Alerting

Simplyblock uses Grafana to configure and manage alerting rules.

By default, Grafana is configured to send alerts to Slack channels. However, Grafana also allows alerting via email notifications, but this requires the use of an authorized SMTP server to send message.

An SMTP server is currently not part of the management stack and must be deployed separately. Alerts can be triggered based on on-time or interval-based thresholds of statistical data collected (IO statistics, capacity information) or based on events from the cluster event log.

Pre-Defined Alerts

The following pre-defined alerts are available:

Alert	Trigger
device-unavailable	Storage device became unavailable.
device-read-only	Storage device changed to status: read-only.
cluster-status-degraded	Storage node changed to status: degraded.
cluster-status-suspended	Storage node changed to status: suspended.
storage-node-unreachable	Storage node became unreachable.
storage-node-offline	Storage node became unavailable.
storage-node-healthcheck-failure	Storage node with negative healthcheck.
logical-volume-offline	Logical volume became unavailable.
critical-capacity-reached	Critical absolute capacity utilization in cluster was reached. The threshold value can be configured at cluster creation time using --cap-crit.
critical-provisioning-capacity-reached	Critical absolute provisioned capacity utilization in cluster was reached. The threshold value can be configured at cluster creation time using --prov-cap-crit.
root-fs-low-disk-space	Root filesystem free disk space is below 20%.

It is possible to configure the Slack webhook for alerting during cluster creation or to modify it at a later point in time.