hmm? how's that for properly elided. trust me they match. :)
Ah, I misunderstood - you were doing 'Dragnet' eliding to protect the
innocent?
the frustrating bit is that either one of these works on their own.
But if they are both invoked w.in the lifetime of a single JVM the
last-one-in isn't properly decrypted.
what I'm wondering now is if there is some issue with the
sequential keytab evaluation within a single kerberos domain.
It also sounds like you have identified roughly the cause. Something
is getting bound to the first configuration that doesn't coexist
nicely with a second configuration? Might be kerberos domain. Might
be ClassLoader? Could you configure a second ClassLoader separately?
Tomcat and/or Eclipse might do that for you which could explain
different results seen from them?
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Java-dev mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/java-dev/email@hidden
