One agent per store, checking every device at the checkout.
The agent runs inside the store network, checks each device the way it should be checked, and reports results to a central board. Here is what it checks, how alerts work and what you can export.
Set up a store in under an hour
- 01
Add your stores
Import a CSV of stores with address, region, time zone and opening hours, or add them one at a time.
- 02
Install one agent per store
Run the installer on an always-on Windows 11 or Linux machine in the back office. It registers with a one-time store token and needs outbound HTTPS only.
- 03
Confirm devices and alert routes
The agent lists what answers on the store subnet. Name each device, assign it to a lane, and choose who hears about failures.
$ sudo ./mvr-agent install --store-token st_9f2c41Registering store Pasadena #12 ...Outbound HTTPS 443 reachableScanning 10.12.3.0/24 for devicesFound 14 devices: 6 POS, 6 card terminals, 2 printersAgent running as service, check interval 60 s
Checks by device type
Each device gets the check that proves it is working, not just powered on.
| Check | Typical devices | What it records |
|---|---|---|
| ICMP ping | POS stations, card terminals, access points | Reachability, round-trip time, packet loss |
| TCP port | Receipt and label printers (9100), POS services (443) | Port open, connect time |
| HTTPS health page | Router admin pages, local POS servers | Status code, response time, TLS expiry |
| SNMP v2c / v3 | Routers, switches, firewalls | Interface up/down, WAN errors, uptime counter |
| Internet path | Primary ISP and LTE backup | Active path, latency and loss to public endpoints |
- Stores
- Lanes
- Network
- Alerts
- Reports
Incidents
Trading hours only
New1
-
Torrance #21 · Receipt printer
Down 3 min Lane 2
Acknowledged2
-
Pasadena #12 · Card terminal
Down 14 min Region West -
Glendale #07 · Primary ISP
On LTE ISP ticket open
Resolved today1
-
Burbank #03 · Access point
Back up 22 min
Alerts that respect opening hours
- Three-strike rule. A device must fail three checks in a row before an alert goes out, so one dropped ping does not wake anyone.
- Per-store hours. Each store keeps its own time zone and opening hours. Outages while closed go into a morning summary.
- Routing by store or region. Send alerts to the store manager, the regional lead, your IT contact, or all three.
- Email and webhooks. Alerts go out by email or as JSON to any HTTPS webhook, including Slack and Microsoft Teams incoming webhooks.
- Acknowledge and close. Whoever picks it up acknowledges the incident, and it closes automatically when the device answers again.
Reports and exports
Weekly and monthly reports count uptime only while stores are open. Rank stores, compare device models and ISPs, and take the numbers into vendor reviews.
- CSV and JSON export of devices, checks and incidents
- Check history kept 30 days on Store, 13 months on Operations and Enterprise
- Scheduled report emails to regional leads every Monday
- Stores
- Lanes
- Network
- Alerts
- Reports
Weekly lane uptime
Week 43
| Store | Uptime | Incidents | Trend |
|---|---|---|---|
| Inland #31 Region Inland | 96.8% | 7 | Worse |
| Pasadena #12 Region West | 98.9% | 3 | Flat |
| Torrance #21 Region South | 99.4% | 2 | Better |
| Burbank #03 Region West | 100% | 0 | Better |