On-call
On-call is the practice of assigning an engineer to respond to production problems during a set period, usually on a rotation. The on-call engineer carries a pager or phone app and must acknowledge alerts within an agreed time. If they do not, the alert escalates to a secondary engineer or a manager.
How it works
A team builds a schedule, often weekly, with a primary and a secondary. With 8 engineers on a weekly rotation, each is primary one week in 8, or 12.5% of the time. Alerts route through a paging tool that tracks acknowledgement, for example within 5 minutes for urgent pages. The engineer follows a runbook, mitigates, and opens an incident if needed.
Google's SRE book suggests SREs spend at most 25% of their time on-call and handle no more than two incidents per 12 hour shift, so each one gets a proper fix and review.
Why it matters in production
On-call is how users get help at 2 a.m. Too many pages cause alert fatigue and burnout, and tired engineers miss real problems. Healthy rotations need actionable alerts, good runbooks and time off after hard shifts.
See also
On the One2N blog
Further reading
