[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [xacml] request's attribute assertion lifetime?
Daniel Engovatov wrote: >... >>Except that I can not incorporate the validity time check for the >>assertions that are used in the evaluation as their validity is only >>compared to the current time before the PDP gets its hand on it. > > > PDP evaluation would be valid against whatever dataset you provided. > You can validate this dataset prior, or after evaluation, or during > enforcement, or after enforcement, or when federal agents come knocking > on your door, or on Mondays, or never. It is a forever fixed > collection of values; one of them is "current-time". > Could you please show me how I could get the validity time interval of an > attribute assertion, like SAML, into the request context, tied to the > associated request's attribute? > (just simple time interval, no conditions, this is about the certificate > lifetime...) But my point was that I do not think it belongs there. Maybe we are not talking about the same thing... Hm... Whoever constructs the context does the validity-check, though one can certainly place whatever information is relevant for the policy into the context, if one wants to write explicit rules against this information. If this responsibility is distributed (such as in a case when a local routine computes the "current-time") this can be resolved by aggregating the used data for validation. For example, whatever protocol is used for PEP-to-PDP calls, it may return whatever local variables were computed during evaluation alongside the decision. Time may be only one of them. If I am to design such a system, I would either rely on a fact that the evaluation duration is inconsequential for the validity check (assuming it happens instantaneously, assuming you do not care about less then a millisecond, and the PEP and PDP use the same time source), or compute "current-time" in the same place were the validity check is performed and pass it along - I guess this is a perfectly compatible strategy. Daniel.
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]