Per-Entity Metadata Working Group

[David Walker <>]

On 09/14/2016 10:45 AM, Cantor, Scott wrote:

On 9/14/16, 12:20 PM,   wrote:

   I still have a number of unanswered questions regarding the
   distribution of per-entity metadata:
   
   1) Do we need preview-main-fallback servers or is a single production
   server adequate?

Notwithstanding Nick's comment, certainly this could be accomodated in many ways, but I tend to believe it's useful to have this ability, at least preview + main anyway. Preview could perhaps only be available from a single source, perhaps, vs. fully redundant.

I agree with Scott.  To me, the main purpose of preview is to deploy broad changes to the metadata in a way that it can be tested by participants.  The concern is only partially testing of TIER-supported software, but also software that is not supported by TIER, or is configured in an unexpected way.  Perhaps even more important, it allows those tests to observe impacts of metadata changes beyond simple software bugs, for example, attribute release decisions based on some new metadata element that a test IdP has been reconfigured to use.

Attachment: signature.asc
Description: OpenPGP digital signature