Alert Management

As we've seen, Flex comes bundled with observability tools and allows Ops and IT administrators to get notified of possible system or application status before it leads to production incidents.

This is done automatically by comparing key Prometheus metrics to specific thresholds for a given duration of time. Once reached, the built-in AlertManager system is used to notify the requested people accordingly.

Dalet Flex supports multiple notification backends:

Note

Keep in mind that multiple backends may be used at a given time, with different severities processing.

A good example is having your team notified over a MS Teams or Slack channel for any little thing adding up but only triggering a PagerDuty alert to anyone's on call for critical alerts.