Scoped attributes are being filtered

Richard Frovarp richard.frovarp at ndsu.edu
Thu Dec 28 12:30:30 EST 2017


I figured it out. Turns out I was bringing in two copies of the metadata 
for the same entity id, the one I was looking at had the correct scope 
asserted, the one I didn't realize was included further down in 
shibboleth2.xml had the wrong scope.

Thanks for the help.

On 12/28/2017 11:18 AM, Boyd, Todd M. wrote:
> To help with troubleshooting, you could try and disable encryption for that SP in relying-party.xml.
>
> Here's an example of an override we have on one of our SPs:
>
>      <util:list id="shibboleth.RelyingPartyOverrides">
> 		<bean parent="RelyingPartyByName" c:relyingPartyIds="https://redacted.com">
> 			<property name="profileConfigurations">
> 				<list>
> 					<bean parent="SAML2.SSO" p:encryptAssertions="false" p:signAssertions="true" p:signResponses="false" />
> 				</list>
> 			</property>
> 		</bean>
>      </util:list>
>
>
>
>
> From: users <users-bounces at shibboleth.net> on behalf of Richard Frovarp <richard.frovarp at ndsu.edu>
> Sent: Thursday, December 28, 2017 10:52:29 AM
> To: Shib Users
> Subject: Scoped attributes are being filtered
>      
> I'm trying to upgrade our IdP, and to validate, I'm testing against a
> native Shibboleth SP.
>
> The SP is filtering the ePPN and scoped affiliation attributes out. I'm
> trying to figure out where my problem is. If I make it so that those two
> attributes aren't scope filtered, the SP reports back the correct values
> with the correct scope / suffix. This makes me think that the data
> leaving the IdP is correct.
>
> To me that indicates that I'm having trouble configuring the SP. The
> metadata for the IdP that the SP has the correct scope. However, that
> scope isn't using the same subdomain as the entity id or even the URL of
> the IdP. Not sure if that makes any difference or not. And of course
> everything is encrypted, so trying to look at the raw messages isn't the
> easiest.
>
> Any ideas as to which end is wrong?
>
> Thanks,
>
> Richard
>



More information about the users mailing list