View and edit your host table
Use the dashboard viewing table to compare status, readings and counts. Search/filter hosts, sort columns and use Table options → Columns to add available fields such as endpoint, location, args and limits. Your table preferences are saved locally. Hover long text to read it fully.
Switch to Edit hosts to add, enable/disable or remove a monitor. Open edit details for timeout, credentials/key, Args, skip cycles and low/high limits. Save changes before returning to view. Low/high limits and skip cycles live in the details dialog so the main edit table stays compact.
Open a host chart
Use the host’s chart control. The chart includes its reading unit, summaries and history. Show details expands above the plot with endpoint/configuration and monitor location. You can use full-screen view when you need more room.
A monitor location tells you where the checks run. A gap in readings does not by itself establish when an agent was online; inspect actual observations and agent status separately.
Datasets and time ranges
Selected dataset shows one collection of readings, with previous/next dataset navigation. Time range selects observations across datasets:
- Last hour
- Last 24 hours
- Last 7 days
- Last 30 days
- Custom From/To, up to 90 days
Custom inputs use your local time; the website sends explicit UTC instants. A range includes its start and excludes its end. The same observation can display a different clock time in another timezone. Alert timestamps use UTC.
Summaries and long ranges
Range summaries use all available readings within the chosen period, not just visible chart dots. Successful values contribute to mean/minimum/maximum; failed observations are counted separately.
For long periods, the range chart retains up to 960 representative points across 160 time buckets: first, last, minimum, maximum and first/last failure where present. That keeps the chart readable and preserves extrema, but does not display every individual failure or brief event. A shorter range gives you more detail.
Archived datasets that cannot be read by this chart are noted. Their missing samples are excluded from statistics. A retention allowance does not mean every archived point is loaded in every view. A range spanning a changed metric/configuration is not a reliable like-for-like comparison; choose a period using one measurement definition.
Failure and limit markers
Red downward arrows below the horizontal axis mark failed observations. They sit away from the measurement line so failures do not distort your response/voltage scale.
Coloured, outlined points indicate readings outside a configured low or high limit, with reference lines and explanatory tooltips. Colours use your current limits, including on historical data. Changing a limit changes that historical colouring. It does not change past readings or identify the exact point that caused a historical alert notification.
Negative physical values remain valid points. No reading/unavailable is different from 0 and from a negative current. See Bluetooth measurements and alerts.
New data without losing your place
The background refresh preserves the period you are viewing. If your view includes the latest collecting dataset, a New data available indicator lets you load updates when you choose. A fixed historic period is not replaced by the newest data.
If you see no readings, check the time range, host enablement, location, agent and any archived-coverage notice. Follow troubleshooting. For summaries you can share, use reports and downloads.