institutions using Shibboleth to authenticate Peoplesoft

[Cantor, Scott]

On 7/28/20, 4:11 PM, "users on behalf of Liam Hoekenga" <users-bounces at shibboleth.net on behalf of liamr at umich.edu> wrote:

>    I believe that we're configured to store POST data in shared memory (the postData settings in the Session element)

RelayState also has to be factored in. If you're using memory for the post data, you can't be clustered, so you should be using memory for the relay state also, and then no cookies would be involved to recover anything.

Oracle is likely correct, your configuration is wrong somewhere. I thought they were blaming the SP for something happening on the SciQuest end.

In any case, testing with Firefox is much cleaner. Then it breaks every time in a consistent way.

-- Scott