Bug 7099601 is not yet accessible. Anyway, I hope the Sun folks don't throw it out because it may lack credibility. Its reproduction requires logback+zookeeper. Plus, one needs to carefully read pre-fix version of AppenderAttachableImpl to be convinced that we are not hallucinating. Syzygy is the word that comes to my mind. I am keeping my fingers crossed. On 11.10.2011 17:10, Joern Huxhorn wrote:
I updated my project at https://github.com/huxi/lbcore224 and added a pom.xml I'm able to reproduce the issue on Mac OS X 10.6 with 1.6.0_u26 if I execute "mvn clean install". I'm also able to reproduce it running the test in IDEA. Last but not least, I submitted http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7099601 but it hasn't been processed yet.
Cheers, Joern.
On 04.10.2011, at 16:05, Ceki Gulcu wrote:
On 04.10.2011 14:46, Joern Huxhorn wrote:
Did the original test in your branch fail using Maven or did it also only fail from IDEA?
I was not sure if I could reproduce the IllegalMonitorStateException from Maven. I just resurrected the WithZookeperTest and ran it with Maven and was able to reproduce the IMSE. BTW, there was no logback configuration file, so the default config was used (ConsoleAppender).
Maybe IDEA and Maven launch the tests differently than Gradle... ? -- Ceki