Incident
An unplanned interruption to a service, or a drop in its quality, which is logged, prioritised and worked on until the service is back.
Also written P1, priority 1, severity 1, incident ticket, outage, bridge call
An incident is anything unplanned that stops a service or makes it worse: online banking is down, a batch job abends, a file to a partner is late. Each one is recorded as a ticket, which tracks what happened, who is working on it and what was done.
Incidents are given a priority, often from 1 to 4. A priority 1, or P1, is the most serious: a major service is down, or a deadline will be missed. A P1 usually starts a bridge call, where everyone needed joins one call until it is fixed, and an incident manager leads, communicates and keeps a timeline.
The first goal is to restore the service, even with a temporary fix, such as restarting a region or skipping a job. Finding out why it happened comes afterwards. That deeper work is often tracked separately as a problem record, which leads to the permanent fix.
For people in support, handling incidents well is a core skill: stay calm, write down what you see and do, say clearly what you know and what you do not, and follow the runbook.
Where can I learn this hands-on?
Right here, at ZedBank. You take a junior job at a simulated bank and learn each part of the work by doing it, with Zed beside you. There's nothing to install.
Start at ZedBankRelated
- Production supportKeeping live systems running: diagnosing failures, fixing what broke, and getting the service back.
- RunbookThe written procedure for operating a system or responding to a known failure, step by step.
- On-callBeing available outside working hours to respond when production systems have a problem, usually in turns with colleagues.
- Emergency changeA change to production that cannot wait for the normal approval process, usually to fix a live problem, approved quickly and reviewed afterwards.
- SLAService level agreement: an agreed target for a service, such as how available it must be or by what time a batch run must finish.