Incidents Management Runbooks

Over the course of time, multiple events can occur, forecasting a possible failure that need to be taken care of or symptomatic of a much deeper on-going problem, leading to incident up to service disruption.

Dalet Flex comes with codified errors and alerts management. Based on the various collected metrics, the system is configured to raise alerts triggered by events.

Alerts are pretty simple: if a specific condition is met for a certain duration, then an alert is being sent with a categorized severity.

Examples can be:

• raise an alert if instance CPU usage is high for 15 minutes or more
• RabbitMQ queue is constantly growing (i.e. ingested messages not being
• ...

Alerts are then codified are displayed as runbooks, collection of recipes about:

• incident description
• associated severity
• detailed customer impact
• operational remoediation process