Downtime is the period when a service, application, or infrastructure component is unavailable or not functioning properly. It's the opposite of uptime and is measured as the total duration a system is inaccessible, offline, or degraded below acceptable performance thresholds.
Downtime can be categorized into two types:
Planned downtime occurs during scheduled maintenance windows for updates, patches, or infrastructure upgrades. Users are typically notified in advance.
Unplanned downtime results from unexpected failures—hardware faults, bugs, attacks, power outages, or configuration errors—and directly impacts your SLA and error budget.
Monitoring tools detect downtime through health checks, heartbeat monitoring, and synthetic monitoring, which test your service continuously. When downtime occurs, it triggers alerts to on-call teams, who follow a runbook to restore service and measure MTTR (Mean Time to Recovery).
Downtime costs money—lost transactions, damaged reputation, and customer churn. That's why organizations invest in redundancy: load balancers, DNS failover, multi-region deployment, and status pages to keep users informed during incidents.
TipTrack both frequency and duration. A system with five 10-minute outages differs operationally from one 50-minute outage, even though total downtime is identical. Use uptime percentage targets (99.9%, 99.99%) backed by your SLA to define acceptable downtime thresholds.
Try it with a free tool
Guides that use this term
Related terms
New articles, straight to your inbox
I’d like to receive product updates and new articles from InfraNest by email. I can unsubscribe at any time. Privacy policy