Due to some relocation one of my Lumi-sensors was out of reach for a couple of days. Now it’s marked as unavailable i the gui. In the logs I can see that it is transmitting data regulary, byt HA refuses to pick it up:
[0x78F8](lumi.weather): last_seen is 337395.2020287514 seconds ago and ping attempts have been exhausted, marking the device unavailable
[0x78F8](lumi.weather): Update device availability - device available: False - new availability: False - changed: False
Are there any simple way to reset the status and make the device available again?
I have seen this problem occasionally, to rectify it I have just re-paired the device again after first checking whether it needs a new battery. After re-pairing, all the correct entity names get picked up automatically however for some reason you will need to re-select the Area.
The problem is that the device is 300 km away in my “summer”-house. I moved the sensor to the same room as the HA-node and assumed it would start working automatically before I left…
I was hoping I could change some configuration or similar to reset the status and make it available again.
I have ssh-access to the Home Assistant-node and 30 years of vi-experience. Just tell me what to do.
Yes, I’ve tried that reconfigure-option withiut success.
Sorry for not giving all information from start, but I thought the answer was something simple like “Do these three steps in the GUI” or “Change this row in xxx.yaml” or something similar. The problem should be quite common I think?
The issue is with the device. In case a battery powered device can not communicate with the network, it is supposed to stop trying for a while to save battery and not overload the network (all devices could be trying to reconnect at the same time). Normally the device should try to reconnect at “regular” intervals (I think there is a recommendation of 15 minutes in the Zigbee specifiction).
However, this behaviour very much depends on the device and I find that some will reconnect without an issue after an outage, and others have to be manually “re”-joined.
I even had a device once that did not rejoin for a very long time (at least 3 days), and when I opened the network for joining a new device, it suddenly proposed itself for rejoining which succeeded (I was sniffing at the time so I could see that it was a rejoin).
So your device could be in the latter case: it is waiting to see an open network. The network would have to be open when the device is looking for it so you would need to search for new devices repeatedly for about 15 minutes.
However, in my experience, when a device does not rejoin by itself when the network is closed, it will not rejoin without manual interaction when the network is open.
According to the logs the device still sends data regularly (I haven’t timed it, but at least once a minute), but it is dropped by HA:
[0x78F8](lumi.weather): last_seen is 337395.2020287514 seconds ago and ping attempts have been exhausted, marking the device unavailable
[0x78F8](lumi.weather): Update device availability - device available: False - new availability: False - changed: False
I guess this means that te sensor are on the network?
Actually this means that ZHA sent multiple packets to the device and that it did not get a response, that it has not communicated since 337395 seconds - or almost 4 days and that it is therefore labeled as not available.