The problem
My Roborock Zeo Mini (device model roborock.wm.a93) repeatedly becomes unavailable in Home Assistant. All 19 entities (sensors, selects, buttons, binary_sensors, switch) go to unavailable simultaneously. The device itself is confirmed online and actively running a wash cycle in the official Roborock app at the same time HA shows it as unavailable — so this is not a device-side/standby issue, and not a Wi-Fi issue.
Troubleshooting already attempted, none of which resolved it:
- Reloading the config entry
- Fully removing and re-adding the integration
- Confirmed the device is online/active in the Roborock app during the outage
What version of Home Assistant Core has the issue?
core-2026.7.2
What was the last working version of Home Assistant Core?
Unknown — first noticed on 2026.7.2
What type of installation are you running?
Home Assistant OS / Supervised (adjust if different)
Integration causing the issue
Roborock
Link to integration documentation on our website
https://www.home-assistant.io/integrations/roborock
Diagnostics information
(Attach diagnostics via Settings → Devices & Services → Roborock → ⋮ → Download diagnostics)
Anything in the logs that might be useful for us?
2026-08-06 17:18:43.466 ERROR (MainThread) [homeassistant.components.roborock.coordinator] Error fetching roborock data: Failed to update data
roborock.mqtt.roborock_session.aiomqtt: failed to receive on socket: [Errno 104] Connection reset by peer
roborock.data.code_mappings: Missing ZeoState code: 0 - defaulting to 1
This looks related to but distinct from #158294 (which was fixed in #158997 for core-2025.12.2) — I'm already on a version well past that fix, so this appears to be a separate/regression issue specific to the Zeo Mini's cloud-only MQTT connection stability.
Additional information
Device is an a01-category device (Zeo washer/dryer, cloud-only — no local API support per the integration docs). Two other Zeo devices on the same account/instance remain stable; only this one repeatedly drops.
The problem
My Roborock Zeo Mini (device model
roborock.wm.a93) repeatedly becomes unavailable in Home Assistant. All 19 entities (sensors, selects, buttons, binary_sensors, switch) go tounavailablesimultaneously. The device itself is confirmed online and actively running a wash cycle in the official Roborock app at the same time HA shows it as unavailable — so this is not a device-side/standby issue, and not a Wi-Fi issue.Troubleshooting already attempted, none of which resolved it:
What version of Home Assistant Core has the issue?
core-2026.7.2
What was the last working version of Home Assistant Core?
Unknown — first noticed on 2026.7.2
What type of installation are you running?
Home Assistant OS / Supervised (adjust if different)
Integration causing the issue
Roborock
Link to integration documentation on our website
https://www.home-assistant.io/integrations/roborock
Diagnostics information
(Attach diagnostics via Settings → Devices & Services → Roborock → ⋮ → Download diagnostics)
Anything in the logs that might be useful for us?
2026-08-06 17:18:43.466 ERROR (MainThread) [homeassistant.components.roborock.coordinator] Error fetching roborock data: Failed to update data
roborock.mqtt.roborock_session.aiomqtt: failed to receive on socket: [Errno 104] Connection reset by peer
roborock.data.code_mappings: Missing ZeoState code: 0 - defaulting to 1
This looks related to but distinct from #158294 (which was fixed in #158997 for core-2025.12.2) — I'm already on a version well past that fix, so this appears to be a separate/regression issue specific to the Zeo Mini's cloud-only MQTT connection stability.
Additional information
Device is an a01-category device (Zeo washer/dryer, cloud-only — no local API support per the integration docs). Two other Zeo devices on the same account/instance remain stable; only this one repeatedly drops.