Hi there,
On 09/08/2017, 19:34, Stefan Ideler via db-wg typed:
>
>
> As a result our route objects regardless of origin are noted in the RIPE database. Our upstreams use it to generate their prefix filters (and not all of them support multiple databases)
I won't comment here on your specific use case or the overall reasoning or motivation for wanting to use the RIPE DB exclusively, even for resources that may come from various other regions. I have an opinion, but it's just that. An opinion.
What I'd like to comment on is the excerpt above. Specifically about upstream providers not supporting multiple databases.
That has proven to never really be an issue. When a network upstream of another network says they don't support additional databases for IRR, so far that has always proven to actually mean they've never needed to think about supporting multiple databases, and never had the demand to do the work to change that.
In other words, it's not that they *can't* support other databases. It's that they never have, and it's just the "way things have been done". It's always ended up that they *can* support multiple databases if asked to.
I speak from experience here. If you go back into the archives of this list, when this issue was first brought to the list, I responded with this exact fear: But what about the upstream support?
Since then I've had the opportunity to speak to folks from some of the larger networks that are often someone's upstream on behalf of AFRINIC members from time to time, and also in my own capacity in the context of an RIR meeting event network. We've also encouraged member operators to just ask their upstreams to change. And this has proved to never have remained an issue to the best of my knowledge.
Summary: While you may (or may not) have other good reasons for the RIPE DB to be open (or not), upstream support of alternates is not likely one of the reasons.
Best regards,
Daniel
