Downtime is the period during which a website, server, application or machine is not working or not reachable by its users. Its opposite, uptime, is the time it is running and available. Hosting and server availability is usually expressed as an uptime percentage (e.g. 99.9%); the remainder is the maximum downtime allowed.
| Term | Meaning |
|---|---|
| Downtime | Time the system is down or unreachable |
| Uptime | Time the system is up and reachable |
| Planned downtime | Announced outage for maintenance, updates or migration |
| Unplanned downtime | Sudden outage caused by failure, attack or error |
| SLA | The service level a provider commits to (e.g. 99.9% monthly uptime) and the compensation if it is missed |
Planned vs Unplanned Downtime
Planned downtime is a deliberate outage for OS updates, hardware replacement, database maintenance or migrations. It is announced in advance and scheduled for low-traffic hours. Many providers exclude announced maintenance from SLA calculations, so check that clause in the contract.
Unplanned downtime happens suddenly because of an unexpected problem. It is the costly kind: visitors cannot reach the site, online stores cannot take orders and email is delayed.
Common Causes
- Hardware failure: disk, RAM, power supply or network card.
- Software and configuration errors: a bad update, wrong web server or DNS settings, a full disk.
- Resource exhaustion: CPU, RAM or connection limits hit during a traffic spike, causing slowdowns or 502/503/504 errors.
- Network and power outages in the data centre.
- Cyber attacks: especially DDoS attacks.
- Human error: deleted files or stopped services.
- Expired domain or SSL certificate: the server runs but visitors cannot reach the site or see a security warning.
How Much Downtime Does an Uptime Percentage Allow?
Maximum downtime per uptime level (365-day year, 30-day month):
| Uptime | Max downtime per year | Max downtime per month |
|---|---|---|
| 99% | 3 days 15 h 36 min | 7 h 12 min |
| 99.5% | 1 day 19 h 48 min | 3 h 36 min |
| 99.9% | 8 h 45 min 36 s | 43 min 12 s |
| 99.95% | 4 h 22 min 48 s | 21 min 36 s |
| 99.99% | 52 min 34 s | 4 min 19 s |
| 99.999% | 5 min 15 s | 26 s |
The gap between 99% and 99.9% looks small but equals roughly three days a year. See also uptime guarantees in web hosting.
Measuring and Tracking Downtime
- Uptime monitors request your site at intervals from several locations and alert you by email, SMS or chat when it fails. Self-hosted open-source options include Uptime Kuma and Zabbix; see server monitoring tools.
- Server logs show when an outage started and why.
- Status pages are where providers announce maintenance and incidents.
- MTBF and MTTR: the longer the mean time between failures and the shorter the mean time to repair, the less downtime.
Is a site down for everyone or just you? Try another network (e.g. mobile data). If it opens there, the problem is likely your connection or DNS cache; if it opens nowhere, the site is down.
How to Reduce Downtime
- Choose the right infrastructure: enough resources and redundant hardware; move growing sites from shared hosting to a VDS or server.
- Set up monitoring so you hear about outages before customers do.
- Back up regularly to a separate location and test restores. See server backup strategies.
- Plan maintenance: test updates first, deploy in low-traffic hours and announce them.
- Spread the load: a CDN serves static content and absorbs spikes; critical systems use load balancing and standby servers.
- Secure the stack: firewall, DDoS protection and up-to-date software.
- Track renewals: enable auto-renew for domains and SSL certificates.
To keep your site online, see our hosting plans and, for projects that need more resources and control, our VDS servers.
Frequently Asked Questions
What is the difference between downtime and uptime?
Uptime is the time a system works; downtime is the time it does not. 99.9% uptime means up to 0.1% of the period may be downtime.
Is 100% uptime possible?
In practice, no. Every component eventually needs maintenance or fails; redundancy shortens outages but cannot eliminate them.
Categories
-
CentOS 7 guides
-
cPanel 20 guides
-
CyberPanel 18 guides
-
Operating Systems 81 guides
-
Juniper Networks 18 guides
-
Control Panels 50 guides
-
Linux 17 guides
-
MySQL 29 guides
-
Network 95 guides
-
Plesk Panel 12 guides
-
Virtual Servers 273 guides
-
SEO 3 guides
-
Cyber Security 8 guides
-
Technology 162 guides
-
VMware 36 guides
-
WordPress 20 guides
-
Artificial Intelligence 22 guides