[quagga-dev 11577] Re: [PATCH] BGP: add aspath_aggregate_mpath that preserves path length
dwalton at cumulusnetworks.com
Mon Oct 13 15:02:06 BST 2014
> The selection mechanism I think could have interoperability / routing
> consistency issues I think. I'm curious why there isn't some capabilities
> defined to explicitly signal which path-grouping selection mechanism is
Can you elaborate on the possible issues?
add-path itself just provides a means of sending more than one path for a prefix. A BGP implementation can then have features that decide what additional paths to send to a peer. It would be up to those features to use capabilties to signal parameters specific to that feature. None of the following need to do capability exchange but these are some features that use add-path:
- If you are an RR use add-path to TX the bestpath per nexthop, this way the clients can choose from all nexthops.
- For Route Servers use add-path to TX the bestpath per neighbor-AS
- TX all paths to some type of collector/analyzer. BMP is a better solution here but prior to BMP this was reasonable.
> Then, for BGP ECMP, what happens if a speaker wants to describe an ECMP
> path to a non-Add-Path speaker?
In that scenario you would need to create the AS_SET with all of the ASNs that you are multipathing over.
More information about the Quagga-dev