Re: [Netconf] Access control [was: ew draft-ietf-netconf-partial-lock-08.txt goes to AD

[Andy Bierman <andy at netconfcentral.com>]

WashamFan wrote:
> Hi,
....
> I review sec8.6.4.1, rfc4741, which tells me <source> parameter
> of <validate> operation might indicate configuration subtree
> rather than the full candidate. Although I am not sure if it is the
> intent.

Actually, the text in this section clearly says
that an empty element for the name of a database is used,
such as <candidate/> or <running/>.  Except the
complexType 'rpcOperationSourceType' in the XSD
does not match this text, because it includes
the 'inline config' option in the choice.

The NETCONF WG needs to decide (for 4741-bis) whether the
the text or the XSD is correct.

The intent of the WG at the time was not really clear,
except that the entire config was being validated
by this operation, not a piece of a config.

If an inline config element was used, then it
would need to be validated as if it were a complete
config database, so must/mandatory top-level nodes not
included in the inline config would still cause
a validation failure.


> washam

Andy