[Gen-art] Last Call review of draft-ietf-trill-oam-mib-06

[Tom Taylor <[e-mail]>]

Resending just to note that I got a bounce from Tissa's address.

Tom Taylor

I am the assigned Gen-ART reviewer for this draft. For background on
Gen-ART, please see the FAQ at

<http://wiki.tools.ietf.org/area/gen/trac/wiki/GenArtfaq>.

Please resolve these comments along with any other Last Call comments
you may receive. Sorry for the late review, but there was a pile of 
reading to do in preparation. I admit I did not read the whole 1800 
pages of 802.1Q.

Tom Taylor

Document: draft-ietf-trill-oam-mib-06
Reviewer: Tom Taylor
Review Date:        2015-08-14
IETF LC End Date:   2015-08-13
IESG Telechat date: 2015-08-20

Summary: Not quite ready. Minor issues and editorials/nits.

Major issues:

Minor issues:

1) Section 5.2 states that IEEE8021ServiceSelectorType has two values. 
In fact, 801-1Q-2014 enumerates more than that. I'd suggest a slight 
change in wording to reflect this:

OLD

IEEE8021-TC-MIB defines IEEE8021ServiceSelectorType with two values:

- 1 representing a vlanId, and

- 2 representing a 24 bit isid.

NEW

The IEEE8021-TC-MIB definition of IEEE8021ServiceSelectorType includes 
the two values:

- 1 representing a vlanId, and

- 2 representing a 24 bit isid.

2) Section 6.2 indicates that TRILL OAM has no support for Link Trace 
Message/Reply. Perhaps text could be added to say why this is so (i.e., 
that Path Trace has been substituted, as indicated in Sec. 10 of RFC 
7455, and has been supplemented by Multi-destination Tree Verification 
Message/Reply).

3) "Reference Overview" in the MIB module header indicates that the 
TRILL MIB module refers to the original CFM document, IEEE 802.1ag-2007, 
instead of IEEE 802.1-Q-2014. Why the older starting point?

4) Has any thought been given to including an indication of when the 
counters were last reset (e.g, due to restart)?

5) description of trillOamMepTxPtmStatus refers to the MEP Initiator 
State Machine. Reference should include pointer to the description of 
this state machine. Where is it defined? -- not in RFC 7455. Similar 
comment regarding the description of trillOamMepTxMtvmStatus.

6) Description of trillOamMepTxPtmMessages: there is no indication in 
RFC 7455 of how this limit is used. More text is needed here. Does 
number of hops affect the count of transmitted messages against this 
limit? Similar comment regarding the description of 
trillOamMepTxMtvmMessages.

7) Surely this document has normative dependencies on 802.1Q and the 
LLDP-MIB, for which no reference is given, BTW.


Nits/editorial comments:

1) Under "Abbreviations" in the MIB module header, definition of SNMP 
Agent, need to spell out NE. Similarly, spell out EMS and NMS in the 
next definition.

2) Description of trillOamMepTable: s/rowsare/rows are/

3) in the descriptions of trillOamMepPtrFlag and 
trillOamMepPtrErrorCode, incorrect section numbers for RFC 7455 are 
given in the references. Should be 8.4.3?

4) Description of trillOamMepPtrIngress: s/PTM/PTR/. Similar comment for 
trillOamMepPtrEgress.

5) Syntax error for trillOamMepPtrIngressPortIdSubtype: syntax should be 
LldpPortIdSubtype. Descriptions of trillOamMepPtrIngressPortIdSubtype 
and trillOamMepPtrIngressPortId should be interchanged. Similar comments 
for trillOamMepPtrEgressPortIdSubtype and trillOamMepPtrEgressPortId. 
Note that syntax is also stated in the Entry definition and has to be 
fixed there, too.

6) Same problem as 5) for the corresponding trillOamMtvrTable objects.

7) Security Considerations, third para, first-to-second lines: 
s/MAC-ACCESS/MAX-ACCESS/