Parameters

This component is multi-instance. The parent key for all of the following parameters is the instance name you assign in your inventory. For the default instance, that key is filesystem_exporter.

Example inventory snippet:

applications:
  - filesystem-exporter as archive

parameters:
  archive:
    namespace: syn-archive
    volume:
      persistentVolumeClaim:
        claimName: archive

`namespace`

type	string
default	`syn-filesystem-exporter`

The namespace in which to deploy this component. By default this is syn-${_instance}.

`images.exporter`

type	object
default	`images: exporter: registry: ghcr.io repository: containeroo/filesystem-exporter tag: v1.2.1`

The container image used for the exporter.

`collector.interval`

type	string
default	`5m`

Controls the exporter flag -collector.interval. This defines how often the exporter refreshes filesystem usage data in the background. The collection-failure and collection-stale alerts scale from this value.

`collector.timeout`

type	string
default	`2m`

Controls the exporter flag -collector.timeout. This defines the timeout for a single collection run. If this exceeds collector.interval, the collection-failure and collection-stale alerts scale from the timeout instead.

`report_child_dirs`

type	boolean
default	`false`

Controls the exporter flag -filesystem.report-child-dirs. When set to true, the exporter also reports metrics for immediate child directories under /data. When set to false, only the root path is reported.

`volume`

type	object
default	`null`

The Kubernetes volume source for the single filesystem mounted by this exporter instance. This parameter is required. On OpenShift, if this volume uses nfs, the component creates a dedicated service account plus a namespaced RoleBinding to the built-in ClusterRole system:openshift:scc:hostmount-anyuid. Applying that binding still requires a sufficiently privileged reconciler, because granting SCC usage is itself a privileged operation.

Example:

volume:
  persistentVolumeClaim:
    claimName: archive

The component hardcodes the volume name to data, mounts it read-only, and always mounts it at /data.

`monitoring_enabled`

type	boolean
default	`true`

Enables the monitoring resources. When enabled, the component renders a ServiceMonitor and a PrometheusRule and adds the same namespace-level monitoring labels described by component-backup-k8up: SYNMonitoring=main on most distributions, or openshift.io/cluster-monitoring=true on OpenShift 4.

`grafana_dashboard`

type	object
default	`grafana_dashboard: enabled: false`

Controls an optional Grafana dashboard ConfigMap.

Example

namespace: example-namespace
volume:
  persistentVolumeClaim:
    claimName: data
grafana_dashboard:
  enabled: true