Unsupported content This version of the documentation is no longer supported. However, the documentation is available for your convenience.You will not be able to leave comments.

Recovering from an ETL hang or failure

You can use the stop_etl.nsh script to recover from an extract, transform, and load (ETL) hang or failure, which can be indicated by any of the following circumstances:

  • You receive one of the following error messages:
    • SCENARIO EXCEPTION - Scenario is already running.
    • Another instance of the script is already running.
  • ETL stops responding (you see no activity for a long time).

After running the stop_etl.nsh script, try running the run_etl.nsh script again.

To stop an ETL run, use the following procedure. 

To stop an ETL run

  1. Log on to the reports server on which the ETL failure occurred.
  2. From a command line, start the Network Shell.
  3. Navigate to the shared/odi/bin directory.

  4. Run the stop_etl.nsh script as follows:

    ./stop_etl.nsh -s <siteID> 

    <siteID> is the site ID for which ETL needs to be stopped. This should be a valid site ID from the BL_SITE table.

    For example, to stop the ETL for a site with site ID 2, run the following script:

    ./stop_etl.nsh -s 2

     

Note

If you are running this script on a Microsoft Windows computer, you are prompted to terminate all ETL processes and and to confirm the termination. The script does not verify that the processes have been terminated.

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*