kubeflow / kubeflow/spark-operator
[QUESTION] How to disable spark.properties generation?
- Dominant language
- Python
- Stars
- 3.2k
- Forks
- 1.5k
- Avg merge
- 5d 10h
- Merged PRs (30d)
- 13
Description
I can't seem to find a way to disable the built-in /opt/spark/conf/spark.properties. Generating config via configmaps is not really a viable option for me since I'm trying to build a single spark image that can be used in multiple scenarios (kubeflow being one of them).
Kubeflow seems to stomp all over the /opt/spark/conf path. So I've built a container that keeps config in /opt/spark/config instead and set the appropriate env variable (SPARK_CONF_DIR). But that seems to be ignored because "--properties-file /opt/spark/conf/spark.properties" is injected during job submission. Is there a way to override or remove that option?
Thanks.
Contributor guide
Research direction
The issue does not name a source file or test. Start by tracing Spark job submission where `--properties-file /opt/spark/conf/spark.properties` is injected, then examine how `SPARK_CONF_DIR` is handled. Done means users can prevent or override the forced properties-file path while keeping configuration under `/opt/spark/config`.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- kubernetes, python
- Domain
- devops, infrastructure
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100