DEBUG-level logging of org.elasticsearch.action by default seems excessive

[DaveCTurner]

In 2bb31fe (v0.6.0!) we added DEBUG-level logging to the default config of the action logger "for easier debugging". This lives on to this day in the default config and results in a good deal of log noise sometimes. For instance, an exception thrown in an implementation of TransportMasterNodeAction#masterOperation results in a retry or else it is propagated back to the caller but is also logged at DEBUG level, complete with stack trace, by TransportMasterNodeAction on the way past:

elasticsearch/server/src/main/java/org/elasticsearch/action/support/master/TransportMasterNodeAction.java

Lines 155 to 166 in 96b4f3d

	ActionListener<Response> delegate = ActionListener.delegateResponse(listener, (delegatedListener, t) -> {
	if (t instanceof FailedToCommitClusterStateException || t instanceof NotMasterException) {
	logger.debug(() -> new ParameterizedMessage("master could not publish cluster state or " +
	"stepped down before publishing action [{}], scheduling a retry", actionName), t);
	retry(t, masterChangePredicate);
	} else {
	logger.debug("unexpected exception during publication", t);
	delegatedListener.onFailure(t);
	}
	});
	threadPool.executor(executor)
	.execute(ActionRunnable.wrap(delegate, l -> masterOperation(task, request, clusterState, l)));

It seems odd to single out this one package for DEBUG-level logs; these days many actions are in other packages (e.g. within org.elasticsearch.xpack.*). Do we still see value in this being part of the out-of-the-box default config? I think we should move to not emitting any DEBUG-level logs by default.

[elasticmachine]

Pinging @elastic/es-core-infra (:Core/Infra/Core)

[nik9000]

I think we should move to not emitting any DEBUG-level logs by default.

+1. I think that is what DEBUG is supposed to mean. When I first saw these were on years ago I thought it was a temporary thing. Then I forgot about it.

[DaveCTurner]

Seeing no objections, I opened #51459