routing tunnel. |
Traffic is transmitted over a primary segment routing tunnel, and a TI-LFA FRR backup path is computed. |
|
Anti-Micro-Loop Switchover
In Figure 5-24, if Device B fails, traffic is switched to a TI-LFA FRR backup path. After Device A completes route convergence, traffic is switched from the TI-LFA FRR backup path to a converged path. If Devices D and F do not complete route convergence, they transmit traffic over the path established before convergence is performed. As a result, a loop emerges between Devices A and F and is broken after route convergence finishes on Devices D and F.
To prevent the loop-induced problem, the implementation is modified. After Device B fails, traffic is switched to the TI-LFA backup path. Device A delays convergence. After Devices F and D complete route convergence, Device A starts route convergence. After route convergence is complete, traffic is switched from the TI-LFA backup path to the converged path.
- The interface directly connected to the local interface fails, or local BFD goes Down.
- No network topology change occurs during the delay.
- A backup next hop for a route is available.
- The primary next hop of the route is the faulty interface.
- The primary and backup next hops are different after the route convergence is complete.
- During the delay of the multiple nodes advertise the Same Route, the route source change event occurs, and the delay stops.
Anti-Micro-Loop Switchback
- Data is being transmitted along the backup path before the link between Device B and Device C recovers.
- After the link between Device B and Device C recovers, if Device A converges earlier than Device B, Device A forwards traffic to Device B that does not finish convergence. Upon receipt of the traffic, Device B forwards traffic along the original path to Device A, causing a loop.
- To prevent a micro loop, after a traffic switchback is performed on Device A, configure an explicit path to forward packets. Device A adds E2E path information (for example, a B-to-C adjacency label) to data packets. Upon receipt of the data packets, Device B forwards packets to Device C based on the carried path information.
After Device B finishes convergence, Device A deletes explicit path information from data packets so that the data packets can be forwarded to Device C using normal SR.