I reviewed already revision 09 of this module [1]. All substantial
objections and suggestions expressed in that review are addressed in
revision 23 and I am satisfied with the result. I especially
appreciate that descriptions were considerably expanded and references
added in many places.

I tested validity of the ietf-ospf module with pyang and Yangson
tools, and found no issues. The comments below are non-substantial and
do not affect practical use of the module.

In summary, I think this YANG module and document is a remarkable
piece of work demonstrating that it is possible to build quite complex
vendor-neutral data model that can be used equally well with several
router plaforms.

Comments:

     - names of locally-defined identities as parameters of XPath
       functions derived-from and derived-from-or-self sometimes have
       the 'ospf:' prefix, sometimes don't. I suggest to be
       consistent, and the option without a prefix looks better to me.

     - RFC 8407 suggests this format of references to RFC:
       RFC XXXX: Title of the Document
       This draft uses a hyphen instead of a colon. I suggest to
       follow the 8407 convention so as to make parsing easier.

     - the title of Sec. 2.8 should be "OSPF Notifications" (plural
       and capitalization)

     - enumerations "nssa-translator-state-type" and
       "restart-status-type" define the value parameter
       for two of their enums but not for the third. This should be
       avoided.

[1] https://datatracker.ietf.org/doc/review-ietf-ospf-yang-09-yangdoctors-lc-lhotka-2017-12-06/