Github user tmckayus commented on a diff in the pull request:
https://github.com/apache/spark/pull/21572#discussion_r195777382
--- Diff:
resource-managers/kubernetes/docker/src/main/dockerfiles/spark/entrypoint.sh ---
@@ -109,9 +81,16 @@ case "$SPARK_K8S_CMD" in
)
;;
+ init)
+ CMD=(
+ "$SPARK_HOME/bin/spark-class"
+ "org.apache.spark.deploy.k8s.SparkPodInitContainer"
+ "$@"
--- End diff --
any place in this entrypoint where $@ is being referenced by a spark-on-k8s
command still needs the shift that was taken out on line 44, correct? For the
unknown command case we just want a passthrough, but the main case statement
historically is expecting the spark-on-k8s command to be stripped. The
easiest/clearest way to preserve this might just be an extra case statement
like this at line 44:
```bash
case "$SPARK_K8S_CMD" in
driver | executor | init)
shift 1
;;
esac
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
