Problem
In Datameer 5.9.4, workbooks may fail with errors of the following nature in the job logs:
WARN [<timestamp>] [ConcurrentJobExecutor-0] (DefaultMrJobClient.java:216) - attempt_<id>: AttemptID:attempt_<id> Timed out after 600 secs
WARN [<timestamp>] [ConcurrentJobExecutor-4] (DefaultMrJobClient.java:211) - Task Id : attempt_<id>, Status : FAILED
WARN [<timestamp>] [ConcurrentJobExecutor-4] (DefaultMrJobClient.java:216) - attempt_<id>: Error: GC overhead limit exceeded
WARN [<timestamp>] [ConcurrentJobExecutor-4] (DefaultMrJobClient.java:216) - attempt_<id>: Container killed by the ApplicationMaster.
Container killed on request. Exit code is 143
Container exited with a non-zero exit code 143
Cause
This issue can be caused from a changed resource allocation either on the Datameer end or the cluster end.
Solution
In order to work around this issue, check to see if there are any intermediary, unkept sheets in the workbook. If found, check to keep all the intermediary sheets and re-run the workbook.
Alternatively, the Hadoop Custom Properties
das.join.disabled-strategies=MEMORY_BACKED_MAP_SIDE
can also be added to the workbook to work around the issue.
Should the issue persist after the above step is taken, please contact Datameer support for further troubleshooting.
Comments
0 comments
Please sign in to leave a comment.