On Fri, Dec 1, 2017 at 2:41 PM, Randy Presuhn < [email protected]> wrote:
> Hi - > > On 12/1/2017 3:37 AM, Balazs Lengyel wrote: > >> Hello, >> >> https://tools.ietf.org/html/rfc7950#section-7.21.2 >> >> o "deprecated" indicates an obsolete definition, but it permits >> new/continued implementation in order to foster interoperability >> with older/existing implementations. >> >> This means that a node that is deprecated MAY or MAY NOT be implemented. >> YANG is considered an interface contract, however "maybe implemented" is >> unusable in a contract. >> > > From a client perspective, access control can have similar consequences. > A contract only outlines some of the kinds of things that > might possibly exist. It doesn't tell you what's actually there or > whether you're allowed to do anything with them. > > >From a vendor perspective the issue Balazs is raising can affect support contracts. YANG status-stmt does change the contract for the specific object. The status deprecated is essentially the same as obsolete. There is no useful warning that an object can be removed from a server in the future. The warning says "the object can be removed right now". The deprecated status could have been defined to be the same as current, except the object will likely go away in the future. Existing implementations MUST support the object until it is obsolete. New implementations SHOULD NOT include the object. For YANG linkage issues, new revisions SHOULD NOT add new statements that use deprecated objects. Existing linkage would be reliable to use until the referenced definition is obsolete. Randy > > Andy > _______________________________________________ > netmod mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/netmod >
_______________________________________________ netmod mailing list [email protected] https://www.ietf.org/mailman/listinfo/netmod
