[Bloat] What does cablelabs certification actually do?

[Bloat] What does cablelabs certification actually do?

[jb]

[-- Attachment #1: Type: text/plain, Size: 1348 bytes --]

http://www.cablelabs.com/specs/certification/

So a number of our users have ganged together to share information and
discovered that Puma6 chipset based cable modems all seem to have a very
bad flaw. Whether this flaw is fixable in firmware isn't known. The only
company initially participating in the discussion - Arris - has gone a bit
quiet.

We built a tiny tool to identify whether a cable modem has the issue:
http://www.dslreports.com/front/puma6.html

The distribution of latencies for a puma6 modem is terrible, stretching over
100ms and beyond often as high as 500ms. This appears to be some
kind of systemic packet loss or delay involving small packets and while it
doesn't influence speed over an established connection it screws up DNS
lookups and short TCP transactions. So its rather a strange issue but very
irritating for any owner of these new modems until it gets fixed.

The modem / chipset is increasingly widely used so we got the register to
warn about the issue, I think they are going to do a new article next week:
http://www.theregister.co.uk/2016/12/03/intel_puma_chipset_firmware_fix/

Anyway while looking at this data and the number of these Puma6 driven
modems out there now, I wondered how it got certified!

And then wondered why certification can't also include verification for
correctly sized buffers as well?

[-- Attachment #2: Type: text/html, Size: 1778 bytes --]

Re: [Bloat] What does cablelabs certification actually do?

[Mikael Abrahamsson]

On Fri, 9 Dec 2016, jb wrote:

> And then wondered why certification can't also include verification for 
> correctly sized buffers as well?

There is nothing stopping this, and it's being worked on (PIE goes into 
DOCSIS 3.1).

http://www.cablelabs.com/wp-content/uploads/2014/06/DOCSIS-AQM_May2014.pdf
https://www.nanog.org/sites/default/files/20160922_Klatsky_First_Steps_In_v1.pdf

Cable Labs (as far as I understand) is an organisation funded by cable 
operators and vendors, and they create standards and tests used by the 
cable industry.

I don't know what tests Cable Labs perform, but there is nothing stopping 
them from validating buffers+AQM in the modems as well, and I do hope they 
do this going forward.

Why not reach out to Greg White who is mentioned in the cablelabs 
DOCSIS-AQM pdf above and ask? Or even better, invite him to this list if 
he's not already here. He's on the IETF AQM WG list, I have posts by him 
in my folder back to 2013.

-- 
Mikael Abrahamsson    email: swmike@swm.pp.se

Re: [Bloat] What does cablelabs certification actually do?

[jb]

[-- Attachment #1: Type: text/plain, Size: 1766 bytes --]

So after writing, I discovered they were asked about the problem with the
Puma6 chipset and stated their job as they see it is mainly to test for
standards compliance etc, and they purposely steer clear of performance in
order to encourage competition and variety. Or some punt like that. (I'm
probably mangling the message bit but that was the gist of it).

I guess there are such a huge number of devices and firmware releases it is
an endless job to evaluate each for performance in depth (beyond top
speed). Perhaps they are like cars before Ralph Nader. Unsafe at any speed?
IoT devices would fall into that category for sure!

On Fri, Dec 9, 2016 at 5:39 PM, Mikael Abrahamsson <swmike@swm.pp.se> wrote:

> On Fri, 9 Dec 2016, jb wrote:
>
> And then wondered why certification can't also include verification for
>> correctly sized buffers as well?
>>
>
> There is nothing stopping this, and it's being worked on (PIE goes into
> DOCSIS 3.1).
>
> http://www.cablelabs.com/wp-content/uploads/2014/06/DOCSIS-AQM_May2014.pdf
> https://www.nanog.org/sites/default/files/20160922_Klatsky_
> First_Steps_In_v1.pdf
>
> Cable Labs (as far as I understand) is an organisation funded by cable
> operators and vendors, and they create standards and tests used by the
> cable industry.
>
> I don't know what tests Cable Labs perform, but there is nothing stopping
> them from validating buffers+AQM in the modems as well, and I do hope they
> do this going forward.
>
> Why not reach out to Greg White who is mentioned in the cablelabs
> DOCSIS-AQM pdf above and ask? Or even better, invite him to this list if
> he's not already here. He's on the IETF AQM WG list, I have posts by him in
> my folder back to 2013.
>
> --
> Mikael Abrahamsson    email: swmike@swm.pp.se
>

[-- Attachment #2: Type: text/html, Size: 2661 bytes --]