Uptime Monitoring
What Uptime Monitoring Is
Section titled “What Uptime Monitoring Is”Uptime monitoring answers one question: is the service reachable and behaving, right now and over time. In LoadFocus there is no separate "uptime monitor" object to create. Uptime is what your existing checks measure, and it is reported in two places: your status pages and your SLA and SLO reports. Your alerts tell you when a check fails.
For an overview of the feature and its plans, see Uptime monitoring.
That distinction matters when you are setting things up. You do not pick "uptime monitoring" from a menu. You create the check that matches what you need to prove is up, and uptime follows from it.
Which Check Measures Which Kind of Uptime
Section titled “Which Check Measures Which Kind of Uptime”| What you need to prove is up | Use |
|---|---|
| A single HTTP endpoint responds correctly | API check |
| A whole API workflow still completes | Multistep API check |
| A user journey still works in a real browser | Browser check |
| A non-HTTP service accepts connections | TCP port monitor |
| DNS records still resolve to the right values | DNS monitor |
| A scheduled job actually ran | Heartbeat monitor |
| A certificate is not about to expire | SSL certificate monitoring, included on every HTTPS API check |
All of these share the same alert channels. API, multistep and browser checks run on a schedule from your chosen cloud regions; a heartbeat monitor instead waits for a ping from your job.
How Uptime Is Calculated
Section titled “How Uptime Is Calculated”Uptime is the time-weighted percentage of successful checks over the window you select. A check that was down for one hour of a 30-day window lowers uptime in proportion to that hour, rather than counting as a single failed run. Where there are no results in the window at all, uptime is shown as a dash rather than as zero, because no data is not the same as downtime.
Where Uptime Is Reported
Section titled “Where Uptime Is Reported”Status pages show live status and 90-day uptime for the services you choose, driven by your existing monitors. You can post incidents, schedule maintenance windows so planned work does not read as an outage, and let visitors subscribe to updates. A status page can be served on your own domain.
SLA and SLO reports turn the same results into a target. Set an SLO per monitor, for example 99.9%, and LoadFocus tracks whether you are meeting it, how much error budget is left, alerts you on a breach, and emails a recurring SLA report. SLA and SLO reports are a paid feature.
Reducing False Alarms
Section titled “Reducing False Alarms”A single failed run is not always an outage. Configure check retries and alert thresholds so a transient network blip does not page your on-call, and use maintenance windows for planned work.
What is uptime monitoring?
Section titled “What is uptime monitoring?”Uptime monitoring is the practice of checking on a schedule whether a service is reachable and responding correctly, then reporting availability over time. In LoadFocus it is produced by your API, browser, TCP, DNS and heartbeat checks rather than by a separate monitor type.
How is uptime percentage calculated?
Section titled “How is uptime percentage calculated?”It is the time-weighted percentage of successful checks across the window you select, so an hour of downtime in a 30-day window reduces uptime in proportion to that hour.
Do I need a special monitor type for uptime?
Section titled “Do I need a special monitor type for uptime?”No. Create the check that matches what you need to prove is working, and uptime is reported from it on your status pages and SLA reports.
Can I show uptime publicly?
Section titled “Can I show uptime publicly?”Yes. A status page shows live status and 90-day uptime for the services you choose, and can be served on your own domain.
How do I stop a brief network blip from paging me?
Section titled “How do I stop a brief network blip from paging me?”Use check retries and alert thresholds so a check must fail more than once before an alert fires, and schedule maintenance windows around planned work.