Jupyterlab doesn't delete correctly

Hey jupyter community,

This is my first post to this forum so please excuse my mistakes :slight_smile: .

Goal

In our company, we are using jupyterhub running on a Fedora linux server as a centralized solution for coding and device control (The server is located in our lab and can talk to a lot of our devices). The goal is to have a stable solution that many people in the company can work with and share their solutions. There a lot of VScode fans in our company and I would love to convince them of the jupyter solution (I am a big fan!).

Problem

At the moment, we encounter a big problem that the codes are not saved correctly from the editor which makes programming close to impossible. Whenever we modify code of a .py file in the editor and then save, the code that actually gets saved is different from what the editor shows. When you reload the jupyterlab page, the editor refreshes and you see that the content is different from what was shown before.

Setup

Our setup is similar to what is discussed here: Install JupyterHub - JupyterHub-Deployment-ENGR101-2019Q1 . We also setup a collaborative server like discussed here: Real-time collaboration without impersonation — JupyterHub documentation . Most of our code is shared on a NAS that is mounted using nfs.

Does this setup make sense like this? What do you think is the biggest issue? Working on the nfs system or the collaborative jupyter?

Thanks in advance for your help!

How different? Can you provide a example? Does this happens to other file extensions, for example .txt?

1 Like

Thanks for your quick reply. Actually, I realized that this is a common issue when the files are linked with nfs and can also be accessed by other software, like other jupyter notebook instances at the same time. I changed the setup now that these two issues are solved. So have files locally and only have a collaborative server and not a personal at the same time.

Does seem reasonable to you?