Re: logging-log4j2 git commit: Test independence.

I thought so too... I think the problem has to do with plugging in and out of SLF4J. Gary On Wed, Dec 3, 2014 at 9:42 AM, Ralph Goers wrote: > Yes, but if MarkerTest is calling MarkerManager.clear() in the @Before > method then all the Markers should be gone. > > Ralph > > On Dec 3, 2014, at 5:

Re: logging-log4j2 git commit: Test independence.

Yes, but if MarkerTest is calling MarkerManager.clear() in the @Before method then all the Markers should be gone. Ralph > On Dec 3, 2014, at 5:47 AM, Gary Gregory wrote: > > On Tue, Dec 2, 2014 at 11:45 PM, Ralph Goers > wrote: > So clearMarkers is called b

Re: logging-log4j2 git commit: Test independence.

On Tue, Dec 2, 2014 at 11:45 PM, Ralph Goers wrote: > So clearMarkers is called both before and after? Then I don’t understand > why the failure occurred in MarkerTest. > IIRC, the problem was that LoggerTest that ran before MarkerTest left the system in a state that caused MarkerTest to fail b

Re: logging-log4j2 git commit: Test independence.

So clearMarkers is called both before and after? Then I don’t understand why the failure occurred in MarkerTest. Ralph > On Dec 1, 2014, at 1:38 PM, Gary Gregory wrote: > > Note that MarkerTest already had: > > @Before > @After > public void clearMarkers() { > MarkerManag

Re: logging-log4j2 git commit: Test independence.

Note that MarkerTest already had: @Before @After public void clearMarkers() { MarkerManager.clear(); } Which was not enough. The next question is: Should org.apache.logging.log4j.junit.InitialLoggerContext have additional clean ups like MarkerManager.clear() and MDC clea

Re: logging-log4j2 git commit: Test independence.

I suppose that is one way to do it. Perhaps a better way would be to call MarkerManager.clear() at the beginning and end of both LoggerTest and MarkerTest. Ralph > On Nov 30, 2014, at 10:57 PM, [email protected] wrote: > > Repository: logging-log4j2 > Updated Branches: > refs/heads/master