GTI Hackettstown NJ · Facility 3530

Mesh backbone:Red troncal: 243 of 423 devices are online and communicating in the live snapshot, so the gateway can push WP5 to them at the scheduled update. The gateway pushes WP5 to every device online at that moment; devices that are offline simply stay on WP4 (unchanged — not harmed). This facility has 6 gateways, all already on WP5 (ggVersion 2): 3 online (H2100003037, H2100004444, H2100004610); 3 are offline — H2100004362 (last seen 2026-06-16); H2100004438 (last seen 2026-04-20); H2100005494 (last seen 2025-02-01). Confirm whether the offline gateways are decommissioned before the push. The push is run remotely by AROYA; a site visit is the exception, not the default.

Your fleet is 348 substrate nodes, 25 climate stations (repeater-class radios with an Atmos-14 attached), 44 range-extender / IO radios and 6 gateways. Across the 348 substrate nodes the attached sensors split into 196 TEROS-12, 143 TEROS One (already next-gen), 5 SQ-521 PAR and 4 bare hosts with no sensor paired. The 44 range-extender radios include 10 H4210 units carrying EE894 sensors. Attached sensors ride on their host radio for the WP5 push — there is nothing to update on the sensor itself.

The gateway updates to WP5 and pushes it to every device that is online and communicating at the scheduled time (the go/no-go is: online, battery ≥ 30%, signal ≥ 50%). Devices that are offline at that moment simply stay on WP4 — they are not harmed, they just don't get the new firmware. The catch is compatibility, not damage: a WP4 device needs a WP4-matched gateway to keep communicating, so the goal is to have as many devices as possible online at the push. The update is run remotely by AROYA; keeping the fleet online is the site's responsibility.

1. P0Bring as many devices online as possible before the scheduled push. Work the "Offline" and "Likely to miss" lists — replace batteries and shore up signal so each meets online + battery ≥ 30% + signal ≥ 50%. Every device up at push time receives WP5; every device down stays on WP4.
2. P1Shore up the "At risk" list. Marginal battery/signal devices may drop out around the push window — plan battery swaps and repeater coverage now.
3. P2Decide what to retire or trade in instead of recovering. For dead/legacy sensors you won't bring back, mark them for a TEROS One / Climate One upgrade in the planner above rather than spending effort reviving them — and claim the free trade-in above first.
4. P3Confirm the gateways/sinks are online in the AROYA app. The gateway carries the push; verify the backbone (especially the silent Gateway I) before the scheduled update.
5. After the push, recover any stragglers. Devices that were offline can still be brought current later by reverting a gateway to WP4, bringing them online, re-running, then returning the gateway to WP5 — there's no time limit. Keep the mixed WP4/WP5 window short, though: mixed firmware on one network long-term can cause connectivity issues.
6. Note: existing WP4 gear keeps working as-is; new AROYA equipment requires WP5, which is the reason to update.

Method & scope. One row per physical device; an attached sensor (TEROS-12, TEROS One, Atmos-14, SQ-521 PAR, EE894) is folded into the radio it plugs into, and sensor-head serials are masked to the last four digits — locate hardware on the floor by the H-prefix radio serial only. Class by serial prefix / model (H111x/H411x = climate station, H21xx = gateway, H32xx/H34xx = substrate node, H13xx/H34xx bare = repeater). All values are from the live SPA device snapshot pulled 2026-07-14 — online, live battery (%), rssi, linkQuality (%) and lastCommunication. Gateway WP4/WP5 state is the live ggVersion from /devices/?facility=3530 (2 = WP5, 1 = WP4). This is a point-in-time snapshot, not a 24-hour telemetry roll-up: 24-hour battery/signal/link percentiles, radio travel time and >60-min gap counts are not computed from a snapshot and render “—”. Readiness reflects the WP5 go/no-go — a device receives WP5 only if it is online with battery ≥ 30% and link ≥ 50% at the scheduled push; otherwise it stays on WP4 (unharmed) and can be recovered later via a WP4 gateway revert. Priorities: P0 = act before the push, P1 = shore up, P2 = plan / trade-in, P3 = verify only. This is a device-readiness report — cultivation metrics are covered in the facility’s cultivation audits, not here.