-
Bug
-
Resolution: Done
-
Minor
-
Logging 5.5.0
-
False
-
False
-
NEW
-
VERIFIED
-
-
Logging (LogExp) - Sprint 215
the configMap grafana-dashboard-elasticsearch, managed by elasticsearch operator is being created + deleted all the time.
It seems it's removed before the reconcile and immediately created afterwards.
A possible root cause is that the reconcile code is including explicitly the deletion of this configMap in some conditions:
func (r *ElasticsearchReconciler) Reconcile(ctx context.Context, request ctrl.Request) (ctrl.Result, error) {
// Fetch the Elasticsearch instance
cluster := &loggingv1.Elasticsearch{}
err := r.Get(ctx, request.NamespacedName, cluster)
if err != nil {
if apierrors.IsNotFound(err) {
log.Info("Flushing nodes", "objectKey", request.NamespacedName)
elasticsearch.FlushNodes(request.NamespacedName.Name, request.NamespacedName.Namespace)
elasticsearch.RemoveDashboardConfigMap(r.Client)
if err := console.DeleteKibanaConsoleLink(context.TODO(), r.Client); err != nil {
log.Error(err, "failed to delete consolelink")
}
return ctrl.Result{}, nil
And elasticsearch.RemoveDashboardConfigMap(r.Client) is doing this:
// RemoveDashboardConfigMap removes the config map in the grafana dashboard
func RemoveDashboardConfigMap(c client.Client) {
This has been added in this pull request:
https://github.com/openshift/elasticsearch-operator/pull/610/files
Also we can even check in the console or by the command line that the "created at" is always showing "now" except for some secnods where we can see a former date.
This has been reported in bz:
https://bugzilla.redhat.com/show_bug.cgi?id=2060426
And in the mirrored jira:
https://issues.redhat.com/browse/OCPBUGSM-41313
This is not harmful at all. My customer has some logs where we can see the add and delete of this configMap all the time and it's provoking just some confusion.
