Page tree

Release 8.2.1



Contents:


Outdated release! Latest docs are Release 8.7: Install Verify

   

After you have installed or made changes to the platform, you should verify end-to-end operations.

NOTE: The Trifacta® platform is not operational until it is connected to a supported backend datastore.

Steps:

  1. Login to the application as an administrator. See Login

  2. Through the Admin Settings page, run Tricheck, which performs tests on the Trifacta node and any connected cluster. See Admin Settings Page.

  3. In the application menu bar, click  LibraryClick Import Dataset. Select your backend datastore.
     

  4. Navigate your datastore directory structure to locate a small CSV or JSON file. 
     
  5. Select the file.  In the right panel, click Create and Transform.
    1. Troubleshooting: If the steps so far work, then you have read access to the datastore from the platform.   If not, please check permissions for the  Trifacta user  and its access to the appropriate directories. 
    2. See Import Data Page.
  6. In the Transformer page, some steps have already been added to your recipe, so you can run the job right away. Click  Run. 
    1. See  Transformer Page.

  7. In the Run Job Page: 
    1. For Running Environment, some of these options may not be available. Choose according to the running environment you wish to test.
      1. Photon: Runs job on the Photon running environment hosted on the Trifacta node. This method of job execution does not utilize any integrated cluster.
      2. Spark: Runs the job on Spark on the integrated cluster.

      3. Databricks: If the platform is integrated with an Azure Databricks cluster, you can test job execution on the cluster.

    2. Select CSV and JSON output. 
    3. Select the Profile Results checkbox. 
    4. Troubleshooting: At this point, you are able to initiate a job for execution on the selected running environment. Later, you can verify operations by running the same job on other available environments.
    5. See Run Job Page.
       
  8. When the job completes, you should see a success message in the Jobs tab of the Flow View page. 
    1. Troubleshooting: Either the Transform job or the Profiling job may break. To localize the problem, mouse over the Job listing in the Jobs page. Try re-running a job by deselecting the broken job type or running the job in a different environment. You can also download the log files to try to identify the problem. See Jobs Page.
       
  9. Click View Results in the Jobs page. In the Profile tab of the Job Details page, you can see a visual profile of the generated results. 
    1. See Job Details Page.
  10. In the Output Destinations tab, click the CSV and JSON links to download the results to your local desktop. See Import Data Page.
     
  11. Load these results into a local application to verify that the content looks ok.

This page has no comments.