We are currently trying to provision Grafana alert rules using a Kubernetes ConfigMap in an OpenShift environment where Grafana open source(v 11.6.4) is deployed. The alert rules YAML was exported directly from the Grafana UI and we are mounting the ConfigMap to /opt/bitnami/grafana/conf/provisioning/alerting as per the standard provisioning approach.
The ConfigMap is successfully created and visible in the cluster, and when we inspect the deployment configuration, the volume mounts appear to be correctly defined. However, the alert rules are not getting provisioned in the Grafana UI. When we exec into the Grafana pod and inspect the /opt/bitnami/grafana/conf/provisioning/alerting directory, the expected alerting file from this ConfigMap is not present. At the same time, other provisioning files are present in their respective directories and are working as expected, which suggests that provisioning in general is functioning.
We have verified that the YAML schema matches the format exported by Grafana and does not appear to have structural issues. There are also no obvious errors shown in the Grafana UI related to alert provisioning.
We are trying to understand if there are any additional requirements or nuances specific to alert provisioning compared to dashboards or datasources, especially in an OpenShift setup. It would also be helpful to know if Grafana silently ignores alerting files under certain conditions (such as naming conventions, file placement, or permissions), or if there are specific logs or debug settings we should enable to troubleshoot why the file is not appearing inside the container.
Any pointers or guidance would be appreciated.
We have also enabled the log level at “debug” but dont see any error or unusual behaviour