Your first on-call shift, and how to get through it

Nobody forgets their first on-call shift. Mine was a long quiet stretch followed by one page I did not understand, at an hour when my brain was not at its best, for a system I had only read about. I fixed it eventually. I also learned more in that one hour than in the month before it.

If you are about to carry the pager for the first time, the fear is normal and mostly misplaced. The job is not to know everything. It is to stay calm, find the right information, and know when to pull someone in. A few things actually help.

You are not expected to know everything

The most common first-shift mistake is believing you have to solve every problem alone. You do not. On-call is not a test of whether you personally can fix anything that breaks. It is a test of whether the right thing happens when something breaks, and pulling in someone who knows the system is the right thing far more often than struggling in silence.

Escalating is not failing. The engineer who wakes a teammate after fifteen minutes of being stuck looks a lot better in the postmortem than the one who burned two hours protecting their pride while the site stayed down.

Read before you touch

When a page comes in, the urge is to do something immediately. Resist it for sixty seconds. Read the alert. Look at what it is actually telling you. Check whether anything changed recently, a deploy, a config push, a dependency.

Most incidents have a recent cause. A change went out, and now something is unhappy. If you can find the change, you are usually most of the way to the fix. Acting before you understand is how a small problem becomes two problems.

Write things down as you go

During the incident, keep a running note of what you saw and what you did, with rough times. It feels like a distraction in the moment. It is the most useful thing you will do.

It helps you think clearly while you are in it, it gives whoever you escalate to an instant picture, and it turns the postmortem from a memory exercise into a copy edit. A tool that keeps that timeline for you automatically is better still, because the worst time to take notes is while you are fixing something.

It gets easier, and that is the point

Your first few shifts are hard because everything is unfamiliar. After a handful, you start to recognize the shapes. This kind of alert usually means that. This service fails in these ways. The unknown shrinks.

That is the real reason on-call is worth doing well. Every shift makes the next one less frightening, and makes you the person a future first-timer escalates to. Be kind to them when they do.

On-call is easier when the information you need is already in front of you. Vigiles gives every incident a timeline and a clear record, so a page is something you can read, not decode. Start free.