[PDB-tech] proposed new attribute: max_allowed_peering_next_hop_latency

[Job Snijders]

On Thu, Apr 21, 2016 at 01:37:46PM +0200, Arnold Nipper wrote:
> > I propose a new attribute for network records called
> > "max_allowed_peering_next_hop_latency".
> >
> > The purpose of this attribute is to signal to other interested
> > parties (human peering coordinators looking for new interconnection,
> > or computer scripts generating route server configs) what the
> > maximum unidirectional latency can be between the two (potentially)
> > peering routers.
> 
> Would that really be a single number only. Imho it could make sense to
> take into the place where you are peering.

Yes, our peering policy has one such number. Exceptions may be made but
that is not part of the policy itself.

Would it be an idea to start with one value, and in a later, separate
discussion brainstorm about regional granularity? Such granularity would
apply to max prefix v4|v6 too, or maybe IRR records. I'd rather see a
generalized mechanism for regional granularity then shove something in
specific to latency.

Kind regards,

Job