Hi,
  My all hive jobs require "mongodb-java-driver, mongo-hadoop-core, 
mongo-hadoop-hive" jars to successfully execute. I don't have cluster access to 
copy these jars. So I use distributed cache (add jar <jar file>) for every job 
to make available these jars for M/R tasks. Because of this approach my user 
file cache (tmp folder) getting filled very quickly. So is there a way to copy 
these jars to distribute cache only once and reuse for all the jobs all the 
time?. So that I can avoid file cache getting filled.

Thanks,
Chandra

Reply via email to