Use Cases and Operational Experience with Multipath TCP
RFC 8041

Note: This ballot was opened for revision 06 and is now closed.

(Spencer Dawkins) Yes

Comment (2016-09-14 for -06)
No email
send info
Should this text

   The FTP interference is expected and due to
   Application Level Gateways running home routers. 
   
be "running in home routers"? I'm not sure how to parse the sentence otherwise.

When reading this text

   This may be
   excessive in some environments, in particular when the client and/or
   the server have a large number of interfaces. 
   
I am reminded that multiple IP addresses per interface are common in IPv6 and dual stack deployments. That would make the situation worse, wouldn't it? If so, would that be worth saying?

Mirja Kühlewind Yes

(Jari Arkko) (was Discuss) No Objection

Comment (2016-09-15)
No email
send info
I would also like to request that the authors look at Dan's other comments:

---

Summary: Ready with issues

A very useful and well written document, which gathers implementation
and deployment experience and expands the list of the Multipath TCP
Use Cases. A few minor issues described below, if addressed, could
improve the clarity and usability of the document.

Major issues:

Minor issues:

1. The 'Introduction' section starts with the statement:

Multipath TCP was standardized in [RFC6824] and five independent
  implementations have been developed.

Saying 'was standardized' seems misleading to me, as RFC 6824 is an
experimental RFC, so not even standards-track (this putting aside the
discussions whether RFCs are standards). Actually at no point this
document mentions that Multipath TCP is Experimental, this seems odd.

2. It would be useful to clarify the statement about the iOS
implementation of Multipath TCP in the Introduction by mentioning what
'single application' is referred.

However, this particular Multipath TCP implementation is currently only used to support a single application.'

3. I am questioning whether the 'Multipath TCP proxies' section really
belongs to the use cases or rather to operational experience. After
all it's about a strategy of deployment of Multipath TCP in cases
where clients and/or servers do not support Multipath TCP but the need
exists probably because of the combination of one or several other use
cases.

4. In section 3.5:

There have been suggestions from Multipath TCP users to modify the
  implementation to allow the client to use different destination ports
  to reach the server.  This suggestion seems mainly motivated by
  traffic shaping middleboxes that are used in some wireless networks.
  In networks where different shaping rates are associated to different
  destination port numbers, this could allow Multipath TCP to reach a
  higher performance.  As of this writing, we are not aware of any
  implementation of this kind of tweaking.

Beyond the potential problems described in the following paragraph, is
such a 'tweak' consistent with the protocol definition, or would it
need to cause changes in the protocol as defined now? A clear
recommendation seems to be needed here.

5. A more clear recommendation would be useful also in 3.8. It is not
clear here whether the segment size selection is a design or a tuning
issue that can/should be added to applications.


Nits/editorial comments:

1. Section 3.12 contains a timed statement 'As of September 2015 ...'
which should be updated or maybe edited to make it less
time-dependent.

2. It seems to me that [RFC6824] and [RFC6181] should be Normative
References as they describe the protocol extensions, and the initial
list of use cases which is expanded by this document. Without reading
these two documents, this one does not make too much sense.

(Alia Atlas) No Objection

Deborah Brungard No Objection

(Ben Campbell) No Objection

Comment (2016-09-14 for -06)
No email
send info
I support Stephen's comment, and have a few minor comments of my own (none of which are showstoppers):

-1, first paragraph: I'm not sure we should describe an experimental RFC as "standardized".

-1, mention of iOS7: Is this really limited to iOS7 and not future versions? Would it make since to say that "Since September 2013...is also supported ... iOS"? (i.e. without the version?)   (Noting that iOS10 released this week...)

- 2.2, paragraph 9: Figure 1 just shows a generic 2 path connection. It doesn't seem to "summarize" the described scenario.

-2.2, 2nd paragraph after figure 1: I'm pretty sure there are in fact real applications that transfer bulk data. Do you mean to say that "Some [or even many or most] real applications do not..."

(Benoît Claise) No Objection

Comment (2016-09-14 for -06)
No email
send info
Thanks for this document.

- This document contains a couple of possible improvements.
I believe this important aspect of the draft should also be mentioned in the abstract.

- Just a thought (no need to reply):
On one side, section 3.1 speaks of "middlebox interference". 
On the other side, this document via [I-D.lhwxz-hybrid-access-network-architecture] proposes just that: a new middlebox function.
This new middlebox might generate some interference for others.  Sarcasm warning: I guess that middleboxes are like children, we can only stand ours. 
I wanted to make that point, in light of all the middlebox discussions these days, for example in the QUIC charter.

- Another thought (again no need to reply) 
   "Since September
   2013, Multipath TCP is also supported on smartphones and tablets
   running iOS7 [IOS7].  There are likely hundreds of millions of
   Multipath TCP enabled devices.  However, this particular Multipath
   TCP implementation is currently only used to support a single
   application.  Unfortunately, there is no public information about the
   lessons learned from this large scale deployment."

Yes, this is really unfortunate.

Below is Qin Wu's OPS directorate review:
I think this document is well written and ready for publication. Here are a few editorial comments:

1.       Section 1, paragraph 2 mentions that three implementation are open sources. Which three of them are open sources? I think it includes apple MP-TCP, but it is not clear in the text.

2.       Section 2.2 paragraph 7 points out using REMOVE_ADDR option may cause operational problem, but I don’t see any discussion on this in the operation experience section, is this an open issue that needs to be addressed in the future or other document?

3.       Section 3.1 talks about the v0.87 Multipath TCP implementation What does V0.87 stands for? Version number? is there any reference to it?

4.       Section 3.2

s/the the default congestion control/ the default congestion control

5.       Section 3.2 last paragraph said that Reports from some users indicate that they seem to favor OLIA. It looks this statement is groundless statement.

6.       Section 3.9

s/ returned to the DNS query/return in response to the DNS query

7.       Section 3.10 said:

“

A better approach would probably be to try a few attempts on

the WiFi interface and then try to use the second interface for the

initial subflow as well.

”

When trying to use second interface for initial subflow? A few attempts on the WIFI interface fails?

8.       Section 3.2

s/accomodate/accommodate

(Stephen Farrell) No Objection

Comment (2016-09-14 for -06)
No email
send info
I was a bit sad that there was no reporting of
experiences with the security aspects of MPTCP.  Have
we really learned nothing worth saying about that?
Have we really seen no attacks on, or tailored to,
MPTCP? It seems odd that the answer to both questions
is "no."

(Joel Jaeggli) No Objection

Comment (2016-09-13 for -06)
No email
send info
Qin Wu <bill.wu@huawei.com> performed the opsdir review. 

he had quite a sfew small editorial comments that may be of use to the authors.

Suresh Krishnan No Objection

(Terry Manderson) No Objection

Alexey Melnikov No Objection

(Kathleen Moriarty) No Objection

Comment (2016-09-14 for -06)
No email
send info
I agree with Stephen's comments.

Alvaro Retana No Objection