> Curious that Maven is seeing these failures while the Ant build isn't.
I think it does. Or it did. It's been eerily quiet -- I agree -- but I haven't been in sync with the latest changes so I don't know what you guys did to shut jenkins up a bit. > I was just looking today at randomizedtesting to see if there was a built-in > way to ignore thread leakage (there isn't AFAICT). You mean ignore them from the outside, right? No, there isn't a way to do this. I mean -- if a test is configured to detect these then why would you want to disable them externally. You can do it from the source code level of course, but that's not solving the maven-only build problem. I don't have many clues to offer here other than just ignoring either those NIO-stuck threads. I've tried installing FreeBSD to experiment locally but it was such a pain that I gave up after my time dedicated for this simply vanished and I was still at the command-prompt level trying to configure things. Not that I wasn't persistent enough, I just can't spend a week on it. :( Dawid --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
