Calling LinkedIn, Amazon and Akamai @ DE-CIX NY

Wed Jan 30 23:55:40 UTC 2019

On 1/31/19 12:36 AM, Töma Gavrichenkov wrote:
> On Thu, Jan 30, 2019 at 23:10 AM Ren Provo <ren.provo at gmail.com> wrote:
>> You probably should remove sessions with networks
>> explicitly *not* participating in route servers versus
>> displaying them on a global shame list.
> And so it begins — yet another discussion on what does the word
> "responsibility" really mean.
>
> Given that e.g. the peering facility in Amazon, according to an
> adjacent NANOG ML thread, is in deep deep trouble since Nov 2018, just
> shutting down sessions with all of the entries in that shame list is
> likely to cause huge disruption and disappoinment.
>
> --
> Töma

What triggered that part of the discussion is a logical fallacy along 
the lines of: if A is true, then B is true. B is true, therefore A is true.

Here: all networks that didn't already change their peering IP are not 
yet connected to the updated route-server. Some networks are not 
connected to any route-server. Therefore, those networks did not yet 
change their peering IP.

I think you can see what's wrong with that statement.. it does not 
follow. That has nothing to do with peering department resources, but 
everything to do with the chosen peering strategy.

Best regards,
Martijn