Morning guys! I often see jupyter hang eternally, what can I do with it? Might be related to occasionally it is not saving files. "Saving started" and kernel not initialized Additional question - how to run python file separately with environment I built? (not through step -> run) Can VS Code service reuse environment? (Just trying to figure out correct expected workflow working with orchest)

What version of JupyterLab do you see (top right Help -> About)?

Yep so I went as far as trying to setup this tool to use by small team and It's a bit complicated to come up with some nice choice. I worked in Jupyterlab (Version 3.1.18), but it is too glitchy at the moment. VS Code service is super responsive and works fine but there is no way to actually develop a pipeline there... Really a complicated choice. I ended up moving notebooks to python files for portability and as they tend to be less glitchy.

As far as I understand real cross-team usage would require setting up orchest locally for everyone, developing and pushing commits to the repo...

We're working on making this more seamless. Right now that would be preferred in order to not have everyone hog each other's compute resources. I'll get an release of Orchest out today that upgrades JupyterLab to the latest release. They move to 3.2.x and I hope they improved the stability. Will do some internal testing.

thanks for clarifications

We'll be launching the k8s version in a couple of months, this will make it a lot more feasible to run Orchest on a small cluster and have multiple people use one and the same Orchest instance where their compute will be fanned out over multiple 'worker' nodes.

@Rick Lamers also can we make environmental variables propagate more? I stumble third+ time upon this. Or they work only in ipynb and during pipeline run

Found an issue in JupyterLab 3.2.4 so holding off on releasing it for now: https://github.com/jupyterlab/jupyterlab/issues/11527 Will keep you posted.

Well, like I said before, probably I`m not a target audience. But mostly it was needed for some edge cases: 1. I tried to use them to make nodejs processing script to work in my environment 2. Now I tried to use them to setup git repository ssh rights in terminal for interactive editing (SSH_KEY env variable) I guess there are more cases where you would expect your project /pipeline level env variable be accessible

Are you using git from VS Code?

@Rick Lamers tried to setup there with env variables, then from jypiter, havent found an option to save my ssh key anywhere

@Rick Lamers ok, but this won't work with jupyterlab buttons as I checked, right?

It should work, the only issue is that it needs to trust the host at least once (which needs to happen through terminal, as the JupyterLab git extension doesn't support asking you)

okay, thanks