A maintenance window is a scheduled time when a team performs planned work on a system, such as patching servers, upgrading a database or moving to new hosting. Users are told in advance that the service may be unavailable or slower.
Why it matters
Planning changes for low-traffic hours and announcing them reduces surprise and customer impact. Monitoring should know about the window too: otherwise it sends alerts for expected downtime and records it in uptime reports. Many SLAs exclude properly announced maintenance from availability calculations, though the terms differ by contract.
Example
Window: Sunday 02:00-03:00 UTC
Work: database major version upgrade
Rollback: restore from snapshot if not healthy by 02:45
Notice: status page post 7 days before, reminder 24 hours before
During the window the monitor is paused or tagged as maintenance, and the site shows an HTTP 503 page with a Retry-After header.
Common mistakes
- Forgetting to pause alerts, so on-call is paged for planned work.
- Overrunning the window with no rollback decision point.
- Serving 200 on a maintenance page, which search engines may index. Use 503.
- Leaving monitors paused after the work is finished.
Related
Get notified when this term updates
Plus occasional practical tips on uptime and monitoring, nothing else. Unsubscribe anytime.
This page is general information written by our editorial team and updated when documentation changes. Servers and providers differ, so test changes safely first. See our editorial policy and disclaimer.