Re: [v6ops] draft-ietf-v6ops-icp-guidance WGLC

[Erik Nygren <[e-mail]>]

I took a pass through this draft and had enough suggestions that
it was easiest to just pass them along as a patch/diff against the
-02.xml source
(see attached).

In general this feels like it will be very useful as a starting points
for people
at Internet content and application providers looking to kick off an
IPv6 effort.
As discussed on some of the other threads, I think its important that we
keep this focused on recommendations that will be useful to people
implementing IPv6 today, rather than on technologies that are still
under development.

Some of the proposed changes include:

* Proposed some compromise text in the "IPv6 Infrastructure" section
  on the PI vs PA address space topic discussed in the other thread.
  It feels like we should be giving concrete recommendations that are
  useful to ICPs now, but while still leaving doors open for the
  future.

* Add a note that support operations and support staff will generally
  also need to have IPv6 connectivity provisioned.  (Outside of universities,
  external server and staff network connectivity and environments are
often independent.)

* Renamed the "Proxy" section to "Surrogate" as what is being described
  there is explicitly a reverse proxy (surrogate) setup.

* Added more context and details and gotchas to the "DNS" section.
  This seems to be one of the top areas where ICPs seem to get
  themselves in trouble (perhaps because it seems like it's
  a "just do it").

* Added some notes on security (such as for host and network device
  firewalls) at appropriate points in the doc as ICPs especially
  shouldn't be thinking of these as an afterthought.

* Added some more notes into the Application Layer section,
  such as some particular notes about logging and testing,
  as this seems to be another area where things are often missed.

* Split up the CDN section into a broader section on "Complex Sites
  and Applications".  About half of the existing CDN section was more generally
  applicable to ICPs with multiple POPs, so "Distributed Locations"
  got its own subsection here.  I also added a subsection on
  "embeded resources" reminding ICPs to check that a rich
  site has more than just the base page available over IPv6.
  (I also took out the reference to my employer
  as that quote was from a few years ago and may not
  be generally helpful.)

* Fixed some typos and made a few other edits, as well as
  some others that I may have missed in the summary above.

I hope this is useful to folks.  I'm happy to discuss any
of my proposed changes in more detail.

  Erik


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

All opinions expressed herein are my own and my not reflect those of
my employer.





On Mon, Aug 6, 2012 at 12:29 AM, Fred Baker (fred) <[e-mail]> wrote:
> This is to open a two week Working Group last Call on
>
> http://datatracker.ietf.org/doc/draft-ietf-v6ops-icp-guidance
>   "IPv6 Guidance for Internet Content and Application Service Providers",
>   Brian Carpenter, Sheng Jiang, 10-Jul-12
>
> Please read it now. We are interested in, among other things, technical commentary on the draft and the working group's perception on its usefulness to its target audience.
> _______________________________________________
> v6ops mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/v6ops