A case of classpath roulette :-)
Would you mind filing a Jira issue for better validation/messaging?
https://issues.apache.org/jira/projects/LOG4J2
I've written similar validation for slf4j bindings elsewhere that we may be
able to leverage (though the general case has many edge cases that I have
Thanks for your answer but that wasn't the problem. In the end there was a
stray log4j 1.x JAR still in the application artifact, after removing that
everything worked fine. This wasn't clear from Log4J2's own debug logging,
according to that everything was fine. I don't know why this kind of cl
