[DATALAB-2496] [GCP][DeepLEarning][Jupyter with TensorFlow] Apache Standalone Spark cluster creation fails

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Done
• Affects Version/s: v.2.5
• Fix Version/s: v.2.5
• Component/s: DataLab Main
• Security Level: Public (Regular Issues)
• Labels: Back-end Debian GCP
• Epic Link: Data engine
• Sprint: Apache DataLab release 2.5
• Assignee: Mykola Bodnar
• Reporter: Vira Vitanska
• Created: 05/Jul/21 15:42
• Updated: 12/Jul/21 11:57
• Resolved: 12/Jul/21 11:57

Description

The bug was found on GCP during Apache Standalone Spark cluster creation only for DeepLearning and Jupyter with TensorFlow

Preconditions:

1. DeepLerning/Jupyter with TensorFlow is in running status

Steps to reproduce:

1. Create Apache Standalone Spark cluster for DeepLerning/Jupyter with TensorFlow

Actual result:

1. Apache Standalone Spark cluster creation fails

Failed to generate variables dictionary. Exception:'gpu_type'.

Expected result:

1. Apache Standalone Spark cluster creation is successfully

 

---

 Jupyter with Tensorflow creation on GCP from autotest