Re: Naive question on multiple TCP/IP channels and please dont start a uS NN debate here unless you really want to.

[David Morris <[e-mail]>]

On Fri, 6 Feb 2015, Phillip Hallam-Baker wrote:

> One part is that we need more than a technical fix for this problem. We need to be
> able to tell Joe or Jane Consumer how often these slowdowns occur and what the cause
> is. The problem being that the cause of the problem might be on the broadband provider
> side or the home user side of the network. 

Not a technical fix, but information gathering would help. I run a small
network behind an SDSL connection. A while ago I noticed a performance
issue with my internet connection. When I called the ISP, they came back
and said my link was saturated with traffic.

It would have been really helpful if they'd given me a simple
characterization of the traffic in terms of protocol. I had to gerrymander
the ability to trace the subnet between my SDSL modem and my firewall.
Learned it was the NTP error attack. Configured the firewall to block
unsolicited inbound NTP traffic or perhaps NTP traffic unrelated to the
servers I use.

There was a technical solution to this problem, but my point here is to
illustrate the difficulty of getting data to diagnose the problem.

Every home/small office router/modem should be able to provide
utilization information over time as raw load, by protocol, and ideally,
by inside IP generating the load. Every ISP should be able to provide
something similar on demand.

Having the ability for the ISP to configure traffic filters at their end
to block last mile traffic for common threats. The end customer needs
to control what is filtered. Most small networks don't run HTTP or SMTP
servers as an example. Even fewer run NTP servers.

Dave Morris