wangyang0918 commented on a change in pull request #10965: [FLINK-15797][k8s]
Reduce log noise of Fabric8FlinkKubeClient
URL: https://github.com/apache/flink/pull/10965#discussion_r373491974
##########
File path:
flink-kubernetes/src/main/java/org/apache/flink/kubernetes/kubeclient/Fabric8FlinkKubeClient.java
##########
@@ -269,14 +271,14 @@ public void eventReceived(Action action, Pod pod) {
callbackHandler.onDeleted(Collections.singletonList(new
KubernetesPod(flinkConfig, pod)));
break;
default:
- LOG.info("Skip handling {}
event for pod {}", action, pod.getMetadata().getName());
+ LOG.debug("Ignore handling {}
event for pod {}", action, pod.getMetadata().getName());
break;
}
}
@Override
public void onClose(KubernetesClientException e) {
- LOG.error("Pods watcher onClose", e);
+ LOG.debug("The pods watcher is closing.", e);
Review comment:
@GJL It is a very nice catch. I do not think the watcher will be stopped
even some exception occurs. Since the default value of `watchReconnectLimit` is
-1. That means the `WatchConnectionManager` will always reconnect. The only one
case is the `WatchConnectionManager#close()` by expected.
However, if the `watchReconnectLimit` is configured by users via java
properties or environment, the watch may be stopped and all the changes will
not be processed properly. I will create a new ticket to track this. A fatal
exception will be thrown when the watcher is closed.
----------------------------------------------------------------
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]
With regards,
Apache Git Services
