How this audit works: the upcoming firmware update is WP4 → WP5, pushed over the air by gateway H2100002390 to every device online at the scheduled time. Go/no-go per device is online + battery ≥ 30% + signal ≥ 50% at the moment of the push. Devices that don't meet the threshold simply won't receive WP5 and stay on WP4 — they are not harmed and you can bring them along later by reverting the gateway to WP4 temporarily and re-running the push. Remote is the default; no site visit is assumed.
⚠ Telemetry pull blocked during this run. The portal_api/devices/?facility_id=2872 pull — the source of per-device firmware, online/battery/RSSI/last-comm, and the retired-pool upgrade list — could not complete because the local AROYA portal browser bridge went offline mid-swarm at ~01:44 UTC and did not recover within this agent's wall-clock budget. The verified facts in this report come from the AROYA public_api room/zone record (which uses an independent path); everything that depends on device-level telemetry is marked NEEDS TELEMETRY REFRESH. Re-run the portal_api pull within 24 h of the scheduled push to bucket every active device into green / lime / amber / red and to size the retired TEROS-12 / Atmos-14 upgrade pool against the fixed AROYA trade-in pricing ($479 each — $629 MSRP minus $150 trade-in credit).
—
GREEN — Will sail through (online + batt ≥ 50% + sig ≥ 70%)
—
YELLOW — Recoverable (above threshold but tight)
—
AMBER — Borderline (below threshold; wake/move before push)
—
RED — Won't receive WP5 (offline / dead battery)
—
Retired / upgrade-pool (TEROS One eligible)
1
WP4 gateway (upgrade target)
What we know from the public_api
11
Rooms on file (9 flower + 1 veg + 1 dry)
69
Zones / cultivation sections
1
WP4 gateway (H2100002390)
SOILLESS
Substrate type across all rooms
Bloomfield Building 1 is laid out as a uniform 9-room flower block (Room 1–9, each with 5 zones), a 24-zone Veg 1 (3-tier rack layout: tables 1–14, bottom 15–19, top 20–24), and a 3-zone Dry Room 1 (back / center / front). Total 69 zones — consistent with a mid-sized commercial cultivation footprint that would typically host on the order of 100–180 AROYA devices (1 substrate node + 1 TEROS-12 per zone in flower, plus repeaters, climate sensors, and the gateway + sink). The exact device count and per-device telemetry will be confirmed by the portal_api refresh below.
Room & zone layout
Room
Type
Substrate
Zones
Room 1
Flower (OTHER)
Soilless
5
Room 2
Flower (OTHER)
Soilless
5
Room 3
Flower (OTHER)
Soilless
5
Room 4
Flower (OTHER)
Soilless
5
Room 5
Flower (OTHER)
Soilless
5
Room 6
Flower (OTHER)
Soilless
5
Room 7
Flower (OTHER)
Soilless
5
Room 8
Flower (OTHER)
Soilless
5
Room 9
Flower (OTHER)
Soilless
5
Veg 1
Veg
Soilless
24
Dry Room 1
Dry
Soilless
3
Sibling facility — Bloomfield Building 2 (Facility 2806)
Bloomfield operates a sister facility — Bloomfield Building 2 (Facility 2806, gateway H2100002306) — audited in parallel as part of this same WP4 firmware tranche. Both buildings show the same Addium-internal correspondence pattern (no external POC surfaced from the email pull). Recommendation: coordinate the WP4 → WP5 push for Building 1 and Building 2 in a single window so the operator only has to walk one go/no-go sequence and the rollback plan is shared between both gateways. The shipped Building 2 report lives at https://ciansullivan-hue.github.io/aroya-reports/Bloomfield_Building_2_Firmware_Readiness.html (publish time and exact URL depend on the Building 2 agent's completion).
Recommended firmware-push sequence
T-7 days · Notify operations. Confirm the push date with the Bloomfield grow team for both buildings together. Remind them that any node deliberately powered down or out-of-mesh at push time stays on WP4 and can be migrated later — there's no irreversible window. Offline devices are not harmed.
T-24 h · Telemetry refresh (BLOCKING for this audit). Pull live battery / signal / last-comm via portal_api /devices/?facility_id=2872&retired=false AND &retired=true for every device, bucket into green / lime / amber / red, and surface any node below the 30%-battery or 50%-signal go/no-go threshold so the on-site team can wake / re-seat / move the repeater before the push.
T-2 h · Mesh prep. Confirm the repeater (H11-prefix) ring is fully online and that the sink + gateway H2100002390 are reachable. Repeaters carry the WP5 image to the nodes and update first.
T-0 · WP4 → WP5 push from gateway H2100002390. Each online device picks up WP5 within the push window. Anything offline simply stays on WP4 — no damage, no warranty issue, no time-bounded recovery window.
T+24 h · Verify. Pull device list a second time. Any device still on WP4 that should have updated → check it's online + batt + sig; if it's healthy and stuck, re-issue the push to it. Anything still offline → flag for retire-trade-in or bring it online before re-running.
If rollback is needed: revert gateway H2100002390 to WP4 from the AROYA portal and the entire mesh rolls back. There is no time limit on the rollback.
Sensor upgrade planner — TEROS One & Climate One
Pricing is fixed at the standard AROYA upgrade tariff:
$419
TEROS One Single — $419 MSRP minus $100 trade-in credit per legacy TEROS-12
$419
Climate One — $629 MSRP minus $150 trade-in credit per legacy Atmos-14
—
Retired TEROS-12 (eligible for TEROS One trade-in)
—
Retired Atmos-14 (eligible for Climate One trade-in)
Retired-pool sizing requires the portal_api/devices/?facility_id=2872&retired=true pull. Once that lands, every retired atm14 / ter12 whose host puck node is also flagged retired becomes an upgrade-eligible row in the device table below. Tick the Upgrade? column to build the order list. Quote totals and unit selection (Single vs Dual TEROS One) persist live across every device that opens this report URL.
Per-device table — live cross-device sync
0shown
Upgrade ticked: 0
TEROS One: 0
Climate One: 0
Quote running total: $0
Live sync: …
Serial
Model
Room
Zone
Bucket
Why / what to do
Upgrade?
Dual / Single quote
Status
Per-device rows pending telemetry refresh. The table above is wired up for cross-device sync, filter, sort, and the Upgrade?/Dual/Single quote total — but it currently shows only the WP4 gateway because the portal_api active+retired device pulls were blocked by the browser-bridge outage. After the operator re-runs opencli aroya raw --path '/portal_api/devices/?facility_id=2872&retired=false&pagination=0' and --retired=true the U[] array at the bottom of this file can be re-populated and the same URL will continue to hold operator state for every device.
Data quality: verified from AROYA public_api /rooms/?facility=2872 (rooms + zones) on 2026-06-04 and from the canonical WP4 Gateways sheet (gateway serial). Live battery / signal / last-comm telemetry and the retired-pool count were not pulled because the local AROYA portal browser bridge disconnected mid-swarm. Re-run the portal_api pull within 24 h of the push. Firmware language: WP4 → WP5 OTAP. Offline devices stay on WP4, are not harmed, and can be brought along later via a temporary gateway revert. Remote push; no on-site visit assumed. State sync: every check / dropdown you set in the table above is saved to aroya-report-sync.ciansul.workers.dev/2872 within 1 s and replays on any other device that opens this URL. Use the Export button to download a JSON snapshot for the change record.