Darren Loher

For backwards compatibility, instead of removing the leaf consider setting [status deprecated](https://www.rfc-editor.org/rfc/rfc6020#section-7.19.2) for this leaf and add a new instance-id leaf with reference to rfc8202 to provide clarity of the...

Obsoleted by #816

@wenovus Can you suggest an update based on feedback above?

This is a last call for comments on this pull request. This pull request is now scheduled to merge on Nov 30, 2022.

FYI: the sflow readme might be helpful to refer to: https://github.com/openconfig/featureprofiles/blob/afe4b12d8f80e5fb4c847a3d72a17217eb7468fa/feature/sflow/otg_tests/sflow_base_test/README.md

I think the code is correct as is. An interface name is permitted to be any arbitrary string which is system (vendor) defined. Therefore, we do expect the format of...

I notice that in metadata.textproto, no DUT is assigned the use of deviations.InterfaceRefInterfaceIDFormat for this particular test. If it is truly not needed then the deviation could be removed. But...

> > It is already unique in this case, isn't it? > > Yes, in this testcase. My question was generic. Shouldn't the format of `id` remain same for all...

> 3\. Put a strict protocol around `interface-id` so that it has common meaning. I don't understand why we can't state: > > * If you have an interface without...

@Ankur19 and @liulk do you have some experience with this issue?