[aklug] Re: IPv6 benefits me...

On Wed, 6 Jun 2012, Erinn Looney-Triggs wrote:

> As I said it is enabled on their server OS as well as the client. I
> would say working with MS systems is probably relevant to everyone,
> unless of course you never connect to an MS system. And it is of course
> relevant if any Linux distribution wants to enable the same feature by
> default in Linux (I don't know if any have).

Anyone interested in static addresses is highly likely to be interested in
setting up a server. The article wasn't clear as to how stateful the random
ID things is, and it explicitly states that the privacy settings (which
apply not to the random ID itself but to the temporary addresses used for
outbound connections) is disabled by default on the server.

So, again, it's not unreasonable to question the relevance. Most people on
this list interested in setting up a server are overwhelmingly likely to be
setting up Linux servers, not MS servers, and for the minority doing the
latter it still doesn't answer the question as to how long the random ID
lasts, the article you posted doesn't explicitly say. Correct me if I
missed it.

> Both, and more, it gets regenerated every couple of days as well, so no
> static.

The article doesn't explicitly say that, and the privacy extensions are
separate from the random ID. If you have references that outline that
explicitly, or if you've run your own experiments, please share.

>> Or, you could just run the one command necessary to put it into a normal
>> EUI-64 based address. So, regardless, you can still essentially have a
>> static IP on non-DHCP networks.

> Of course, but a default is a default, you were talking about how
> systems will have a static, I was saying no not for a large chunk of
> systems by default. You could also just assign a static regardless of
> router advertisements or DHCP, the choice is always yours, I was just
> pointing out that by default it wasn't as simple as it looked,
> especially in Microsoft's case.

I think you're splitting hairs. One single netsh command to disable random
IDs is all it takes. That makes it very trivial to get the essentially
static IP address we're talking about here. I appreciate you pointing out
MS's default behavior, but in this instance it doesn't appear to be
difficult at all to get the desired behavior.

You do, however, bring up one more good question for ISP network
administrators. Given the ease of finding available addresses in a subnet
(especially with the local administration bit) a DHCP network is still going
to have a hard time blocking people from assigning themselves statics. I
imagine they'd have to put a filter on the router to only allow the DHCP
pool range to be routed to prevent that, eh? I'd have to imagine, though,
that there's tools tying the filter list to acknowledged DHCP leases, since
it's an issue that affects both IPv4 & IPv6 networks.

         --Arthur Corliss
           Live Free or Die
---------
To unsubscribe, send email to <aklug-request@aklug.org>
with 'unsubscribe' in the message body.
Received on Wed Jun 6 11:34:29 2012