Databricks kicking me out every 2 minutes

[sminamioka]

After the session expire every 2 minutes as show below, I can't log in afterwards as shown further below. 

What the heck is this? Then I have to wait the platform own will to allow me to log in many minutes later.

[CynAuad]

If it helps, what worked for me to avoid wasting so much time was setting up the notebook in VS Code. I only launch Databricks when I import the notebook and run it. In other words, I first connect to spin up the cluster, then I get logged out. I log back in to run it... it’s not real-time... but at least I’m not spending an hour on a 15-minute exercise.

[Ashwin_DSA · Accepted solution]

Hi @CynAuad @sminamioka @learnawscloud @Kowsi 

Thank you for your patience while we investigated this issue.

We flagged this to our teams internally, and the team has deployed a fix, and based on our checks, the behaviour should now be back to normal. This incident has affected a limited set of environments, which is why it was harder to reproduce internally and took longer to raise and resolve than we would have liked.

Please try again and let us know if you still see the problem. We are treating this as a priority, and any confirmation or remaining error details you can share will help us validate the fix and follow up quickly.

Thank you again for flagging this and for bearing with us.

If this answer resolves your question, could you mark it as “Accept as Solution”? That helps other users quickly find the correct fix.

[CynAuad]

I can definitely tell it's improved—it's been 30 minutes and it hasn't logged me out yet.
Thank you so much!

[Ashwin_DSA]

Hi @CynAuad,

Thanks for the quick response. Glad it’s working as expected. Let me know if you see this happen again. 

[Ashwin_DSA]

Hi @CynAuad,

As we continue our investigation to prevent this issue from recurring, could you please share the details of your login setup? Are you a user in Entra, and do you use Entra SSO for logging in?

If this answer resolves your question, could you mark it as “Accept as Solution”? That helps other users quickly find the correct fix.