Re: Usable extension headers [Re: New Version Notification for draft-voyer-6man-extension-header-insertion-08.txt]

[Ole Troan <[e-mail]>]

>>> We also tested against Alexa Top N targets for www, dns and mx, where perhaps Andrew’s tests were between end systems that were more “cooperative” across intermediate open / transit networks, so his values would be more akin to the “best case” numbers.
>>> 
>>> Certainly not a bad time to re-run tests, and maybe run tests between co-operative end points/domains.  Nearly 4 years have passed since the RFC7872 tests were run.  How much have things changed?
>> 
>> RFC7872 was (IMNSHO) based on the flawed belief that any end-host/service end-point should accept arbitrary extension headers.
>> The document offers little with regards to the deployment of extension headers within a limited _network_ domain.
> 
> No, but it makes it clear the tests are run across the Internet and transit ASes, and has some methodology to try to detect where the drops are happening.   Readers can draw their own conclusions.  The measurements showed clear issues in transit networks.  Specific / limited domains can of course to some extent do as they please; the issue is if such behaviour leaks.

The way the numbers are represented in that document, makes it way too easy for a casual reader to draw the wrong conclusions.
Which also affects how we in the IETF perceive what we can do with EHs.

>> I'm all in support for continuing monitoring and testing. Testing the "right thing"(tm) and understanding the results is a discipline on its own. If there is any learning that I can take from 7872, is that the RFC format isn't the right channel for measurement results.
> 
> Well, there was WG consensus both to adopt and publish it.  If someone conducts new tests, and perhaps takes your comments more on board, then I’d still support them being published via v6ops - it’s v6 operations after all.

Indeed, somewhat flippantly; there's very little there isn't WG consensus for in v6ops.

Cheers,
Ole