We are able to successfully install ArcGIS Pro 2.2 with the ‘pre-packed’ installer for enterprise. So, we move to test with Jupyter Notebook.
ArcGIS Pro 2.2 ships with the ArcGIS API for Python 1.4.1 installed. The initial installation of ArcGIS Pro 2.2 creates a default read-only conda environment named arcgispro-py3.
ESRI guides us with this way to upgrade ArcGIS API for Python and add new Python packages.
To upgrade the ArcGIS API for Python package, use the Python Package Manager to create a new environment, and then install the latest release of the ArcGIS API for Python, and then try to upgrade like:
conda upgrade -c esri arcgis
Or add new any conda Python package via Python Package Manager GUI.
But, both wouldn’t work in enterprise, because of the privilege and security firewall.
The question is:
Is there any other effective way for enterprise customers to create a new environment, install the latest release of the ArcGIS API for Python, and then add new Python packages? Saying, point to the local driver like RStudio.
Cloning your environment is currently your only option if you don't have admin privileges on your machine.
If you do have admin privileges, then conda installs of packages work without incident.
I have used the conda approach successfully to install newer versions of python, spyder, etc etc and the 1.5 api.
Support for the idea and related links can be found here
It is an issue under enterprise environment.
I guess, Esri ArcGIS Pro and relevant development teams are required to help clearify.
We may submit the remedy to tech support team for tracking ..