Skip to main content
Question

Backup analysis AWS S3 or Git

  • December 12, 2019
  • 1 reply
  • 0 views
A
  • Anonymous
Hello Community, I am playing around with the backup content function for Celonis analysis. https://python.celonis.cloud/docs/pycelonis/en/latest/notebooks/99_Use_Case_Version_control.html#Create-backups-of-all-analyses-that-are-published-in-separate-workspace-directories While investigating, I realised that you have to store your file locally, before you can push it to Git, or AWS S3. I was wondering if anyone tried to directly send the backup to S3, or Git?

1 reply

S
Level 0
  • simon.rieze
  • Level 0
  • December 13, 2019
Hi Paul, Currently the backup_content functions download directly to a Path object from Pythons built in pathlib. pathlib currently seems to only support filesystem paths, so downloading directly to s3 or a git hosting provider seems to be impossible for now. You could add functionality to your Jupyter notebook to directly upload each downloaded file to S3 though: https://docs.aws.amazon.com/code-samples/latest/catalog/python-s3-s3-python-example-upload-file.py.html In a use case where I am doing something similar I created a git repo directly in the ML workbench and at the end of my scheduled notebook the following commands run: !git add . !git commit -m "Update $(date +"%D")" !git push Would a similar solution would work for you as well? Best regards, Simon Riezebos
Terms & ConditionsAccessibility statement