Metadata requireSignedMetadata=false Behavior

[Cantor, Scott]

On 6/8/15, 10:01 AM, "users on behalf of Nate Klingenstein" <users-bounces at shibboleth.net on behalf of ndk at internet2.edu> wrote:
>
>Since v2 remains a supported distribution for awhile, I hoped TestShib could support it too.

I would hope the opposite, very much so. It is supported for people running it, not people who need testshib.

>For the reason why I used the ID's, please see my other message.

Showing an example of what to change should not be totally invalidated because an ID changes. A simple note like "Note that the ID attribute can be anything and is used for logging purposes." seems like it would do the trick.

I'm harping on this because it's not as if we're shipping a new version any time soon, so the file is what it is right now.

>I'm sorry for misinterpreting how the trust engine works.

A trust engine can do anything is the point, and you could chain trust engines together with multiple keys if you really didn't care who signed what (which itself is a problem, so that's the main reason edugain really doesn't work like that).

-- Scott