Hi,
I have 2 smartlight air conditioner wi-fi dongles, they basically convert a dumb AC unit into a smart one: smartlight ac dongle
They work with ESPHome, they have been working fine but I notice that multiple times per day they become unavailable. They have good wifi signal and a reserved IP adress, I’m not sure what else I can do to fix this.
Any cue what could fix this? It does not affect operation but it does not look right.
Here’s a screenshot for one of them but they both do the same thing.
You could remove the “unavailable” from your graph or simply not look at it. Less than 1s unavailability is likely not worth of investigation. If it bothers your peace of mind, try to catch the event on esphome logs.
Likely this connection drop as it is immediately followed by return to last state
You could add WiFi signal strength to sensors and see if that correlates to unavailable time. Send notification with last recorded rssi when it become unavailable or state changes to help.
I had similar problem but in different context, unavailable device. But device was on Thread. After investigations turned out that WiFi was on channel 4 and was disturbing Thread. I changed channel on WiFi and all go. Try to change the channel of WiFi.
I got this error in the log, for now that’s all I got… not sure this helps.
Logger: homeassistant.components.esphome.manager
Source: components/esphome/manager.py:555
Integration: ESPHome ([documentation](https://www.home-assistant.io/integrations/esphome), [issues](https://github.com/home-assistant/core/issues?q=is%3Aissue+is%3Aopen+label%3A%22integration%3A+esphome%22))
First occurred: 1:48:06 AM (2 occurrences)
Last logged: 1:48:06 AM
* Air Conditioner 7c7ce7: [W][component:237]: Component midea.climate took a long time for an operation (63 ms).
* Air Conditioner 7c7ce7: [W][component:238]: Components should block for at most 30 ms.
Expecting to see errors for disconnection from a logging process that depends on connection is likely to lead to disappointment. You might be able to see something that happens just before the issue, but probably not exactly what the issue is (since the logging depends on actual connection being present).
They are much more detailed.
If you direct connect to device it may become more stable and not disconnections I’m not sure how to get live logs under normal function.