Understanding Server SLAs and Uptime Guarantees

Service Level Agreements (SLAs) define the minimum level of service a hosting provider commits to delivering. Understanding what is actually covered — and what is not — helps you set realistic expectations and plan your own redundancy strategy.

What Is an SLA?

An SLA is a contractual commitment from a hosting provider that specifies measurable service guarantees, typically including network uptime, hardware availability, and the remedies (usually service credits) when those guarantees are not met.

Understanding Uptime Percentages

# What uptime percentages actually mean in annual downtime:
# 99.0%   = 3 days, 15 hours, 36 minutes
# 99.5%   = 1 day, 19 hours, 48 minutes
# 99.9%   = 8 hours, 45 minutes, 36 seconds
# 99.95%  = 4 hours, 22 minutes, 48 seconds
# 99.99%  = 52 minutes, 33.6 seconds
# 99.999% = 5 minutes, 15.36 seconds (five nines)

# Monthly equivalent:
# 99.9%  = ~43 minutes downtime per month
# 99.95% = ~22 minutes downtime per month
# 99.99% = ~4.3 minutes downtime per month

What SLAs Typically Cover

Network Uptime

The most common SLA metric. It guarantees that the network infrastructure — switches, routers, and uplinks — will be available. This is usually the highest percentage (99.99% or more) because providers have redundant network paths.

Hardware/Server Uptime

Guarantees that the physical server hardware will be functional. Typical SLA: 99.9-99.99%. Covers hardware failures like disk crashes, memory errors, and power supply failures.

What SLAs Do Not Cover

• Software issues — Your application crashing is not an SLA event
• Configuration errors — Misconfigured firewalls or services
• DDoS attacks — Often excluded or handled separately
• Scheduled maintenance — Usually excluded with advance notice
• Force majeure — Natural disasters, wars, government actions
• Third-party services — DNS, CDN, or upstream provider outages
• OS/software vulnerabilities — If your unpatched server gets hacked

Understanding Service Credits

When an SLA is breached, the typical remedy is service credits — a percentage discount on your next bill.

# Typical SLA credit structure:
# 99.9% - 99.5% uptime:    10% credit
# 99.5% - 99.0% uptime:    25% credit
# 99.0% - 95.0% uptime:    50% credit
# Below 95.0% uptime:      100% credit

# Important: Credits are usually capped at 100% of one month
# and must be claimed within a specific timeframe (often 30 days)

Building Your Own High Availability

No single server — regardless of SLA — can guarantee 100% uptime. If your application requires near-zero downtime, you need to build redundancy yourself.

Application-Level Redundancy

# Strategy 1: Multiple servers with a load balancer
# - 2+ application servers in different locations
# - Database replication (primary to replica)
# - Shared session storage (Redis)
# - Health checks and automatic failover

# Strategy 2: DNS-based failover
# - Multiple A records pointing to different servers
# - Health check service removes failed servers from DNS
# - Cloudflare load balancing handles this well

# Strategy 3: Active-passive standby
# - Primary server handles all traffic
# - Standby server receives database replication
# - Manual or scripted failover when primary fails

Monitoring Your Own Uptime

# Set up independent external monitoring:
# Free options:
# - UptimeRobot (50 monitors free, 5-minute checks)
# - Hetrixtools (15 monitors free, 1-minute checks)
# - Freshping (50 monitors free)

# What to monitor:
# 1. HTTP/HTTPS endpoint (is your app responding?)
# 2. TCP port checks (is the port open?)
# 3. SSL certificate expiry
# 4. DNS resolution
# 5. Response time thresholds

Questions to Ask Your Provider

1. What is the network uptime SLA?
2. What is the hardware/server uptime SLA?
3. What constitutes downtime for SLA purposes?
4. How do I claim service credits?
5. What is the maximum credit available?
6. Are scheduled maintenance windows excluded?
7. How much advance notice is given for maintenance?
8. What DDoS protection is included?
9. What is the mean time to repair (MTTR) for hardware failures?

The Bottom Line

An SLA is a commitment, not a guarantee of zero downtime. For business-critical applications, combine a reliable provider with your own redundancy strategy, monitoring, and incident response plan.