Permission denied on schema

jlbellier Registered Posts: 22

Hello everybody,

I am working on a flow to update tables on 2 schemas Sch1 and Sch2 in a PostgreSQL database. The update is done through SQL Script recipes. My version of Dataiku is 9.0.1.

For the first Schema, everything works well.

For Sch2, I get the following error in the log : Permission denied on Sch2.

In Dataiku, several Database connections have been set up, but I do not have the hand on it so I cannot see how they are configured).

In my PostGreSQL database where I want to write the data, I can see groups but I do not see the content.

I there any information in Dataiku's logs to see which user is used for the SQL recipes, and also for query executions (in my case script SQL recipes) ? It is my personal user that is used ?

I am asking these questions to understand how the connections are managed and, as a consequence, which team I need to contact to fix this issue.

Thank you in advance for your help.

Have a good day.

Best regards,


Operating system used: Windows


  • Turribeach
    Turribeach Dataiku DSS Core Designer, Neuron, Dataiku DSS Adv Designer, Registered, Neuron 2023 Posts: 1,737 Neuron

    Dataiku connections use a specific user account so if you need to move data between schemas you will need to have that account granted permissions to write on both schemas. You will need to speak with your Dataiku Administrator to get the user ID used in the Dataiku connection and then speak with your Database Administrator to have it grant the correct permissions. Alternatively you could get around this issue by using two connections, Conn1 to write in Sch1 and Conn2 to write in Sch2 which use different user IDs. However this setup will be inneficient since Dataiku will not allow you to push the data between these two connections so you will need to sync the data to an intermediate dataset in a local file system or a bucket.

  • jlbellier
    jlbellier Registered Posts: 22

    Hello Turribeach,

    Thank you for your response. In the meanwhile, I found the problem. In fact, the input dataset of the SQL recipe on Sch2 was created on Sch1, so the recipe needed to execute a "cross-schema query".
    After recreating the input data set on Sch2, it worked fine.

    Best regards,


Setup Info
      Help me…