On-Call

On-call is an arrangement where one or more people are responsible, for a set period, for responding to production alerts. The person on call is often paged by a phone call, SMS or app notification, and is expected to acknowledge it within an agreed time. Historically the device used was a pager, and the word "paged" stuck.

Why it matters

Services fail at inconvenient times, and a clear owner prevents the situation where everyone assumes someone else saw the alert. A well-run rotation spreads the load fairly, defines who is the primary and backup, and links to an escalation policy if the first person does not respond.

Example

A small team might run:

Primary:   rotates weekly, Monday 10:00 to Monday 10:00
Secondary: the previous week's primary
Ack target: 5 minutes; escalate to secondary after 10

Handover includes open incidents and known flaky alerts.

Common mistakes

• A rotation of one person, which guarantees burnout and a single point of failure.
• Paging without runbooks, so the responder has to guess.
• Not compensating or giving time off in return, where employment rules or policies apply.
• Letting noisy alerts wake people repeatedly.

Related

• Escalation policy
• Runbook
• Alert fatigue
• MTTA

यह पेज सामान्य जानकारी है, जिसे हमारी संपादकीय टीम ने लिखा है और दस्तावेज़ीकरण बदलने पर अपडेट किया जाता है। सर्वर और प्रदाता अलग-अलग होते हैं, इसलिए बदलाव पहले सुरक्षित तरीके से जाँचें। देखें: संपादकीय नीति और अस्वीकरण.