Re: [RRG] Consensus check: renumbering

[Scott Brim <swb@employees.org>]

On 8/23/08 6:39 PM, Tony Li allegedly wrote:
> Hi Noel,
> 
> |    > I don't think that end users will care one whit about 
> |which namespace
> |    > they're renumbering. The work's the same. ;-)
> |
> |Well, no it's not, actually - which is one of the chief points 
> |of having more
> |than one namespace.
> 
> 
> Let me come at this from a different perspective: the truly painful event
> that is the large issue for folks like Eric is the prospect of having to
> reconfigure each and every host within the site.  This is work that scales
> linearly or super-linearly with the size of the site.

Well,

  - changing the locator on an individual device is easy.

  - changing locators for a whole site is hard and must not be depended
    on (one time seems tolerable but is still undesirable).

  - if there is an identifier, independent of topological location, that
    is used for authentication, then it is probably in databases all
    over the place and changing it will be hard even for an individual
    device.

> To the site administrators, whether they are changing IPv4 addresses, IPv6
> addresses, FQDN's or public/private key pairs, the pain is pretty much the
> same.  Touching everything *hurts*.

Certainly true for a whole site, because the problem is instantiation of
policy.  Changing addresses for individual devices is easy.

> That said, yes, for most schemes this event happens when you renumber the
> identifier namespace.  However, I really don't want to get hung up on which
> namespace it is that causes this level of pain.

Which meaning of identifier are you using?

Scott (starting to catch up)

--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg