Example: Configuring EVPN-MPLS Active-Standby Multihoming with ESI
This example covers the steps necessary to configure Ethernet VPN-MPLS (EVPN-MPLS) active-standby multihoming using either Ethernet Segment Identifier (ESI) per physical interface (IFD) with Link Aggregation Control Protocol (LACP), or ESI per logical interface (IFL) with Connectivity Fault Management (CFM).
Understanding LACP and CFM in EVPN-MPLS Active-Standby Multihomed Environments
EVPN-MPLS active-standby multihoming with ESI leverages advanced networking techniques to provide robust redundancy and high availability. By configuring active-standby roles, you ensure that traffic is consistently forwarded through a single active link while multiple standby links remain ready to take over in case of a failure. This setup is critical for maintaining uninterrupted service and ensuring network resilience. The feature supports both LACP-based and CFM-based solutions, each tailored to different interface types—aggregated ethernet (AE) and non-AE interfaces, respectively. We support active-standby multihoming in EVPN fabrics only with MPLS.
For the LACP-based solution, you configure the system to use LACP out-of-sync mechanisms. This approach communicates the non-designated forwarder (DF) state to the customer edge (CE) device, ensuring it does not use non-DF links for traffic forwarding. The configuration involves setting the DF election granularity and enabling LACP out-of-sync for non-DF states via specific CLI commands. This method provides effective load balancing and redundancy management, making it ideal for environments using AE interfaces. The LACP-based solution is used for ESI per IFD in EVPN deployments. ESI per IFL is not supported with LACP in active-standby EVPN-MPLS deployments, since LACP will bring down the affected AE and all of its member links.
The CFM-based solution manages interface status transitions on non-AE interfaces. By configuring CFM, you enable the CE to detect DF/non-DF transitions through interface-status-tlv messages, which adjust the interface status accordingly. This method ensures robust fault management and status monitoring, crucial for scenarios where non-AE interfaces are used. Specific CLI commands are used to set up the CFM maintenance domain (MD), maintenance association (MA), and maintenance end point (MEP) configurations, providing a detailed framework for managing high availability and redundancy. The CFM-based solution is used for ESI per IFL in EVPN deployments. CFM is not supported if the CE uses an AE for the multihomed interface.
Example Prerequisites
You can refer to the following table for platform and release requirements specific to this example. You an can use the same platform for all devices, or a mix of platforms.
Hardware requirements |
|
Software requirements |
Junos OS Evolved Release 24.2R1 and later is required for the ACX and PTX platforms. |
Configure ESI per IFD with LACP
Configure ESI per IFL with CFM
Verification
Use the following commands to check various elements in either the LACP or CFM based deployments.
Command | Verification Task |
---|---|
show evpn instance extensive | On a PE device, check the status of the DF and non-DF nodes. |
show lacp interfaces | On a CE device, check the status of the DF and non-DF interfaces. |
show interfaces <interface-name> | On a CE device, check the status of the DF and non-DF nodes in CFM. |
show oam ethernet connectivity-fault-management interfaces detail | On a CE device, check the status of CFM. |
Release Information
(PTX10001-36MR, PTX10004, PTX10008, and PTX10016)—Support for ESI per IFD with LACP and ESI per IFL with CFM added in Junos OS Evolved Release 24.2R1.
(ACX7332 and ACX7024X)—Support for ESI per IFD with LACP added in Junos OS Evolved Release 24.2R1.
(ACX7100-32C, ACX7100-48L, ACX7348, ACX7509, and ACX7024)—Support for ESI per IFD with LACP added in Junos OS Evolved Release 23.4R1.