Inter-AS Option B Multi-Homed
Load inter.as.l3vpn.option.b.multi.homed.init.cfg
#IOS-XE
config replace flash:inter.as.l3vpn.option.b.multi.homed.init.cfg
#IOS-XR
configure
load bootflash:inter.as.l3vpn.option.b.multi.homed.init.cfg
commit replace
y
A link has been added between R3 and R4. Configure a multi-homed inter-AS option B setup, so that R7/R8 and R9/R10 have reachability. Everything has been preconfigured except for the inter-AS setup.
Use R1-XR1 as the primary link for VPN_A, and R3-R4 as the primary link for VPN_B. In the case of link failure, the other remain link should provide backup.
Answer
Explanation
This lab asks us to configure a multi-homed inter-AS option B setup. The VPN_A VRF should use the R1-XR1 primarily, and the VPN_B VRF should use the R3-R4 link primarily.
This is similar to the option A multi-homed setup, in that we use standard BGP tools to configure the multi-homing. However, we now have a single VPNv4 session between ASBRs instead of one IPv4 unicast session per VRF. For this reason, we set local pref on matching extcommunity values. On R1 and XR1, we match the VPN_A extcommunity and set LP=110. On R3 and R4 we match the VPN_B extcommunity and set LP=110.
Otherwise, this configuration is just basic option B configuration.
A static route is needed on XR1 to fix the label forwarding issue.
Next-hop-self is needed on each ASBR’s VPNv4 session with internal routers to terminate the VPN LSP
There are three LSPs: PE-ASBR, ASBR-ASBR, ASBR-PE
The RT filter needs to be disabled on all ASBRs
Verification
We can verify that VPN_A traffic uses the R1-XR1 link:
VPN_B traffic uses the R3-R4 link:
Shut down the R1-XR1 link and verify that VPN_A traffic can use the R3-R4 link:
Bring the link back up, shutdown the R3-R4 link, and verify VPN_B traffic can use the R1-XR1 link:
Summary
The goals of this lab are very similar to the previous lab, but we are using option B instead of option A. All VPNv4 routes are exchanged over a single eBGP session in option B, so we cannot use a per-neighbor route-policy like we used in option A. Instead, we identify routes based on the extcommunity value and set the LP on matching routes on each ASBR. This configuration is a little less command-intensive because we do not need to define the individual VRFs and per-VRF neighbors on every ASBR.
Last updated