Creating route objects without mnt on existing ones
Hi there It seems to be impossible to create route objects for inet(6)nums, if there are other route objects to which your maintainer has no access. Even if you have mnt-by + mnt-routes on the inet(6)nums. Due to that, it's much more difficult to create new route objects for our customers, as we have to contact the maintainer of all existing objects in order to get mnt-by/mnt-routes on their route objects. In my opinion, that makes zero sense. If you have mnt-by/mnt-routes on an inetnum, you should be able to create objects, regardless of existing ones. Best regards Patrick
Hi there Update: Max gave me the tip that it is possible to force delete the existing route objects & re-create them with the own maintainer. This way it works to add another route object :-) Best regards Patrick On 15.01.21 14:33, Patrick Velder wrote:
Hi there
It seems to be impossible to create route objects for inet(6)nums, if there are other route objects to which your maintainer has no access.
Even if you have mnt-by + mnt-routes on the inet(6)nums.
Due to that, it's much more difficult to create new route objects for our customers, as we have to contact the maintainer of all existing objects in order to get mnt-by/mnt-routes on their route objects.
In my opinion, that makes zero sense. If you have mnt-by/mnt-routes on an inetnum, you should be able to create objects, regardless of existing ones.
Best regards Patrick
Hi there
Update: Max gave me the tip that it is possible to force delete the existing route objects & re-create them with the own maintainer. This way it works to add another route object :-)
Yes. However, I have yet to hear a comprehensible explanation for the current rule. And ... this notion of a "force delete" seems like a kludge to me. Why should you not be able to go from 1 to 2 route objects for the prefix without first having to go to 0 with all the authorizations you listed in place? What if someone decides to generate a prefix list at exactly the wrong moment? Regards, - Håvard
HI guys This is not an issue about the mechanics of the RIPE Database. AFAIK these are the rules that have been developed over time by the routing community. The RIPE Database software simply enforces these rules. If you think the rules need updating to change the who/why/when/how ROUTE(6) objects can be created then I suggest you start a discussion about it on the Routing WG. As the RIPE NCC is hosting an Open House on Routing Security and RPKI next week, maybe they can also cover the issue of ROUTE(6) object security as well? cheersdenisco-chair DB-WG On Friday, 15 January 2021, 16:19:55 CET, Havard Eidnes via db-wg <db-wg@ripe.net> wrote:
Hi there
Update: Max gave me the tip that it is possible to force delete the existing route objects & re-create them with the own maintainer. This way it works to add another route object :-)
Yes. However, I have yet to hear a comprehensible explanation for the current rule. And ... this notion of a "force delete" seems like a kludge to me. Why should you not be able to go from 1 to 2 route objects for the prefix without first having to go to 0 with all the authorizations you listed in place? What if someone decides to generate a prefix list at exactly the wrong moment? Regards, - Håvard
participants (3)
-
Havard Eidnes
-
Patrick Velder
-
ripedenis@yahoo.co.uk