Re: [v6ops] Brian Haberman's Yes on draft-ietf-v6ops-pmtud-ecmp-problem-04: (with COMMENT)

[joel jaeggli <[e-mail]>]

On 10/14/15 8:20 AM, Brian Haberman wrote:
> Brian Haberman has entered the following ballot position for
> draft-ietf-v6ops-pmtud-ecmp-problem-04: Yes
> 
> When responding, please keep the subject line intact and reply to all
> email addresses included in the To and CC lines. (Feel free to cut this
> introductory paragraph, however.)
> 
> 
> Please refer to https://www.ietf.org/iesg/statement/discuss-criteria.html
> for more information about IESG DISCUSS and COMMENT positions.
> 
> 
> The document, along with other ballot positions, can be found here:
> https://datatracker.ietf.org/doc/draft-ietf-v6ops-pmtud-ecmp-problem/
> 
> 
> 
> ----------------------------------------------------------------------
> COMMENT:
> ----------------------------------------------------------------------
> 
> Given the current state of affairs, I think this is a quite useful
> document.  I just have a question for my own knowledge.
> 
> Is there any data that indicates how many of these PTB messages are due
> to tunneling?

I have some although I doubt that they are valid statistically for the
whole internet. by way of characterization the vast majority of sessions
for which pmtud messages were distributed at zynga had an mss of 1440 so
these are otherwise unsuspecting machines facing a constraint on an
intermediate hop.  the pool of ipv6 enabled words with friends users was
actually pretty high, e.g. tmobile verizon and other lte phones with
ipv6 or dual stack pdp contexts, and those would be fine. if on ther
other hand the same user has a problem when attached to their home CPE
then they might have a tunnel.

older work in this space is stuff like

http://www.nlnetlabs.nl/downloads/publications/pmtu-black-holes-msc-thesis.pdf

as a content provider however I'm not that concerned with the total
population only that of my customers where it turns out this is rare,
albiet not rare enough to ignore.

> 
> 
> ______________________________________________
> v6ops mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/v6ops
>