Adam Simpson

Already addressed by PR #712

I think the original purpose of this leaf was to support multi-instance ISIS (RFC 8202). In MI-ISIS each of the potentially N different instances on the router needs a unique...

To be clear, what I am suggesting is that maybe this leaf should be replaced with a new leaf called instance-id or even multi-instance-id of type uint16. Rather than merely...

@earies sorry for the slow reply. Agree with your comment that "if we keep the instance leaf, make it a uint16 then we have 2 identifiers for a protocol instance...

I feel this modeling is too specific to one use case. I am wondering if it can be generalized. A packet matching a PBF rule with this new decap-lookup action...

Looks OK to me. The description for decap-network-instance states that this network-instance is expected to have entries that perform decapsulation as a next-hop action. What if it has other types...

So is this the conclusion? Implementations that support multi-instance ISIS (RFC 8202) need an integer instance-id. The OC model can either introduce a new leaf and deprecate the existing one...

> @dplore > > 1. from my point of view, we can deprecate `/network-instances/network-instance/tables/ ` completly. +1 on deprecating /network-instances/network-instance/tables. As configuration objects they serve no purpose other than anchoring...

@dplore, the SRLinux implementation has two different controls: one to control whether the ACL is shared by multiple interfaces or not and another to control whether stats are enabled or...

> Taking a step back here, it seems this problem statement isn't unique to BGP communities (or even BGP) but rather configuration hierarchies, groupings and expected inheritance across years of...