Maintenance
The Maintenance page puts your fleet on a map so a repair trip can be planned around where the problems actually are. A list tells you twelve monitors need attention; a map tells you that nine of them are on one road.
Prerequisites
An active workspace. Maintenance carries the same visibility as Alerts — if it's missing from your sidebar, see Access Beacon.
Finding the Monitors That Need a Visit
Filter the map down to the monitors worth driving to:
| Filter | Options |
|---|---|
| Reporting window | How many days of history to assess (14 by default) |
| Uptime | All, Good, Moderate, Critical, or Offline |
| Sensor error margin | All, Good, Moderate, or Critical |
| Days offline | Monitors silent for at least this long |
| Cohort | One of your Vertex groups — searchable |
| Grid | A spatial area — searchable |
| Tags | Device tags applied in Vertex |
You can work in map or list view, and collapse the sidebar for a larger canvas.
Filter on sensor error margin as well as uptime. A monitor reporting reliably but with a critical error margin won't show up in any offline list, yet it's producing data you shouldn't trust — and it needs a visit just as much as a dead one.
Planning a Route
Once the map shows the right monitors, build an itinerary. Beacon orders the stops and calculates road routes between them, showing distance and duration for each leg, so a day's work can be sequenced before anyone leaves.
Routing starts from a home location — AirQo's default is its Kampala head office — and can use your current position instead.
A practical sequence:
- Filter to Critical uptime over the last 14 days.
- Add a cohort or grid filter to keep the trip within one area.
- Build the route and check the total distance against what a day allows.
- Tighten the filters if it's too long — Days offline is the usual lever, since longest-silent monitors are the highest priority.
Exporting the Map
Export for a field team that won't have Beacon open:
| Format | Use |
|---|---|
| Enriched CSV | A spreadsheet of the filtered monitors with their health data |
| GeoJSON GIS Layer | For GIS tools, or a phone mapping app |
Include Active Maintenance Route adds the planned itinerary to the export.
LoRaWAN Gateway Coverage
The page includes a tool for modelling LoRaWAN gateway coverage against your monitors, which estimates signal attenuation by distance and reports coverage statistics. You can filter monitors by whether they fall inside gateway coverage.
Gateway definitions are stored in your browser only. They are not saved to your account, not shared with colleagues, and will be lost if you clear site data or switch machine. On first use the tool pre-populates a set of sample Kampala gateways, which are illustrative rather than a record of real infrastructure — replace them before drawing conclusions.
Treat this as a planning aid for reasoning about coverage, not as an inventory of your gateways.
Troubleshooting
The map is empty
Problem: No monitors appear.
Solution: The filters are almost certainly too narrow — combining a short reporting window with a strict uptime band and a days-offline threshold can exclude everything. Reset to All and reapply one filter at a time. Also confirm your monitors have been deployed in Vertex: an undeployed monitor has no coordinates and cannot be placed on a map.
A monitor is in the wrong place
Problem: A pin is at the wrong location.
Solution: Coordinates come from the monitor's deployment record in Vertex. Correct it there — see Deploy a Device to a Site.
Route distances look implausible
Problem: The calculated route seems wrong.
Solution: Check the home location the route starts from. Road routing also depends on an external routing service; if it's unavailable, distances may fall back to straight-line estimates, which will read short in hilly or poorly connected terrain.
My gateways disappeared
Problem: Gateways you entered are gone.
Solution: They were in browser storage, which is cleared by clearing site data, using a private window, or switching browser or machine. This is a limitation of the tool, not a fault — keep your authoritative gateway list elsewhere.
What's Next
- Alerts — What's flagged, and when.
- Reports — Produce a PDF record.
- Device Details — Diagnose before you drive.