holdenk commented on a change in pull request #30892:
URL: https://github.com/apache/spark/pull/30892#discussion_r548237305
##########
File path:
resource-managers/kubernetes/core/src/main/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodsSnapshot.scala
##########
@@ -65,14 +72,31 @@ object ExecutorPodsSnapshot extends Logging {
case "pending" =>
PodPending(pod)
case "running" =>
+ // If we're checking all containers look for any non-zero exits
if (shouldCheckAllContainers &&
"Never" == pod.getSpec.getRestartPolicy &&
pod.getStatus.getContainerStatuses.stream
.map[ContainerStateTerminated](cs => cs.getState.getTerminated)
.anyMatch(t => t != null && t.getExitCode != 0)) {
PodFailed(pod)
} else {
- PodRunning(pod)
+ // Otherwise look for the Spark container
+ val sparkContainerStatusOpt =
pod.getStatus.getContainerStatuses.asScala
+ .find(_.getName() == sparkContainerName)
+ sparkContainerStatusOpt match {
+ case Some(sparkContainerStatus) =>
+ sparkContainerStatus.getState.getTerminated match {
+ case null => PodRunning(pod)
+ case t =>
Review comment:
We can also have the situation where there is no terminated state, so it
should be pod running, but I can collapse it into a single case/switch.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
