Re: [v6ops] ULA draft revision #2 Regarding isolated networks

[Brian E Carpenter <[e-mail]>]

On 27/05/2014 00:36, Liubing (Leo) wrote:
> Hi, All
> 
> We're going to update the ULA draft. Before making a new version, I think it would be helpful to confirm/discuss several important topics which were discussed in last IETF meeting. 
> 
> I'd like to discuss the topics in different mail threads respectively.
> (Current Draft link: http://tools.ietf.org/html/draft-ietf-v6ops-ula-usage-recommendations-02)
> ******************************************************************************
> 
> #2 Regarding isolated networks
> 
> Current draft is a little blurry on what the "isolated" means. Based on former discussion, we'll make it more comprehensive description as the following (just a summary, not the specific revision wording):
> 
> - "Temporarily isolated" or "Forever isolated". In general, ULAs fit both cases. Whatever it is temporarily or forever, when administrators need some prefixes to be on-demand and free to use, ULAs are good choice. However, for the temporarily isolated cases, the administrator needs to consider once it gets to connected, the hosts might need to be renumbered; or NAT might be involved if renumbering is not acceptable. If renumbering or NAT for some reason is considered as heavy burden, then the administrators need to carefully consider the adoption of ULAs.

Such networks (which in the IPng design phase we called "dentist's office
networks") are in the class where fully autonomic operation, with
no expert manager, is the normal case. As we observed in 6man, these
networks get automatically renumbered after every power cut. So we
should simply state that renumbering will happen in such networks
whenever necessary. I don't think we should even mention NPTv6.

The two cases are different though. If a network is running with
an ISP prefix and becomes isolated, it is not *forced* to use a
ULA prefix, but it might need to renumber when it gets reconnected
with a different ISP prefix. If it's forever isolated, a ULA would
be appropriate (and fail-safe if it does, in reality, get
connected to an ISP).

> - "Isolated to all networks" or "Isolated to the public Internet". These are two separate scenarios. However, in the perspective of adopting ULAs, there is no essential difference between them. So long as it doesn't connect to the global Internet, ULAs fit them as well. Comparing to other alternatives (an arbitrary GUA, or documentation prefixes), ULAs could provide a lower possibility of collision if they are generated according to the standard method. And the ACL rules for ULAs are much convenient to be set than arbitrary prefixes, to prevent the prefixes leaked if the isolated networks occasionally connected to the global networks. 

Yes. It seems to me that in "Internet of Things" scenarios, this
would be the normal mode. A group of IoT devices will be connected
to some sort of application-layer gateway, and by using ULA for
that, isolation from other networks is easier to ensure.

   Brian