[afripv6-discuss] What have you done for IPv6 lately, since the
1st of January, 2013?
Guy Antony Halse
G.halse at ru.ac.za
Wed Feb 20 16:51:30 SAST 2013
On Wed 2013-02-20 (17:04), Mukom Akong T. wrote:
> Nevertheless, turning on Web is a big win. Just curious did you or
> anyone with web issues investigate with using some kind of Proxy to
> circumvent the limitation of not wanting to enable native IPv6 on your
> web server/
As it happens, for various reasons, the actual web server isn't natively v6
capable. (The web server software is; the jail it runs in isn't.)
When we tried to turn on v6 for World IPv6 Launch Day, it was in fact
tunnelled (6tunnel, http://toxygen.net/6tunnel/, to be specific). However,
the concept is the same as proxying.
This had an interesting effect that I hadn't banked on -- it made external
IPv6 connections appear as internal v4 connections (the v4 address of the
proxy). See below for the implications of this.
> Also I'd be curious to know --- what are the specific aspects of a web
> server that will break simply because one enabled IPv6?
guy at www:/www/static/intranet$ cat .htaccess
Allow from 22.214.171.124/17
Deny from all
Which is a daft idea anyway, but not my doing. There are lots of examples
of this, and also a number of scripts that have dependencies on address space.
To fix this we need to do a complete audit of all content, find all the
dependencies on IPv4, and work out whether they can be fixed (or
alternatively, whether they don't serve a useful purpose).
In short, its much easier for me to turn on v6 on services that are
completely under my control than it is for services that have user-submitted
content. I suspect that is generally true elsewhere, which is why I said
web isn't necessarily a good indicator.
Manager: Systems, IT Division, Rhodes University, Grahamstown, South Africa
Email: G.Halse at ru.ac.za Web: http://mombe.org/ IRC: rm-rf at irc.atrum.org
*** ANSI Standard Disclaimer *** J.A.P.H
More information about the afripv6-discuss