attilapiros commented on a change in pull request #32221:
URL: https://github.com/apache/spark/pull/32221#discussion_r615667822
##########
File path:
resource-managers/kubernetes/core/src/main/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodsWatchSnapshotSource.scala
##########
@@ -58,10 +58,14 @@ private[spark] class ExecutorPodsWatchSnapshotSource(
snapshotsStore.updatePod(pod)
}
- override def onClose(e: KubernetesClientException): Unit = {
+ override def onClose(e: WatcherException): Unit = {
logWarning("Kubernetes client has been closed (this is expected if the
application is" +
Review comment:
> Actually, that's what @attilapiros meant, but it's not in reality. As
of now, it's frequently possible to observe K8s side exceptions, especially
HTTP-related ones, during application termination after `SparkContext.close()`
starts to doing several clean-ups.
In this case I am fine to leave it as it is.
--
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]
