Anyone successfully run an impala script?

[#1]

Hi,

I would like to run a script (~200 lines) but have problems.

If I manually run each statement in the script, it is mostly fine. But if I use -f option to run this script in command line, it fails after 5~10 statement and shows the following error message:

       ERROR: MetaException: javax.jdo.JDOObjectNotFoundException: No such database row
       FailedObject:6[OID]org.apache.hadoop.hive.metastore.model.MDatabase
       NestedThrowables:
       org.datanucleus.exceptions.NucleusObjectNotFoundException: No such database row

Sometimes, this problem occurs in impala shell. But I can easily solve this problem by running the same statement again. However, I have no idea how to solve this problem in script mode.

Based on the above observation, I think I can mitigate this problem by  leaving some buffer time between statement executions. Or Maybe I can change configurations?

I am using Impala 1.0.1. (Actually in this VM: http://4zoas.com/cloudera-ccp-data-science-vm1-cdh4.3f-vmware.zip)

Any ideas would be greatly appreciated.

[#2 · Accepted solution]

Finally, I found a workaroud.

I wrote a python script which reads all statements from the script and slowly feeds each statement into "impala-shell -q".

[#3]

This certainly seems like it's worth filing a bug. Could you submit a bug with a test case at issues.cloudera.org?