How uptime is measured
A monitor sends health checks at fixed intervals and marks each check as up or down. A basic calculation is successful checks ÷ total checks × 100. Measuring outage duration can capture gaps between checks more accurately.
At 99.9% uptime, downtime can total 43 minutes 12 seconds in a 30-day month or 8 hours 45 minutes 36 seconds in a 365-day year. Those figures assume continuous measurement with no exclusions.
An SLA may exclude scheduled maintenance or count downtime only after several failed checks. Compare the measurement period, test locations, failure threshold and exclusions before comparing percentages.
Uptime vs latency and bandwidth
Uptime describes availability, not speed or capacity.
| Metric | What it measures | Typical unit |
|---|---|---|
| Uptime | Time the service is usable | Percentage |
| Latency | Time a request takes | Milliseconds |
| Bandwidth | Data transferred per second | Mbps or Gbps |
A proxy can be available but slow. It can also respond quickly during healthy periods while experiencing frequent outages.
What lowers proxy uptime
A proxy is usually considered up when its gateway accepts a connection and can forward a test request. Availability can fall because of:
- Gateway failures: the proxy host or service stops accepting connections.
- DNS or routing problems: clients can’t reach the proxy address.
- Capacity limits: overloaded systems produce delays or a connection timeout.
- Unhealthy exit nodes: the gateway works, but requests can’t leave through part of the pool.
- Maintenance or configuration errors: a deployment temporarily interrupts service.
A target website’s outage doesn’t necessarily indicate proxy downtime. A 502 Bad Gateway or 503 Service Unavailable can come from the proxy or the destination, so identify the sender before classifying the failure.
How to monitor uptime
- Send a small HTTPS request through the proxy to an endpoint you control.
- Record DNS, connection, TLS and HTTP failures separately to locate the fault.
- Test from at least two networks and monitor the proxy locations your application uses.
- Separate configuration errors from outages. A proxy sends 407 Proxy Authentication Required when the test omits valid proxy credentials or supplies invalid ones.
FAQ
What does 99.9% uptime mean?
It means the service was available for 99.9% of the defined measurement period. It still permits about 43 minutes of downtime in a 30-day month if no time is excluded.
Is uptime the same as availability?
Often, but definitions vary. An SLA may distinguish a server process running from the service being reachable and usable, so check how availability is calculated.
Does high proxy uptime guarantee successful requests?
No. Requests can still fail because of invalid credentials, target-site outages, rate limits or blocks. Uptime only measures whether the proxy service itself is available under the stated test conditions.