I've been getting the following error periodically lately for several Matter Switches. They all occur at the same time. It appears something is trying to do a refresh. I don't have anything doing that. Wondering where it's coming from.
dev:28142026-02-28 08:01:47.592 AMerrororg.codehaus.groovy.runtime.metaclass.MissingMethodExceptionNoStack: No signature of method: GenericMatterSwitch.refresh() is applicable for argument types: () values: []
Possible solutions: refreshCmd(), flash(), every(), parse(java.lang.String), every(groovy.lang.Closure), grep() (method refresh)
Can't tell on this. The time on the Job changes. It appears it runs about every 40 minutes or so. I don't get the error every time. May only happen once every few days. So no way to match up time vs that job.
EDIT: If that is what is triggering the error then there must be something that says what that job is doing and what can cause the error.
Watching the Scheduled Jobs log and it appears that Matter job runs about every 10 minutes. There is also a 'periodicCleanupJob' that seems to run about every 10 minutes.
I have run out of ideas… I do have a couple of rules that do refreshes on Sonoff outlets to keep the led from flashing (distracting).
Maybe you have something similar and mistakenly included the matter devices, but that should show in the device events.
I don't have anything that does a refresh. The only thing that even runs on time is a battery check and none of these are included there.
Just wondering if someone from HE could chime in and let me know if that JOB is what could be causing the issue. Maybe it's checking or polling the devices and if for some reason they are offline at that point in time.....
It's not a big deal, I would just like to know what is causing it so I know whether to look further or not. All the switches still seem to work just fine.
EDIT: just did some checking and there were several of them that had the error in 10 minute increments for 3-4 times. So that would seem to indicate it is being caused by that JOB. So it would be nice if the error was a little more clearer as to what is wrong. Just a guess, but if it indicates the switch is offline for some reason, say that, not that the refresh command doesn't exist. Maybe something to the effect no response.
Just noticed that the Scheduled Jobs log doesn't show the IDs, have to look at the JSON detail to get them. The lack of the delete option for it is a dead give away though - regardless, checking it on my Matter hub verifies that it is an integration spawned job.
Spent some time thinking about this and have some observations/questions.
I understand there needs to be some background tasks running. I probably never paid much atttention to that before. Looking at the scheduled integration jobs most of them are self explanatory. For example there is a sunsetevent, which I assume gets the sunset time. Then there is a sunjob. I assume that gets the sunrise time???. Perhaps that should be named sunriseevent?? But I could be totally wrong on those. But maybe some documentation on those kind of tasks so when we get an error we might be able to figure out where it's coming from.
Back to the particular scheduled job that started this conversation. "MatterNodeCheckJob". My main question is, What does this do? Is it necessary to poll matter devices every 10 minutes? Kinda reminds me of those old Z-Wave devices that had to be polled to keep alive. (I have a box of those somewhere).
If those jobs are going to produce an error, please make the error something that means what the problem is. Saying the device doesn't support a refresh command doesn't really address what the issue was. Or to me it doesn't. When one spends time trying to run down an error and there is no good way to resolve it is kinda frustrating.
