Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Question - Cluster execution on a unique cell in a .py file #1165

Open
KiDayz opened this issue Apr 3, 2024 · 9 comments
Open

Question - Cluster execution on a unique cell in a .py file #1165

KiDayz opened this issue Apr 3, 2024 · 9 comments

Comments

@KiDayz
Copy link

KiDayz commented Apr 3, 2024

Hey,
I'm using the databricks extension in its pre-release version (v2.0.1)

I'm trying to execute a cell in a .py file (not the all file) with a cluster on Databricks (so no local kernel).

It seems that I have correclty setup the connection, my cluster is enabled & I see the 'Databricks Connect enabled' in VSCode. I thought, reading kartikgupta-db comment on this thread : #472, that I could run indiviual cell using my cluster but it seems that the only options available are the 'upload & Run file' & 'Run file as workflow'.

Am I doing something wrong here ? Or is the option not available yet ?
Many thanks !

@kartikgupta-db
Copy link
Contributor

If you have Databricks Connect enabled, then the notebooks should already be sending all the spark and dbutil commands to a Databricks cluster. We do not support executing full cell on Databricks yet.

@KiDayz
Copy link
Author

KiDayz commented Apr 3, 2024

Hey,
Thanks for your response. Do you have by any chance a roadmap or an ETA for the support of Python code execution by cell on a databricks cluster ?

@pernilak
Copy link

I want this feature as well. E.g., when using native python to read files from UC Volums, it failes when running in VSCode as it runs locally, while it works in Databricks Workspace as it runs on the cluster. We have other examples as well - leading to our developers wanting to work in Databricks as apposed to working from VSCode.

@MrTeale
Copy link

MrTeale commented May 2, 2024

+1 on this. This is actually one of the few blockers I have to developing solely in VSCode rather than the UI.

@kartikgupta-db - If you have a rough understanding of what would need to change for this to be implemented and would accept a PR, I'd be willing to have a go. Just need some guidance on getting started

@antonlindahl-sb1u
Copy link

antonlindahl-sb1u commented May 7, 2024

Pluss 1 on this one!

We struggle with API calls to mlflow starts to track locally. Also when python runs locally it can't find the path to UC Volumes (of course).

Would be very nice to get this feature and a ETA as well!

@KiDayz
Copy link
Author

KiDayz commented Jul 1, 2024

Hello everyone !
Hope you all are well :)

Any news on an ETA on this feature from the Databricks dev team ?

Thanks

@kupalinka-lis
Copy link

plus one here to keep an eye on it!

@pernilak
Copy link

pernilak commented Sep 5, 2024

If you have Databricks Connect enabled, then the notebooks should already be sending all the spark and dbutil commands to a Databricks cluster. We do not support executing full cell on Databricks yet.

@kartikgupta-db any update?

@KiDayz
Copy link
Author

KiDayz commented Oct 1, 2024

Hey @kartikgupta-db ,
Any news on the development of this feature yet ?

Many thanks !

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

6 participants