Training job for `rl_distributed_robotschool` failed

Problem notebook:
```
https://github.com/aws/amazon-sagemaker-examples/blob/master/reinforcement_learning/rl_roboschool_ray/rl_roboschool_ray_distributed.ipynb
```

Training jobs for both primary cluster and secondary cluster failed

Start with reinforcement_learning/rl_roboschool_ray/rl_roboschool_ray_distributed.ipynb, the linked problem notebook, and inspect the training setup for the primary and secondary clusters. Reproduce the failure if possible and determine what prevents both training jobs from completing successfully.

Written by the indexing model from the issue text.