From what I read it seems enough to just change the value in the configuration of the Thing.
I assume the people that experienced this issue and changed the value to atleast 360 or 600 seconds no longer have the error, correct?
I got a chance to do a quick check of the code and we only use a fixed interval of 60 seconds for checking the connection and connect again when the connection is no longer active.
If nobody is experiencing any issues after changing the Thing values it might still be good that I change this values to also 360 seconds but this would make it less urgent…
I can’t recall if we set the default for the Thing or if it’s a system setting, will also check that.
do you have other systems accessing the MELCloud API besides openHAB frequently (the MELCloud app, other automation systems, …)?
First I tried with 600 sec were the system worked nice again, after the MELcoud service came back.
Now the last days I tried with even longer polling-time for the device-thing. It is now 3600 sec, but a test-rule for sending me a telegram-message on each “last-communiication” is sending more unregular, but still 3-4 time per hour. I had expected 1 message each hour (3600 sec).
I am not sending any commands to the heatpump, as the desired temperatur is kept constant right now. But the room-temperatur-item is updated once in a while and more often than 3600 sec.
I have installed the MELcloud app on my Android-phone, too. But normally not using it, so that app deep-sleeing and should not generate any traffic to the cloud.
I am also facing this issue for a few days now. I am using Openhab 4.0.3.
The error messages “Login error, reason: java.lang.IllegalStateException: Expected BEGIN_OBJECT but was STRING at line 1 column 1 path $” occurs always once per minute, independently what has been set for pollingInterval in the things file. Also after modifying and rebooting.
When stopping the melcloud binding, waiting for one hour and it starting again, the error message occurs immediately.
So, the proposed workaround does not help on my side.
not sure how you configured it but there is also a connection check which is fixed at 60 seconds. This reconnects if not connected already, it’s possible this is why you get variable results.
I will probably change this to also 300 seconds.
You need to wait more than 1 hour. Some say 3 hours some say longer.
If the ban hasn’t been lifted it’s normal that you get the error immediately as we can’t login to MELCloud.
Before activating the binding again, try to connect and login to MELCloud via their website (from the same network as openHAB is running). If that works, you can enable the binding again.
I have 2 units. I was getting the login error and disabling the MelCloud account for several hours and reenabling it worked but I was locked out very quickly. Now configured 300s polling interval on each… It’s been a few hours since I made this change and still working. Thanks, everyone, for the advice!
Thank you very much, the needed procedure became clearer by your description. Now it works again
Just as a feedback. Putting the pollingInterval to 600s definitely works. Its running fine for several days now. Thank you for the information.
Initially I had set my 3 AC indoor units to poll every 600 seconds - then lowered them to 360s each. Been working for a week now.