Submitting Feathr job to databricks ML 10.2 cluster might get NullPointer exception
Open
- Dominant language
- Scala
- Stars
- 1.9k
- Forks
- 247
- PR merge metrics
- No merged PRs in 30d
Description
repro step:
1. use a databricks ML 10.2 cluster (or other "ML" cluster)
2. run the databricks quick start notebook *inside* the cluster
3. the get_offline_feature() job will fail with a NULLPointerException,
Weird thing is -if the job is submitted to the current cluster; however submitting to a new cluster will not see this error.
Contributor guide
Assessment
This issue has not been assessed yet.