Folks,
After PIG-219, our Hudson nightly builds are now running the same
junit tests in both local and mapreduce modes (using the test
target). This creates 2 sets of output for each test. Hudson can't
handle this properly. You will see 2 links for each test case but
they point to the same output.
Should Hudson just run junit's in mapreduce mode?
Or should Hudson run both modes but only collect output from one set
of tests?
Nige
On May 8, 2008, at 3:30 AM, Apache Hudson Server wrote:
See http://hudson.zones.apache.org/hudson/job/Pig-trunk/177/changes
