Hi @drgenious  First, please test your script outside of Oozie. If it is working outside of Oozie, then it should work from Oozie as well. As for  error "No module named impala.dbapi" it could be that there is some version dependency issue with impyla and its related libraries refer ---> https://github.com/cloudera/impyla/issues/227   ... View more

Got to know from the dev team that they have modified the column definition. We ran the MSCK repair table and we are able to run the select distinct query.  Vertex errors may not relate to memory issues. Hope this helps the community.  ... View more

  The error message indicates that there is an inconsistency between the expected schema for the column 'db.table.parameter_11' and the actual schema found in the Parquet file 'hdfs:/path/table/1_data.0.parq'. The column type is expected to be a STRING, but the Parquet schema suggests that it is an optional int64 (integer) column. To resolve this issue, you'll need to investigate and potentially correct the schema mismatch. Here are some steps you can take: Verify the Expected Schema: Check the definition of the 'db.table.parameter_11' column in the Impala metadata or Hive metastore. Ensure that it is defined as a STRING type. Inspect the Parquet File Schema: You can use tools like parquet-tools to inspect the schema of the Parquet file directly. Run the following command in the terminal: bash   parquet-tools schema 1_data.0.parq Look for the 'db.table.parameter_11' column and check its data type in the Parquet schema. Compare Expected vs. Actual Schema: Compare the expected schema for 'db.table.parameter_11' with the actual schema found in the Parquet file. Identify any differences in data types. Investigate Data Inconsistencies: If there are data inconsistencies, investigate how they might have occurred. It's possible that there was a schema evolution or a mismatch during the data writing process. Resolve Schema Mismatch: Depending on your findings, you may need to correct the schema mismatch. This could involve updating the metadata in Impala or Hive to match the actual schema or adjusting the Parquet file schema. Update Impala Statistics: After resolving the schema mismatch, it's a good practice to update Impala statistics for the affected table. This can be done using the COMPUTE STATS command in Impala:   This step ensures that Impala has up-to-date statistics for query optimization. Here's a high-level example of what the Parquet schema inspection might look like: parquet-tools schema 1_data.0.parq Look for the 'db.table.parameter_11' column and check its data type in the Parquet schema. If the data type in the Parquet schema is incorrect, you may need to investigate how the data was written and whether there were any issues during that process. Correcting the schema mismatch and updating Impala statistics should help resolve the issue. ... View more

@drgenious This is an OS-level issue that will need to be addressed at the OS level by the system admin. The bottom line here is that thrift-0.9.2 needs to be uninstalled   There are various things that could be happening: 1) Multiple python versions. 2) Multiple pip versions. 3) Broken installation. Solution: 1 - You can try to create the Python virtual environment to connect to impala-shell virtualenv venv -p python2 cd venv source bin/activate (venv) impala-shell Solution : 2 (i) Remove easy-install.pth files available in, /usr/lib/python2.6/site-packages/ /usr/lib64/python2.6/site-packages/ (ii) Try running impala-shell If you found that the provided solution(s) assisted you with your query, please take a moment to login and click Accept as Solution below each response that helped. ... View more

@drgenious, Has the reply helped resolve your issue? If so, please mark the appropriate reply as the solution, as it will make it easier for others to find the answer in the future.   ... View more