# EVPN Multihoming **URL:** \ **Category:** MPLS\ **Created:** [November 15, 2026, 10:58am UTC](https://forum.networklessons.com/t/evpn-multihoming/58914 "2026-11-15T10:58:01Z")\ **Posts on this page:** 1\ **Page:** 1
**Author:** ![davidilles](https://cdn-forum.networklessons.com/user_avatar/forum.networklessons.com/davidilles/32/9686_2.png) [@davidilles](https://forum.networklessons.com/u/davidilles)\ **Post date:** [November 15, 2026, 10:58am UTC](https://forum.networklessons.com/t/evpn-multihoming/58914/1 "2026-11-15T10:58:01Z")
Hello, everyone. I have this simple EVPN VPWS Multihoming setup: ![obrázok](https://cdn-forum.networklessons.com/uploads/default/original/2X/b/b6e75c85b370995a991667bb669e7dabdcaa083c.png) The great part about this is that many of the routes that you see in traditional LAN multihoming such as T2, T1 per ES or T4 are irrelevant/ignored in this scenario. My question is. If you take R1-XR as an example, it will receive routes from R4-XR and R5-XR and notice that the ES matches for both of them. R1-XR now knows that they connect to the same ethernet segment - 45.x.x.x. R1-XR can now load-balance traffic when sending it towards the other end of the VPWS. This is the confusing part for me: ![obrázok](https://cdn-forum.networklessons.com/uploads/default/original/2X/2/29c2b37433bd682354bb7e654d948f9cc6007773.png) ![obrázok](https://cdn-forum.networklessons.com/uploads/default/original/2X/d/d250701b3082ad1651cc466fc6ccb8436d0ba8df.png) R1-XR will load-balance to both R4-XR and R5-XR, yet the entry above is a little confusing. Why did it generate a local label of 24010? Also, doesn’t the MPLS entry above state "if a packet is received with an MPLS label of 24010, it’s swapped with 24008 (since R4 and R5 both generated the same label value) and sent over EVI 100? I don’t understand why did it generate a local label of 24010. When SW1 even generates any kind of traffic, it won’t label it with 24010 at all, so how does ECMP work with a multhomed VPWS? Thank you David