Wednesday, October 17, 2018

VPLS Loop Prevention

image.png

Whats the best way to prevent loops in the above diagram. Core routers are configured for VPLS, and carrier is providing us with 2 fibres presenting us with 3 vlans from remote sites.

I've looked at using VPLS Multihoming, but as soon as i specify an active interface the lab breaks. I am passing multiple vlan's through the routers which might be the problem

Routing instance config:

BGP1

instance-type vpls;

vlan-id all;

interface ge-0/0/5.11;

interface ge-0/0/5.12;

interface ge-0/0/6.11;

interface ge-0/0/6.12;

route-distinguisher 4001:10000;

vrf-target target:65000:10000;

protocols {

vpls {

no-tunnel-services;

site site2 {

site-identifier 2;

site-preference primary;

}

}

}

BGP0

instance-type vpls;

vlan-id all;

interface ge-0/0/5.11;

interface ge-0/0/5.12;

interface ge-0/0/6.11;

interface ge-0/0/6.12;

route-distinguisher 3001:10000;

vrf-target target:65000:10000;

protocols {

vpls {

no-tunnel-services;

site site2 {

site-identifier 2;

site-preference backup;

}

What im struggling with is how these failover. Adding "active interface" stops it working entirely. Ive read that a physical link down will cause them to flip from active/standby, but im unable to lab this in GNS3 as physical interfaces are always up, even when admin down



No comments:

Post a Comment