I've Done A Really Bad Thing

[Liam Proven]

On 23 May 2013 01:38, Paul Smith <paul at mad-scientist.net> wrote:
> Contrary to Liam's comment re "no
> possible causal connection", there have been many setups over the years
> where DNS does not work properly over IPv6, which leads to a significant
> slowdown of the entire network, as name resolution tries to use IPv6 and
> eventually times out and falls back to IPv4 DNS servers.

OK, fair call. I have not met that or anything like it, but I will
take your word.

> Even today not
> all network providers have their act together WRT IPv6.

*Very* few have, IME. Working IPv6 WAN connectivity is rarer than
melon-sized hailstones.

>   Disabling IPv6
> was the recommended solution for these problems in many instances.

I have /never/ encountered that. I used to disable it as a matter of
course in the early part of the century, as no networks that any of my
or my clients' machines ever used supported it, and thus it was
needless bloat. These days, on gigabyte-class machines, I just ignore
it.

I have only used IPv6 in labs/training environments, never in live
systems. It's that rare.

I believe it's quite common in the world of mobile phone networks, but
it's effectively invisible and hidden from customer/user view there.

> By far the most common reason for a network to feel sluggish with no
> discernible cause is DNS resolver issues :-/.

I can certainly believe that!

--
Liam Proven • Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk • GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com • Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 • Cell: +44 7939-087884