shibd.log is blank when testing IdP at sp.testshib.org

[Cantor, Scott]

On 6/16/15, 2:38 PM, "users on behalf of Khawer Masood" <users-bounces at shibboleth.net on behalf of shehzada at gwu.edu> wrote:

>I want to be respectful of people’s time, but should we not rely on the Test Shib any more … ?

If you're using testshib for longer than a week, no, that's not its purpose.

> What purpose does it serve if other than to validate installs before we role out vital services in production? 

Its purpose is to assist people who are new to Shibboleth or SAML to help them mock something up that works and resembles a real scenario so they can get their feet wet. It is most definitely not for validating installs of people already using Shibboleth.

>It was a key part of the training I attended recently, so I’m surprised to hear that we should outright forget about it. If we are not to rely on Test Shib, how do we validate our installs in the interim (other than our own SP)?

You're ruling out the correct option, so...

>I’m happy to wait until someone has the time to “fix” Test Shib — I understand there are more pressing matters. It would be helpful to know so I can report to my project manager whatever delay this adds to our project timeline.

That's just not the point of testshib, I don't know what else to say.

As a general matter, the consortium membership has not prioritized testshib becoming a more robust service or us investing time/money into it, so it remains running as is in a purely voluntary situation.

>PS: I’m happy to volunteer to help with the server upkeep of Test Shib if that is an option.

Kevin Foote and Nate are currently operating it as a volunteer thing, so if they're willing/interested, you could contact them. But as much as that's appreciated, you could apply a fraction of that time to just running an SP...

-- Scott