Segmented Pseudowire
draft-ietf-pwe3-segmented-pw-18

[Ballot comment]
Comment updated for version 17 of the draft.

Thanks for the efforts updating this document and taking my concerns
into account. I am clearing my Discuss, but moving one last issue from the Discuss to this Comment as follows:

I still have trouble understanding the phrase "two control planes." As far as I am concerned there is only one control plane, but there may be many control plane protocols, control plane domains, or instances of the same control plane protocol. I am assuming that one or all of these meanings is covered by the text in the draft, and that the working group understands the term even if I don't.

---

There is also one Comment left over from before. In Section 7 you have:

> There are three MPLS to MPLS PW control planes:
>        -i. Static configuration of the PW.
>      -ii. LDP using FEC 128
>      -iii. LDP using the generalized FEC 129

Option i. is not a control plane.
Suggest you change to:
  There are three different MPLS to MPLS PW setup mechanisms

[Ballot comment]
Comment updated 2010-06-05 after discussions.

Some elements of the previous Discuss have been moved to be Comments. Some Comments have been cleared.

---

Sub-TLVs of the S-PE TLV

I agree that there is no *need* to mention ordering; this is standard for LDP. I don't really think that saying so would "clutter the text".

I understand that the whole S-PE TLV is OPTIONAL. This is clear in the text.

I can *deduce* that the sub-TLVs not marked as OPTIONAL are mandatory. I don't really think that saying so would "clutter the text".

The motivation for my original question was: how does a receiver know when an S-PE TLV is malformed, and what do they do about it?

It seems that you are saying that it is "obvious to one normally skilled in the art." I feel that you could cover the case and achieve perfect clarity with a very few words.

---

Expand "PSN" in the Abstract

---

Section 3

Suggest you delete "and in the vast majority of cases this will be
sufficient" as it tends to imply that the whole I-D is pointless.

---

Section 3                     

How much of the motivation stated here is a duplication of what is in
draft-ietf-pwe3-ms-pw-arch? has it been updated to relfect recent
changes to that draft? Would it be better to write less here and point
out to the requirements and architecture documents?

---

Section 3, Bullet i.

>    When it is not possible, desirable or feasible to establish
>    a PW control channel between the terminating source and
>    destination PEs.

This is not a complete sentence.

Suggest
s/confine MD5 key administration/confine key administration/

---

Section 3

s/pertining/pertaining/

---

Section 3

>  There are four different signaling protocols that are defined to
>  signal PWs:
>      -i. Static configuration of the PW (MPLS or L2TPv3).
>      -ii. LDP using FEC 128
>    -iii. LDP using the generalized FEC 129
>      -iv. L2TPv3

Option i. is not a signaling protocol.
Options ii. and iii. are the same signlaing protocol.

Suggest you change to:
  There are four mechanisms to establish PWs:

This also shows up in Section 7

---

Terminology...

Terms like S-PE and T-PE turn up unintroduced and without even a
reference.

---

Section 6

>  The performance may be further
>  degraded by the emulation process, and performance degradation may be
>  further degraded by traversing multiple PW segments.

The "degradation may be degraded" ?

---

Section 7 first paragraph

It would be nice to replace the passive voice with something more
specific.

---

Section 7.1 covers two of the scenarios listed in Section 7 that "must
be considered in detail." Could you split this into two Sections so
that each scenario gets its own section?

[Ballot discuss]
Discuss updated 2010-06-05 to reflect progress made.

A lot of progress made.

---

1. On March 12, Stewart Bryant wrote in an email to the PWE3 WG...

> I think that we need to proceed a follows:
>
[SNIP]
>
> Then I think that we need to do three things
>
> 1) Ask MPLS WG if they are happy with the
> LDP material.
>
> 2) Ask L2TPext WG if they are happy with the
> L2TP text.

The emails were sent out on 2010-06-03. I am holding this Discuss while discussions progress on the mailing lists.

---

2. In the area of how end-to-end MS-PWs are planned, routed, and set up, especially when different segments use different signaling protocol instances or types, or when some segments are signaled and some manually provisioned.

This is a very fundamental issue to me and I would expect a document with this scope to cover it fully. But perhaps we can avoid me having to hold my nose on this document by including a punt. For example, you could say, "All elements of the establishment of end-to-end MS-PWs including routing and signaling are out of scope of this document and any discussion in this document serve purely as examples."

---

3. I still cannot square the language of "two control planes" that shows up multiple times in the document. I think this means "different signaling protocol instances or types".

---

4. Section 12.2

You should include at least an informational reference to
draft-ietf-mpls-mpls-and gmpls-security

[Ballot discuss]
Hilarie Orman's (late) secdir review raised one important issue that merits a
paragraph or two in the security considerations section.  Specifically, she
noted that

"The looming security problem is that the only way to make sure that
connections are made correctly is to have a table of endpoints and
pre-placed shared keys between them.  Is this at all realistic, given
that multi-segment PWs are for situations in which an organization has
developed an architecture too rich for single-segement PWs?"

Since the underlying protocols for exchanging control information rely
on pre-placed keys, and we really don't have anything better (e.g., no
automated key management solution) to offer right now, I think we
should document the complexity and fragility of this solution.

[Ballot comment]
The Gen-ART Review by Sean Turner on 31-Aug-2009 raises a few questions:

  - Do you really want 7 authors?  The RFC Editor prefers 5 or less.

  - Should the Introduction be Section 1?

  - Sec 7.4.1: Indicates the PW Switching Point description string is
    80 characters long.  Should it say more about the character set?

  - Sec 8.4.1: Figure shows both AC "Up" and AC "UP".  Should they all
    be "Up"?

  - Sec 16: s/Author's/Authors'

  - There are a couple of places where " , " should be changed to ", "

[Ballot comment]
Do you really need seven authors on the front page?

---

It would be really good if the Introduction was the first section in
the draft.                                                        t

---

Expand "PSN" in the Abstract

---

Section 3

Suggest you delete "and in the vast majority of cases this will be
sufficient" as it tends to imply that the whole I-D is pointless.

---

Section 3                                                                 

How much of the motivation stated here is a duplication of what is in
draft-ietf-pwe3-ms-pw-arch? has it been updated to relfect recent
changes to that draft? Would it be better to write less here and point
out to the requirements and architecture documents?

---

Section 3, Bullet i.

>    When it is not possible, desirable or feasible to establish
>    a PW control channel between the terminating source and
>    destination PEs.

This is not a complete sentence.

s/would require tp/would require to/

Suggest
s/confine MD5 key administration/confine key administration/

>    A second example might involve a single AS where the PW
>    setup path between the terminating PEs is computed by an
>    external entity (i.e. client-layer routing protocol).
I think you mean s/i.e./e.g./  but I am not sure what a "client-
layer routing protocol" is, and I would suggest you simply
remove the bracketted text.

s/pertining/pertaining/

---

Section 3

>  There are four different signaling protocols that are defined to
>  signal PWs:
>      -i. Static configuration of the PW (MPLS or L2TPv3).
>      -ii. LDP using FEC 128
>    -iii. LDP using the generalized FEC 129
>      -iv. L2TPv3

Option i. is not a signaling protocol.
Options ii. and iii. are the same signlaing protocol.

Suggest you change to:
  There are four mechanisms to establish PWs:

This also shows up in Section 7

---

Terminology...

Terms like SPE and TPE turn up unintroduced and without even a
reference.

---

Section 6

>  The performance may be further
>  degraded by the emulation process, and performance degradation may be
>  further degraded by traversing multiple PW segments.

The "degradation may be degraded" ?

---

Section 7 first paragraph

It would be nice to replace the passive voice with something more
specific.

---

Section 7.1 covers two of the scenarios listed in Section 7 that "must
be considered in detail." Could you split this into two Sections so
that each scenario gets its own section?

---

Section 7.3

The title of this section...
> 7.3. LDP FEC 128 to LDP using the generalized FEC 129
...implies mapping in one direction only.
But I think the ST-PE can use either one of the two signaling types
with the S-PE being called on to map to the other type.

---

Section 9.1

The following paragraph needs polish.

>  Single-hop pseudowires are signaled using the Virtual Circuit
>  Connectivity Verification (VCCV) parameter included in the interface
>  parameter field of the PW ID FEC TLV or the interface parameter sub-
>  TLV of the Generalized PW ID FEC TLV as described in [RFC5085]. When
>  a switching point exist between PE nodes, it is required to be able
>  to continue operating VCCV end-to-end across a switching point and to
>  provide the ability to trace the path of the MS-PW over any number of
>  segments.

Presumably the SS-PW is only signaled using the VCCV parameter if VCCV
is required?

It is confusing to say "when a switching point exists between PEs"
because you have chosen to call those switching points PEs as well!

[Ballot discuss]
Sorry for a very lengthy Discuss. I think this work has significant merit and is in line with the requirements and architecture documents that have been previously approved. Nevertheless, I think this document is in need of some considerable work before it should be published as an RFC. I have tried to separate important points as Discusses, and smaller points as Comments.

---

On March 12, Stewart Bryant wrote in an email to the PWE3 WG...

> I think that we need to proceed a follows:
>
[SNIP]
>
>
> Then I think that we need to do three things
>
> 1) Ask MPLS WG if they are happy with the
> LDP material.
>
> 2) Ask L2TPext WG if they are happy with the
> L2TP text.

I had a quick scan of the MPLS archive and didn't find any mail on
this topic. Did these mails get sent?
(Yes, most of the relevant people participate on all of the lists,
but we should keep the process clean.)

---

Section 2

Many of the terms are repeated from draft-ietf-pwe3-ms-pw-arch. But the
definitions are subtly different. If this is intentional it should be
highlighted. If it is not intentional, you must fix it, and I suggest
you simply list the terms and point at the Architecture for the formal
definitions.

---

The Abstract says...

>  This document describes how to connect pseudowires (PW) between
>  two distinct PW control planes or PSN domains.

But there is an example in Section 3 which contradicts this...

>    A second example might involve a single AS where the PW
>    setup path between the terminating PEs is computed by an
>    external entity (i.e. client-layer routing protocol). Assume
>    a full mesh of PWE3 control channels established between
>    PE-A, PE-B and PE-C. A client-layer L2 connection tunneled
>    through a PW is required between terminating PE-A and PE-C.
>    The external entity computes a PW setup path that passes
>    through PE-B. This results in two discrete PW segments being
>    built: one between PE-A and PE-B and one between PE-B and
>    PE-C. The successful client-layer L2 connection between
>    terminating PE-A and terminating PE-C requires that PE-B
>    performs the PWE3 switching process.

Anyway, I find this example very suspiscious!

Why does the client layer care about the routing in the server
layer?

But if it does, why doesn't it set up an LSP tunnel A-B-C and
insert the single segment PW into that tunnel.

Note that this example of yours is not the case of using MS-PW
for scaling improvements which (for some reason) is not listed
in this section.

---

Section 4

>  There are two methods for switching a PW between two PW control
>  planes. In the first method (Figure 2), the two control planes
>  terminate on different PEs.

"switching between two control planes" Huh?
The control plane has got nothing to do with the figures that follow
that are about the architecture or the data plane.

I think that the bulk of the functionality is, indeed, control plane-
related. But Section 7 is (or should be) talking about now to map
between PW setup techniques at S-PEs, not how to switch between
conrol planes. Perhaps this sort of language could be used through-out
the document.

---

Anyway, what is this obsession with multiple control planes?


>  In Figure 3 SPE1 runs two separate control planes: one toward TPE1,
>  and one Toward TPE2. The PW switching point (S-PE) is configured to
>  connect PW Segment 1 and PW Segement 3 together to complete the
>  multi-hop PW between TPE1 and TPE2. PW Segment 1 and PW Segment 3
>  MUST be of the same PW type, but PSN Tunnel 1 and PSN Tunnel 2 need
>  not be the same technology.

In the primary example in the architecture document, this doesn't
apply. Indeed, there may be separate LDP sessions, but the control
plane is the same.

On the other hand, Section 3 lists four examples of PW set up mechanisms
one of which is the management plane. That makes the discussion of
control planes very confusing. Furthermore, the document gets a bit
confused about whether it is talking about different control plane
protocols while giving specific statements only about one protocol.
For example, in Section 4 there is a paragrpah that starts to talk about
LDP/L2TP and ends up talking about specific LDP messages...

>  The general approach taken for MS-PWs is to connect the individual
>  control planes by passing along any signaling information immediately
>  upon reception. First the S-PE is configured to switch a PW segment
>  from a specific peer to another PW segment destined for a different
>  peer. No control messages are exchanged yet as the S-PE does not have
>  enough information to actually initiate the PW setup messages.
>  However, if a session does not already exist, a control protocol
>  (LDP/L2TP) session MAY be setup. In this model the MS-PW setup is
>  starting from the T-PE devices. Next once the T-PE is configured it
>  sends the PW control setup messages. These messages are received by
>  the S-PE, and immediately used to form the PW setup messages for the
>  next SS-PW of the MS-PW. If one of the S-PEs doesn't accept an LDP
>  Label Mapping message then a Label Release message may be sent back
>  to the originator T-PE depending on the cause of the error. LDP
>  liberal label retention mode still applies, hence if a PE is simply
>  not configured yet , the label mapping is stored for future use. A
>  MS-PW is declared UP only when all the constituent SS-PWs are UP.

---

Section 7.1 suggests handling the mapping of a signaled PW segment to
a static PW segment by modeling the static segment on an AC. This is
reasonable, but I wonder if it fits perfectly. In a single segment PW
case, what happens at the egress PE if there is no AC configured when
the PW is signaled? Does the PW come up and wait for the AC to be
configured, or does the PW signaling fail?

In the MS-PW case, you definitely need to allow the signaled segment to
be established independently of the configuration of the static segment.

---

Sections 7.1 and 7.2

I don't like the imbalance between the cases where the downstream
segment is always signaled, but the upstream segment is either static
or signaled. In the former case, the S-PE is called upon to be
configured and take an active role being responsible for initiating
the establishment of the second segment. In the latter case, the PE is
required to be passive and wait for the first segment to be set up so
that it can learn the parameters to use.

Why can't a segment be set up in advance and then used either by to
be switched from a static segment or from a signaled segment?

I can appreciate that it is possible that the model you want to use
concentrates only on the provision of individual end-to-end services.
In this case my concern is not important, but your Introduction needs
to highlight the model (as should the Architecture draft!) and should
state explicitly that you do not support the construction of "ready-
to-use" segments without a specific end-to-end service to use them.

In fact, reading on, I find...

>  The Pseudowire Identifier (PWID) , as defined in [RFC4447] is a
>  unique number between each pair of PEs. Hence Each SS-PW that forms
>  an MS-PW may have a different PWID.

[NB s/) ,/(,/]

This tends to suggest that you require configuration of S-PEs to include
PWID mapping in both the dual-signaled and the signaled/static cases. In
which case, I don't see why you need to require support of only end-to-
end setup in the dual-signaled case.

Should you break sections 7.2 and 7.3 into separate end-to-end and
segment-by-segment sub-cases?

---

Section 7.2.1

The term "MS-PW path" is used and has not been properly defined in this
document.

I think you are not concerned about the path of PWs (although you might
be), but really care about the sequence of switching points. That is,
you care about the segment sequence of the PE sequence, not the nodes
along the path of the segments themselves.

Anyway, it would be really wise to define this term clearly.

---

Section 7.2.1

>  To avoid this situation one of the T-PE MUST start the PW
>  signaling (active role), while the other waits to receive the LDP
>  label mapping before sending the respective PW LDP label mapping
>  message. (passive role). When the MS-PW path not statically
>  configured, the Active T-PE (the ST-PE) and the passive T-PE (the
>  TT-PE) MUST be identified before signaling is initiated for a given
>  MS-PW.

I am confused! Do these "MUST" conditions apply only *if* the operator
wants to avoid the situation Or is it mandatory in all cases?

Later you allow manual configuration of the roles. Clearly this allows
for misconfiguration, so the protocol may see cases where both T-PEs
believe they are active. How is this case handled, or do we accept that
if this happens diverse routing might happen?

That would tend to suggest that the previous "MUST" only applies to
"if the operator wants to avoid this situation."

---

Section 7.2.1 and Section 7.4

7.2.1 seems to imply that is important to be able to assign the
sequence of S-PEs to be traversed. Presumably you need to signal
this sequence.

7.4 seems to define exactly what you need, but says the TLVs are
only for reporting. Is that intentional? Why don't you need to
signal the sequence?

---

Section 7.4 and Section 7.4.1

Need to introduce consitency in terminology. They "TLV types"
described in 7.4 are the "sub-TLV types" in 7.4.1.

---

Section 7.4.1

Does it matter what order the sub-TLVs are present in?
Most are described as optional. Some do not have any mention of
mandatory or optional. One has "optional" in lower case.

Can the S-PE TLV be empty?

---

Section 7.4 and Section 7.6

The policy-driven removal of S-PE TLVs is good and reasonable,
however it means that the loop detection mechanisms of 7.6 will
not work.

You either have to flag this up, or you have to insist that a
domain identifier is inserted and never removed so that macro-
looping can be detected.

---

The utility of the S-PE TLV is described in 7.4 as
>  The S-PE TLV MUST be
>  sent if VCCV operation is required beyond the first MS-PW segment
>  from a T-PE.

But Section 9.5.2.5 shows how to do MS-PW Path Trace using VCCV.
9.5.2.5 does not appear to rely on the path supplied using the
S-PE TLV.

What am I missing?                                                     

---

Section 12.2

The reference to RFC 3036 needs to be turned into a citation and
should be to RFC 5036.

You should include at least an informational reference to
draft-ietf-mpls-mpls-and gmpls-security

[Ballot comment]
I also support Lars Eggerts discuss regarding congestion control. It seems that multi-segment in its various combinations have many potential ways of breaking the congestion control loop. Thus that fact and what can be done on the signalling plane by the S-PE is important.

[Ballot discuss]
Section 8.7:

"There are some
  situations where an additional amount of interworking must be
  provided between the two data planes at the S-PE, however this is
  outside the scope of this document."

Why isn't the situations that aren't explicitly supported not enumerated if you know about them. This appears to me to be saying: Yes, there are some cases where thinks doesn't work. But we are not going to tell you when. Have fun finding your self. I do believe the document should be clear when it works or not. I don't have a problem with them not being solved, but you need to inform when it does fail.

Section 9.5.2.4:

-iii. T-PE1 builds a second VCCV echo request based on the
            infomation obtained from the control plane (S-PE TLV). It
            then increments the TTL and sends it out to T-PE2. Note that
            the VCCV echo request packet is switched at the S-PE
            datapath and forwarded to the next downstream segment
            without any involvement from the control plane.

The above text seems to assume that the MS-PW only has two segments. My impression was that MS-PWs could consist of more than 2 segments.

Also section 9.5.2.5 seems to lack appropriate wording for more than 2 segments.

[Ballot discuss]
DISCUSS: RFC3985 describes some mandatory congestion control functions
  that PWEs operating in a single PSN domain must adhere to. I'm missing
  a section in this document that describes how similar functions are
  provided for a segmented PWE. It may well be the case that the
  functions that each PSN employs for each PWE segment are already
  resulting in the correct behavior for the entire segmented PSN. It may
  also be the case that some additional functionality is needed - the
  document should discuss this.

[Ballot comment]
The Gen-ART Review by Sean Turner on 31-Aug-2009 raises a few questions:

  - Do you really want 7 authors?  The RFC Editor prefers 5 or less.

  - Should the Introduction be Section 1?

  - Sec 7.4.1: Indicates the PW Switching Point description string is
    80 characters long.  Should it say more about the character set?

  - Sec 8.4.1: Figure shows both AC "Up" and AC "UP".  Should they all
    be "Up"?

  - Sec 16: s/Author's/Authors'

  - There are a couple of places where " , " should be changed to ", "

IANA questions/comments:

NOTE: please see our question about Action 4.

Upon approval of this document, IANA will complete the following actions:

ACTION 1:

New assignment in the "Control Message Attribute Value Pair" registry at
http://www.iana.org/assignments/l2tp-parameters

Attribute
Type Description Reference
--------- ------------------ ---------
TBA-L2TP-AVP-1 PW Switching Point AVP [RFC-pwe3-segmented-pw-13]


ACTION 2:

In the "TLV TYPE NAME SPACE" registry at
http://www.iana.org/assignments/ldp-namespaces

The value 0x096D has already been assigned temporarily.
IANA will remove the temporary state of the assignment and make it
permanent.
The assignment in the registry will then be:

Range Description Reference
----------------- ----------------------------------- ---------
0x096D Pseudowire Switching Point PE TLV
[RFC-pwe3-segmented-pw-13]


ACTION 3:

In the "Status Code Name Space" registry at
http://www.iana.org/assignments/ldp-namespaces

The value 0x0000003A has already been assigned temporarily.
IANA will remove the temporary state of the assignment and make it
permanent.
The assignment in the registry will then be:

Range/Value E Description Reference
------------- ----- ---------------------- ---------
0x0000003A 0 PW Loop Detected
[RFC-pwe3-segmented-pw-13]


ACTION 4:

QUESTION: In section 13.4, authors request the following assignment:
"Sequencing not supported" in the "L2TPv3 Result Codes" registry.

This registry has sub-registries such as:
"Defined Result Code values for the StopCCN message"
"Defined Result Code values for the CDN message"
"Defined Result Code values for the MSEN message"
"General Error Codes"

However, the draft does not specify the sub-registry in which this
assignment should be made. Moreover, the text in section 8.7.1 says:

"the switching node SHOULD NOT allow
the session to be established by sending a CDN message with
Result Code set to 17 "sequencing not supported" (subject to
IANA Assignment)"

... which seems to imply that the sub-registry would be "Defined
Result Code values for the CDN message" and that the value should
be 17. However, 17 has already been assigned in that sub-registry.

Please let us know which sub-registry this assignment should come
from.


ACTION 5:

New registry at http://www.iana.org/assignments/pwe3-parameters

NOTE: "IETF Consensus" should be changed to "IETF Review," per
RFC 5226.

Registry Name: Pseudowire Switching Point PE TLV Type
Reference: [RFC-pwe3-segmented-pw-13]
Range Registration Procedures Notes
-------- ----------------------------- ----------------------
0 IETF Review
1-64 Expert Review
65-127 IETF Consensus
128-254 First Come First Served reserved for vendor proprietary
extensions
255 IETF Review

"Type" range is 8bit (0-255)

Type Length Description Reference
----- ------ -------------------------------------- ----------
0x01 4 PW ID of last PW segment traversed
[RFC-pwe3-segmented-pw-13]
0x02 variable PW Switching Point description string
[RFC-pwe3-segmented-pw-13]
0x03 4/16 Local IP address of PW Switching Point
[RFC-pwe3-segmented-pw-13]
0x04 4/16 Remote IP address of last PW Switching
[RFC-pwe3-segmented-pw-13]
Point traversed or of the T-PE
0x05 variable FEC Element of last PW segment traversed
[RFC-pwe3-segmented-pw-13]
0x06 10 L2 PW address of PW Switching Point
[RFC-pwe3-segmented-pw-13]

draft-ietf-pwe3-segmented-pw-13.txt

Document Shepard Write-Up

(1.a) Who is the Document Shepherd for this document? Has the
Document Shepherd personally reviewed this version of the
document and, in particular, does he or she believe this
version is ready for forwarding to the IESG for publication?

Stewart Bryant (stbryant@cisco.com)
Yes, I have reviewed the document and I believe it is ready
for forwarding to the IESG.

(1.b) Has the document had adequate review both from key WG members
and from key non-WG members? Does the Document Shepherd have
any concerns about the depth or breadth of the reviews that
have been performed?

Yes, the document has received adequate review in the PWE3
WG.

(1.c) Does the Document Shepherd have concerns that the document
needs more review from a particular or broader perspective,
e.g., security, operational complexity, someone familiar with
AAA, internationalization or XML?

No.

(1.d) Does the Document Shepherd have any specific concerns or
issues with this document that the Responsible Area Director
and/or the IESG should be aware of? For example, perhaps he
or she is uncomfortable with certain parts of the document, or
has concerns whether there really is a need for it. In any
event, if the WG has discussed those issues and has indicated
that it still wishes to advance the document, detail those
concerns here. Has an IPR disclosure related to this document
been filed? If so, please include a reference to the
disclosure and summarize the WG discussion and conclusion on
this issue.

I have no specific concerns

(1.e) How solid is the WG consensus behind this document? Does it
represent the strong concurrence of a few individuals, with
others being silent, or does the WG as a whole understand and
agree with it?

I am comfortable that the document represents WG consensus and
has been reviewed by a significant number of active WG
participants.

(1.f) Has anyone threatened an appeal or otherwise indicated extreme
discontent? If so, please summarise the areas of conflict in
separate email messages to the Responsible Area Director. (It
should be in a separate email because this questionnaire is
entered into the ID Tracker.)

None indicated.

(1.g) Has the Document Shepherd personally verified that the
document satisfies all ID nits? (See
http://www.ietf.org/ID-Checklist.html and
http://tools.ietf.org/tools/idnits/). Boilerplate checks are
not enough; this check needs to be thorough. Has the document
met all formal review criteria it needs to, such as the MIB
Doctor, media type and URI type reviews?

There are no nits issues that gate publication.

This document is not subject to MIB doctor or other reviews.

(1.h) Has the document split its references into normative and
informative? Are there normative references to documents that
are not ready for advancement or are otherwise in an unclear
state? If such normative references exist, what is the
strategy for their completion? Are there normative references
that are downward references, as described in [RFC3967]? If
so, list these downward references to support the Area
Director in the Last Call procedure for them [RFC3967].

Yes, the references are split appropriately.

All normative references are existing RFCs

(1.i) Has the Document Shepherd verified that the document IANA
consideration section exists and is consistent with the body
of the document? If the document specifies protocol
extensions, are reservations requested in appropriate IANA
registries? Are the IANA registries clearly identified? If
the document creates a new registry, does it define the
proposed initial contents of the registry and an allocation
procedure for future registrations? Does it suggest a
reasonable name for the new registry? See [RFC5226]. If the
document describes an Expert Review process has Shepherd
conferred with the Responsible Area Director so that the IESG
can appoint the needed Expert during the IESG Evaluation?

The IANA considerations section exists and is correct.

(1.j) Has the Document Shepherd verified that sections of the
document that are written in a formal language, such as XML
code, BNF rules, MIB definitions, etc., validate correctly in
an automated checker?

There are no sections that use a formal language.

(1.k) The IESG approval announcement includes a Document
Announcement Write-Up. Please provide such a Document
Announcement Write-Up? Recent examples can be found in the
"Action" announcements for approved documents. The approval
announcement contains the following sections:

Technical Summary
This document describes how to connect pseudowires (PW)
between two distinct PW control planes or Packet Switched
Network (PSN) domains. The PW control planes may belong to
independent autonomous systems, or the PSN technology is
heterogeneous, or a PW might need to be aggregated at a
specific PSN point. The PW packet data units are simply
switched from one PW to another without changing the PW
payload.

This document is a product of the PWE3 working group.

This document is Standards Track

Working Group Summary

This document is a solutions document to the "Requirements
for Multi-Segment Pseudowire Emulation Edge-to-Edge (PWE3)"
(RFC5254). It conforms to "An Architecture for Multi-Segment
Pseudowire Emulation Edge-to-Edge" draft-ietf-pwe3-ms-pw-
arch-07.txt



Document Quality
There are no concerns about protocol quality. There are
understood to be implementations of this protocol.

Personnel
Who is the Document Shepherd for this document?

Stewart Bryant (stbryant@cisco.com)

Who is the Responsible Area Director?
Ralph Droms