Survey banner
The Dataiku Community is moving to a new home! We are temporary in read only mode: LEARN MORE

Folders

Solved!
lldiazr
Level 2
Folders

Hi, I want to save a model written in a python recipe in a Folder, however it generates an error that the Folder must be declared as output. How do I declare the Folder to be able to save pkl's there?

 

0 Kudos
1 Solution

Please always use code blocks to post error messages and code (the </> icon on the toolbar, see sample below) as they are much easier to read and copy. It's pretty clear from the error you get (see below) that the GCP service account you are using is not permissioned properly in your GCP environment. Contact whoever issued that service account and give them the error and they should be able to solve your problem.

 

Job failed: Error in python process: At line 505: <class 'Exception'>: Error: None: b"Failed to connect to GCS, caused by: 10Exception: Cannot find or access bucket sbx-analitica-bkt- x
1, caused by: GoogleCloudStorageException: sdx-conn-dIp@sbx-analitica.iam.gserviceaccount.com does not have storage.buckets.get access to the Google Cloud Storage bucket.
Permission 'storage.buckets.get' denied on resource (or it may not exist)."

 

 

View solution in original post

0 Kudos
4 Replies
Turribeach

You need to add the folder to the output of the Python recipe. This can be done at the recipe creation time or at the recipe Inputs/Outputs tab by selecting the Use Existing/New Folder options. The GUI isn't that good in this aspect as the options are less visible as text links below the Create Dataset button.

 

Screenshot 2024-06-09 at 08.24.43.png

0 Kudos
lldiazr
Level 2
Author

Thank you
That is indeed how I have it and that is how it looks ..... However when executing I am getting the following error and I am not sure if it is indeed a permissions issue or I am doing something wrong in the code. 

0 Kudos

Please always use code blocks to post error messages and code (the </> icon on the toolbar, see sample below) as they are much easier to read and copy. It's pretty clear from the error you get (see below) that the GCP service account you are using is not permissioned properly in your GCP environment. Contact whoever issued that service account and give them the error and they should be able to solve your problem.

 

Job failed: Error in python process: At line 505: <class 'Exception'>: Error: None: b"Failed to connect to GCS, caused by: 10Exception: Cannot find or access bucket sbx-analitica-bkt- x
1, caused by: GoogleCloudStorageException: sdx-conn-dIp@sbx-analitica.iam.gserviceaccount.com does not have storage.buckets.get access to the Google Cloud Storage bucket.
Permission 'storage.buckets.get' denied on resource (or it may not exist)."

 

 

0 Kudos
lldiazr
Level 2
Author

Thank you very much for your cooperation

0 Kudos