You encounter a large number of outages in the production systems you support. You receive alerts for all the outages, the alerts are due to unhealthy systems that are automatically restarted within…

A fundamental SRE principle for alerting is that alerts should be actionable and require immediate human attention. If systems automatically recover within a minute without requiring intervention, the alerts are generating noise and contributing to alert fatigue, directly causing staff burnout; eliminating such non-actionable alerts allows staff to focus on real problems.

Redefining the related SLO so that the error budget is not exhausted would potentially mask underlying reliability issues rather than solve the problem of excessive, non-actionable alerts causing burnout.

Distributing alerts to engineers in different time zones only shifts the burden of alert management and does not reduce the number of non-actionable alerts or address the root cause of burnout from excessive notifications.

Creating an incident report for each minute-long, self-recovering outage would significantly increase workload and administrative overhead, directly contributing to staff burnout rather than preventing it.