Uptime

Uptime measures how reliably a service stays online, expressed as the percentage of a period during which it was reachable. "Three nines" (99.9%) allows roughly 8.7 hours of downtime a year; "four nines" (99.99%) allows about 52 minutes. The opposite figure is downtime. Uptime is distinct from latency and throughput: a server can be up but slow, or fast but frequently offline. For anything unattended — a scraper, a scheduled job, a monitoring agent — sustained uptime is what keeps the work flowing without manual restarts.

For automation, plan for failed connections. Our free proxy list shows recorded check results. The catalogue retains historical records, including entries marked as not working. A recorded success does not guarantee a connection now. Review the status and timestamp, then validate an endpoint before use. Authentication or payment does not guarantee availability or acceptance by a destination. Related services: s4m proxies and VPN.

In your own tooling, treat uptime defensively: poll a lightweight health endpoint, retry with exponential backoff, and switch to a healthy exit if one goes quiet. For automation against the s4m free list, re-check a proxy immediately before a run, since public entries can drop between validations. Paid proxies and the VPN are built to stay available, but a resilient client that expects the occasional blip is what keeps long jobs finishing.