My Jupyter notebook keeps disconnecting and/or the kernel is dying.


Why?

You'll run into disconnect issues when the Jupyter instance on your computer can't establish a connection to the live instance living on our server. 

This connection to a "live" instance is different than that of our other missions, which run inside of a virtual machine. A "live" connection means we can give you an authentic Jupyter experience, but it also means that the connection is more susceptible to interruptions, like opening up a new Dataquest tab or trying to work on a project when traffic on your network is high.

This happens most frequently when your network has dips in connection quality (e.g multiple connections to the network other than yours) — things that aren't severe enough for you to notice when you're browsing the web (so your webpages will load normally), but significant enough to sever the connection you've established to the Jupyter instance on our server.

⚠️In areas where the internet connection is regulated, your ISP can block or filter these types of connections, making it seem like you can't connect even though other websites are working properly. The internet at your work and at other public places have the possibility of throttling your connection as well, which can lead to frequent connects/disconnects from the server.

How can I fix it?

Outside of moving to a different network, you can try resetting the containers to work around this issue:

1
Make sure you only have one Dataquest tab open.
2
Click the Dashboard option from the dropdown menu in the top-right corner of your page.
3
Open a different mission and run a simple command, eg print("hello").
4
Click back to your Dashboard.
5
Navigate back to the mission you were working on and give your notebook time to initialize.