Availability goals and reports
Review observed availability, data coverage and results for completed periods.
On this page
What you need
Use an HTTP monitor in the workspace. Editors and owners can manage goals and report schedules; viewers can read existing results.
How to do it
- Open Service-level objectives and create a goal for the HTTP monitor. Choose a target percentage and evaluation period.
- Read availability together with known, unknown and unobserved time. Choose the separate maintenance-adjusted result only when you need it.
- Open Reports → Schedules and create a daily, weekly or monthly schedule. Check its local time, timezone and language.
- Open Reports → Analytics. Choose a report, period, grouping and filters. Save the selection under a name to reopen it on another device. Its menu lets you rename, replace or delete it. Relative periods stay current; fixed dates stay fixed. Saving does not extend result expiry. Use the table icon to inspect chart data and export the complete selection as CSV or PDF.
Good to know
Reports support periods up to 365 days, with hourly grouping up to 14 days. Availability uses known uptime and downtime; data coverage and unknown time remain separate. MTTR and MTTA include the full incident duration for the respective completion cohort. Certificate inventory and the last heartbeat are dated snapshots. Missing historical counts are never estimated. Interactive results expire after 24 hours.
Create a schedule from the current analysis or select report sections in a schedule. Daily, weekly and monthly schedules use completed calendar periods in the chosen timezone. Groups resolve when each report starts. CSV contains every filtered table row. PDF uses the same snapshot and is limited to 100 pages, 20 MiB and 60 seconds. Exceeding a limit produces an error; use CSV or a smaller selection. Saved period reports remain available for 90 days.
If something goes wrong
If an evaluation is uncertain, inspect data coverage and pending observations rather than rounding missing time into success. A queued report needs time to finish; check its status instead of repeatedly starting the same period.
Related guides
- Dashboard and filters
See your monitors, recent readings and open incidents in one place.
- Monitor a website or HTTP API
Check your website or public API for availability, response time and certificate expiry.
- Incidents and delivery history
Check a problem's timeline separately from the delivery of its messages.