CODEWITHSUNDEEP
pythongoogle-cloud-platformjupyter-notebookgoogle-cloud-dataproc
Mariusz Kraj
Mariusz Kraj

Reputation: 107

Can't create a Python 3 notebook in jupyter notebook

I'm following this tutorial and I'm stuck when I want to create a new Jupyter Notebook (Python 3). The cluster is created using this command:

gcloud beta dataproc clusters create ${CLUSTER_NAME} \
 --region=${REGION} \
 --image-version=1.4 \
 --master-machine-type=n1-standard-4 \
 --worker-machine-type=n1-standard-4 \
 --bucket=${BUCKET_NAME} \
 --optional-components=ANACONDA,JUPYTER \
 --enable-component-gateway

When I accessing the JupyterLab and try to create a new notebook I can see: enter image description here and then enter image description here

console works fine

Thanks in advance!

Upvotes: 4

Views: 2217

Answers (2)

Omar Jarjur
Omar Jarjur

Reputation: 1066

This is a consequence of the new feature to support storing notebooks either in GCS or on the local disk.

To enable that, the top-level directory inside of Jupyter (and JupyterLab) was made into a virtual, read-only directory that just links to the GCS and local disk locations via virtual subdirectories. The intention is that you choose a storage location by navigating this virtual file hierarchy.

So, if you try to create a file at the top-level, you will get an error because that top-level is read-only, but if you navigate into one of the subdirectories, then things will work as expected.

The user experience around this clearly needs to be improved, and that error message is not helpful at all, so we are working on improving the experience here.

However, if you choose one of the two storage options (by navigating into one of the subdirectories), then things should work as expected.

Upvotes: 6

David Rabinowitz
David Rabinowitz

Reputation: 30448

Please navigate to GCS or Local Disk on the left panel as an intermediate solution, it should work afterwards.

Upvotes: 1

Related Questions