Early Review of draft-ietf-lime-yang-connection-oriented-oam-model-03
|Requested rev.||01 (document currently at 07)|
|Team||YANG Doctors (yangdoctors)|
|Requested by||Mehmet Ersue|
Opsdir Last Call review of -05 by Joe Clarke (diff)
Genart Last Call review of -05 by Paul Kyzivat (diff)
Secdir Last Call review of -05 by Sandra Murphy (diff)
Carl Moberg reviewed earlier draft-ietf-lime-yang-oam-model, which is now replaced by draft-ietf-lime-yang-connection-oriented-oam-model. Carl should review again.
|Reviewed rev.||03 (document currently at 07)|
|Review result||On the Right Track|
|Draft last updated||2018-01-22|
This is my review of draft-ietf-lime-yang-connection-oriented-oam-model-03 and the included YANG module. Overall, the document is in reasonable shape but needs more attention around one fundamental aspect, and a short list of minor issues. My understanding is that this module is intended to be implemented as a network service model, per the module classification RFC (RFC8199). But e.g. the ‘mip' list has a key 'interface' of type 'if:interface-ref'. That type is defined in ietf-interfaces to be a leafref with path "/if:interfaces/if:interface/if:name” assuming that the module context is a single network element. This seems self-contradicting and needs to be resolved. I would suggest running the YANG module through 'pyang -f yang' to provide consistent formatting of the module. I think it would also be worth considering putting the content in the canonical order. E.g. the location of the description statement in the mip-address grouping makes it hard to read. The 'session' list is 'ordered-by user', but does not provide any description or explanation to the meaning of the order. The content of the description strings are in need of language fixes.