kubeflow / spark-operator

Kubernetes operator for managing the lifecycle of Apache Spark applications on Kubernetes.
Apache License 2.0
2.8k stars 1.38k forks source link

Will Spark-submit command cache anything by default if download dependencies fail ? #2303

Open conmeotuxedo opened 3 weeks ago

conmeotuxedo commented 3 weeks ago

What question do you want to ask?

I mounted the volume into a Persistent Volume (retain reclaim policy) in EKS and first time I run it got error because of a wrong jar path download (jar path doesn't exist)

image

Later on I updated the correct path and run again ( same volume) it just started container then 2s later it show error status without logs ( I ran kubectl logs pods/.... and it return "entrypoint.sh exec error)

image

Then I reverted the code but still got the same error that container started by immediately error status

So I would like to ask by default will Spark-submit command cache anything as default behaviour if download dependencies fail ?

Additional context

No response

Have the same question?

Give it a 👍 We prioritize the question with most 👍