apache pig - Hadoop Load and Store -
when trying run pig script has 2 "store" same file way
store alert_message_count 'out'; store warning_message_count 'out'; it hangs, mean not proceed after showing 50% done.
is wrong? cant store both results in same file(folder)?
normally hadoop mapreduce won't allow save job output folder exists, guess isn't possible either (seeing pig translates commands series of m/r steps) - expect form of error message rather hang.
if open cluster job tracker, , @ logs task, log yield of note can diagnose further?
might worth checking pig mailing lists (if haven't already)
if want append 1 dataset another, use union keyword:
grunt> all_count = union alert_message_count, warning_message_count; grunt> store all_count 'out';
Comments
Post a Comment