[DATALAB-2398] [Jupyter][TensorFlow with Jupyter][Zeppelin]: Implement specific python versions usage via virtual envs

• Type: Improvement
• Status: Closed
• Priority: Major
• Resolution: Done
• Fix Version/s: v.2.5
• Component/s: DataLab Main
• Security Level: Public (Regular Issues)
• Labels: AWS AZURE Debian DevOps GCP RedHat pull-request-available
• Epic Link: Common features for all templates
• Sprint: Apache DataLab release 2.5
• Story Points: 40
• Assignee: Mykola Bodnar
• Reporter: Vira Vitanska
• Created: 13/May/21 09:34
• Updated: 01/Jul/21 11:48
• Resolved: 23/Jun/21 07:24

Description

The default list of python versions defined during the configuration which later can be selected by users. For now it is allowed to point the Python version 3.7.9.

To improve stability of python libraries installation during a notebook deployment, specific versions of python in insolated environment is required.

 Python version should be the same for local & remote Spark kernels.

Issue Links

• is related to: DATALAB-2414 [Rstudio][TensorFlow with RStudio][Spark]: Implement specific python versions usage via virtual envs
• links to: GitHub Pull Request #1175
• links to: GitHub Pull Request #1183
• links to: GitHub Pull Request #1188