Zafar Gilani created HIVE-4099: ---------------------------------- Summary: Execution Error, return code 2 from org.apache.hadoop.hive.ql.exec.MapRedTask Key: HIVE-4099 URL: https://issues.apache.org/jira/browse/HIVE-4099 Project: Hive Issue Type: Bug Affects Versions: 0.7.1 Environment: GNU/Linux x86_64, kernel 2.6.32-131.0.15.e16.x86_64, 16 cores, 48 GB main memory, 16 mappers, 8 reducers, mapred.java.child.opts set to 2g. Reporter: Zafar Gilani
Join query fails with "Execution Error, return code 2 from org.apache.hadoop.hive.ql.exec.MapRedTask". hive.log: ERROR exec.MapredLocalTask (SessionState.java:printError(365)) ERROR ql.Driver (SessionState.java:printError(365)) Select and insert queries work fine. Simplest of join fails. Data-set size: Two tables being joined, have 27k records each, each record having three fields. Already tried and failed: - Add contrib jar to the hive classpath - Set Hadoop mapred.child.java.opts to 2 to 8g of memory - Set Hive mapred.child.java.opts to 2 to 8g of memory - Set hive.auto.convert.join to true (regular join to mapjoin) - Set hive.optimize.skewjoin to true (handle skewness in data) - Set hive.mapjoin.maxsize to 1000000 (small table rows, both tables have <27k rows) -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira