On Feb 23, 2011, at 11:11 AM, Tim Bruijnzeels wrote:
> On Feb 23, 2011, at 10:57 AM, George Michaelson wrote:
>> Can I ask relying party tool writers to consider this in their tools? 
>> Warnings are useful, but over-aggressive pruning or de-validation would not 
>> be good.
> 
> Yes. Our current implementation is indeed over-aggressive but it's on our 
> list to sedate it.

Having said that... yes we will sedate it, in particular with regards to 
*stale* manifests and CRLs which now result in 'invalid'. Missing CRLs I am not 
sure of yet, but it can be considered.

Missing manifests are problematic. Manifests are required by the res-cert draft:
http://tools.ietf.org/html/draft-ietf-sidr-res-certs-21#section-4.9.8.1

In our implementation we use the manifests to do top-down validation. I.e. for 
a TA we look in the manifest to find all published objects and child CA certs. 
The child CA certs have manifests listing their stuff and so on.. This is the 
easiest way do a recursive top-down validation. (Note that we also pre-fetch 
the directory for efficiency, but we only process the items listed on the 
manifest) Since the res-cert draft requires manifests I think this should be 
okay. If this assumption is wrong, please speak up.. I think, but don't know 
for sure, that other RP tools are using a similar approach.


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

Reply via email to