Not all incidents need you to get out of the bed. Incidents effect primarily in these 4 categories. You can also create incidents from your code using webhooks.
Books and transactions failure
Website or app crashes
Security vulnerabilities
Server crash and burnouts
Dashboards failing to load
Delayed response times
Network outages
Queues overflows
DB connections failing
Cron job failures
Poor web page performance
RabbitMQ or Queuing stuck
Errors via Sentry/Rollbar
Server consumption overload
CI/CD alerts
Auto scaling alerts
Logging alerts
CPU, Memory, Disk I/O alerts
There are three statuses for each incident namely Triggered, Acknowledged and Resolved
Every new incident will be in triggered state. It would mean that someone will need to look into this and address the problem.
Actively alerting via Escalation policy in this status
Every new incident will be in triggered state. It would mean that someone will need to look into this and address the problem.
No alerts for current incident. Repeated incidents are automatically suppressed.
Every new incident will be in triggered state. It would mean that someone will need to look into this and address the problem.
If incident occurs again, alerts will start from the top of escalation policy
Soon as you are alerted, mark your new incident as Acknowledged signifying that you are working to Resolve the incident. When in Acknowledged state, we do not send alerts and suppress repeating alerts automatically.