> On 19 Dec 2016, at 13:48, Juergen Schoenwaelder 
> <[email protected]> wrote:
> 
> On Mon, Dec 19, 2016 at 11:37:57AM +0100, Ladislav Lhotka wrote:
>> 
>> I am not proposing to ban candidate or something but IMO it needn't be part 
>> of the base NETCONF (or whatever protocol) spec.
>> 
> 
> Lets recall that candidate is a _capability_. Nobody is required to
> implement it.

Yes, this is what I wrote. My point is that this capability could/should be 
moved out of the NETCONF spec to a separate document. The latter can then be 
modified without affecting the former. 

> 
>> A typical problem of candidate combined with NACM is that user A edits item 
>> X and B edits Y in candidate. If B doesn't have write access to X and A to 
>> Y, then none of them is able to make a commit.
>> 
> 
> The problem is caused by allowing users with inconsistent access
> rights to both use candidate. So you get what you asked for. But I

Well, as soon as you let clients manage their private data (user accounts, 
routing instances etc.) you get these "inconsistent" access rights 
automatically.

> assume you can still do <discard-changes> and recover.

..., or find somebody with superuser privileges to perform the commit, but it 
is clumsy either way. Automated solutions are likely to break.

Lada

> 
> /js
> 
> -- 
> Juergen Schoenwaelder           Jacobs University Bremen gGmbH
> Phone: +49 421 200 3587         Campus Ring 1 | 28759 Bremen | Germany
> Fax:   +49 421 200 3103         <http://www.jacobs-university.de/>

--
Ladislav Lhotka, CZ.NIC Labs
PGP Key ID: E74E8C0C




_______________________________________________
netmod mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/netmod

Reply via email to