[PWE3] High level comments on draft-ietf-pwe3-mpls-eth-oam-iwk-03

Yuji Tochio <tochio@jp.fujitsu.com> Sun, 01 August 2010 13:54 UTC

Message-ID: <4C557CAF.4070202@jp.fujitsu.com>
Date: Sun, 01 Aug 2010 15:54:55 +0200
From: Yuji Tochio <tochio@jp.fujitsu.com>
User-Agent: Thunderbird 2.0.0.24 (Windows/20100228)
MIME-Version: 1.0
To: pwe3@ietf.org
Content-Type: text/plain; charset="ISO-2022-JP"
Content-Transfer-Encoding: 7bit
Subject: [PWE3] High level comments on draft-ietf-pwe3-mpls-eth-oam-iwk-03
Precedence: list

Hi Nabil and all,

As I promised in IETF#78, I tried to review the whole draft on my
way home from Maastricht.
Here what I list is high level comments on this. For details (page
by page), I will send to Editor independently.

Please reply/consider following comments or concern for
completing this work, before WG LC.

Regards,
Yuji

---------------------
1) As I had comment in Maastricht, the OAM architecure of
Etherent should be clarified in this document. It is noted that
both 802.1ag and Y.1731 define clearly this and PW (IETF) should
address how this is incooperated in PW OAM architecture with
respects to these SDOs.
(Other SDOs would be kind enough to ask for comments by sending
LSs...)

2) Related to 1), the usage of local and remote (MEP) as mostly
used in section 3 is really unclear and confusing. Please clarify this.

3) As alco commented in Maastrichit, this draft should clearly
indicate which OAM architectures as defined in [PW-OAM-MSG-MAP]
are refrered -- Single Emulated mode (Figure 3 in OAM-MSG) or
Coupled mode (Figure 4) or both.

4) As to 2), the editor (Nabil) replied that only coupled mode is
considerede. If so this document should clearly indicate that only
couplned mode is only one supported or considered.

5) If 3) is the way for this draft progressing, it should add that
ETH OAM for CE-CE may be indipendent of this document since the
ME (MD) of CE1 -- PE1 may be differrnt from that of CE2 -- PE2
(, therefore CE-CE may be independent of this document)

6) If 4) is against for the scope of this draft, this draft should consider
single Eumlated mode for making ETH OAM happy :-) (As Italo also
commented)

7) It should be moted that there are some detect issued by CCM
reception. This draft only considers dLOC as defined as G.8021 (ITU-T).
However, for CCM reception, there are number of defects as dUNL, dMMM,
and dUNP. It is noted that following listed Consequent actions are aBLK
where the traffic is blocked. So Naturally one question is arised :
‘How about the action for PW when Traffic is blocked by MEP?’

8) Partly related to 7), Y.1731 (and G.8021) defines LCK and MEP
defected LCK is issued defect to cleint layer. How about considering dLCK?
The issue is that this draft seems to consider the peer (not server/client)
case. Even if peer case, does this draft propose to contribute tto PW
transit staus ?

---------------------

[PWE3] High level comments on draft-ietf-pwe3-mpl… Yuji Tochio