Today · 45 min, 2 hrs
45–120 min
Humans carry the effort
- Incident triggers notification to on-call SRE
- Switch between multiple monitoring tools and cloud consoles
- Hunt for an outdated runbook
- SSH in, run diagnostics, execute fix manually
- Update the ticket, hope for the best







