Metadata validity interval and fun with timezones

[Cantor, Scott]

On 10/31/13, 5:27 PM, "Byte Flinger" <byteflinger at gmail.com> wrote:

>Server time (OS time) is the same on both. This was mot happening last
>time I tried so I suspect it has to do with the time change on Sunday.

If they're on track with each other, you're talking about comparing
something issues seconds ago. I don't see how a time change can affect
that comparison.

> I do wonder why the other setup which apart from Shibboleth attribute
>and relying-party configuration (and the fact that it's installed on the
>cloud while the non-working one is installed on a redhat vm) is working
>fine though.

That isn't possible, so you're overlooking some difference in the messages
by definition here. That assumes it's not the simple case of an SP that
doesn't implement skew and the servers are just a few seconds off. Which
is, as I said, a bug in the SP.

-- Scott