For Some reason my Danfoss is eating batteries in a few days. After researching the problem I found on my Debug Log a problem which seems strange to me. Device begins with not reason (apparently) to wake every few seconds, instead of 900.
I am using the latest Z-wave binding from cd Jackson.
I suspected network healing, but this is not the reason. Device works after I set a new wakeup frequency for about 3 hours and then switches to crazy mode again.
Openhab: 2.4.0M4
Z-wave binding : latest from Jackson’s topic
The latest version is in the snapshot builds, but milestone M4 is not far behind. You should probably uninstall the binding (through Karaf), delete the jar, and install through Paper UI, Habmin, or Karaf. Then see if your issue persists.
Which version of the binding are you currently running? In your other post, you mention a snapshot version. In the Karaf console, run list -s | grep zwave, and post the result. Have you manually woken the device?
Note that the binding will not poll the device unless the device wakes up. So, if you have the wakeup period set reasonably long, then this should not really be an issue. If the device is really waking every few seconds, then you need to change the devices configuration - not the binding polling period.
I’m not sure why the binding should poll every few seconds either though - unless you have the polling period set very low. If you have a debug log that shows the behaviour, then I’m happy to have a look at it.
Ah - ok - that makes more sense . If the wakeup is changing, I would keep an eye on the logs to see if something is changing it - eg maybe there is an issue with the UI and it is sending the period to a low value?
Thank you, I couldn’t pin point something like that! but you can read them better than me.
The strange part, this happens always about 1-2 hours after I set the wake value, without having any interface running. Maybe the device waits every time to update it’s ‘wakeup interval’?
This definitely looks like a device config issue. The binding is not sending anything to the device other than the “go back to sleep” command, but the device seems to be waking up every 8 seconds.
My suggestion would be to set the wakeup period to something “normal” (normally I suggest 1 hour, but for this type of device you probably want something in the 5 to 15 minute range so that it has a better response for thermostat updates). Keep debug logging running, and hopefully, these updates should come in at the rate you have programmed. You then want to catch the point in the log where the device starts sending wakeup requests at the 8 second rate and then check the log between the previous couple of wakeups to see if the binding sent anything to change the wakeup period.
If there was a command sent, then we can try and work out where it came from. If the device is changing by itself, then I’m not sure I have a good suggestion on how to proceed .
I had a look at the log around this time and there is nothing that the binding is sending to change the device configuration. The only commands I can see at all are the “Go back to sleep” commands that the binding is sending when the device wakes. Either something else is changing the device configuration (which is pretty unlikely) or the device is changing config by itself.
I’m not sure what else to suggest really (sorry).
One interesting point in here is the device reports that it can’t be set to a value below 60 seconds. This leads me to believe that the device has a problem as if something external was setting it to 8 seconds, it should reject it. So -:
Either -:
the device has a problem in that it’s allowing other devices to set the wakeup to below 60 seconds, AND, another device is setting the wakeup to 8 seconds
or
the device has a problem is is setting itself to 8 seconds due to a software or hardware issue
I’m not sure if it is supported, but this is not something that you can enable - these are reports from the device on what features it supports. If the device doesn’t support it, you can’t change this.
. If the wakeup is changing, I would keep an eye on the logs to see if something is changing it - eg maybe there is an issue with the UI and it is sending the period to a low value?
.