https://issues.apache.org/bugzilla/show_bug.cgi?id=48746
--- Comment #12 from Stefan Bodewig <[email protected]> 2010-06-14 06:30:21 EDT --- There also is bug 46805 which somehow may also be related. Of the suggested changes over there we have performed one (join timeout) but neglected two that we may want to rethink: * unconditionally call process.destroy and don't close the streams at all (rely on process.destroy to do that). I'm not convinced this is going to help or even poses a problem. At the point where we close the streams either process.waitFor has succeeded or we have invoked process.destroy * first close the streams and then stop the threads At first glance I'd expect this change to make it more likely to miss output written by the process but at the same time it may force buffers to get flushed. I'll look into bringing real testcases for bug 5003 into Ant's trunk so we can test a few things. I don't think I ever checked in the ones I used by the end of 2008 and don't have them anymore, so I'll have to recreate them. -- Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug.
