Have You Tried Turning It Off and On Again? Origin and Meaning

“Have you tried turning it off and on again?” is the stock answer of IT support, and the catchphrase of Roy Trenneman in the British sitcom The IT Crowd (Channel 4, 2006–2013). The advice is much older than the show. The show made it a meme.

Where the line comes from

The IT Crowd was created by Graham Linehan and first broadcast on 3 February 2006. Roy, played by Chris O’Dowd, works in the basement IT department of Reynholm Industries. He answers the phone with the least effort possible, and “Have you tried turning it off and on again?” is his default. Wikipedia lists it among his stock suggestions, which he eventually records on tape so he does not have to say them himself.

The line appears in the first episode, “Yesterday’s Jam” (series 1, episode 1): “Hello, IT. Have you tried turning it off and on again?” (episode transcript on Movie Mistakes). It comes back many times across the series.

The advice itself has no single inventor. Help desks were telling people to restart their machines long before 2006. We found no documented first use of the exact phrase, so we will not invent one.

Why turning it off and on again actually works

The joke is funny because the advice is lazy. It is also correct surprisingly often. A restart puts the system back into a known state. Most bugs that a reboot fixes are bugs of state, not of code:

• Memory leaks. A program keeps allocating memory and never frees it. A restart frees everything.
• Stuck processes. A process waits on a lock or a resource that will never arrive. Killing it ends the wait.
• Corrupted state. A cache, a counter or a connection pool ends up with a value nobody planned for.
• Hardware and drivers. A device gets into a bad mode. Power-cycling resets it to its initial configuration.
• Pending updates. Some changes only apply on the next start.

Ops teams use the same idea on purpose. In Kubernetes, a liveness probe that fails makes the kubelet restart the container. It is “turning it off and on again”, automated:

livenessProbe:
  httpGet:
    path: /healthz
    port: 8080
  periodSeconds: 10
  failureThreshold: 3

The catch: a restart hides the symptom, not the cause. If a service needs a reboot every Tuesday, the leak is still there. Good engineers restart to recover, then look at the logs to find out why it works now.

Why it is funny for developers and sysadmins

Everyone in IT has said it, and everyone has been annoyed to hear it. It is the support version of a placebo that actually works. The person on the phone often already tried, or swears they did. Then they restart, and it works.

The IT Crowd captured the tone perfectly: Roy is not wrong, just tired. That mix of competence and exhaustion is what every help desk worker recognises.

Shirts for the help desk

Our shirt prints Roy’s question in full. It sits in the sysadmin collection with other ops jokes, including There’s no place like 127.0.0.1.

Unisex fit, S to 5XL. Check the size guide. More gift ideas in the Christmas gift guide for programmers.

Turning it off and on again, the professional way

A restart is a fair first step, if you do it with care. A few habits make the difference between a fix and a guess:

• Save the logs before the restart. Once the process is gone, its state is gone too.
• Note the time. If the problem returns at the same hour, look for a scheduled job.
• Restart the smallest thing that can fix it: the service before the server, the container before the node.
• Count the restarts. One is maintenance. A weekly one is a bug report.

On a Linux server with systemd, restarting a single service and reading its recent logs takes two commands:

sudo systemctl restart nginx
journalctl -u nginx --since "10 min ago"

The second command is the important one. The restart brings the service back. The logs tell you what to fix so you do not have to restart it again next week. Roy would not approve of the extra effort, but your on-call rota will.

FAQ

Which episode of The IT Crowd has “have you tried turning it off and on again”?

It appears in the first episode, “Yesterday’s Jam” (2006), and returns throughout the series.

Who says it in The IT Crowd?

Roy Trenneman, played by Chris O’Dowd.

Why does restarting a computer fix problems?

It clears memory, ends stuck processes and resets hardware to a known state. It fixes state problems, not the bug that caused them.