(COMPACT) setState not processed because States database not connected
Opened this issue · 12 comments
Describe the bug
Seeing this entry in my logs multiple times.
To Reproduce
Steps to reproduce the behavior:
I don't know, how to reproduce this.
Expected behavior
Should not occur as database seems to be ok.
Screenshots & Logfiles
Versions:
- hm-rpc version: 1.15.19
- hm-rega version: 4.0.0
- JS-Controller version: 5.0.17
- Node version: 18.19.0
- Operating system: raspberry pi os bullseye
- CCU model: debmatic
- CCU firmware: 3.73.9.112
Additional context
The messages are still being logged, although I stopped the hm adapters! 🤔
I am using compact mode
JS-Controller version: ?
please specify, it is essential for this ticket. You can find it e.g. in hosts tab in admin or in iob diag
5.0.17
Didn't find it before.
Okay. Can you put Adapter to debug log and Show log of stoppage until Error occurs
I set the adapter to debug an restarted it. At 2023-12-24 19:58:42.990 I stopped the adapter again. As you can see, the error messages continue, although the adapter is stopped.
iobroker.current.log
I cannot really replicate this. Tried a compact group with two hm-rpc instances, stopping one manually. The stopped one has no reaction to state changes etc anymore. Also code-wise it looks like all timers are clean.
Is this reproduceable every time, also when whole group has been restarted? Group running together with controller?
The group is set to 'default' (seeing this in the instance view). When I click on the 'edit' symbol it shows me this:
then I clicked 'cancel' and afterwards 'edit' again. Now I get this:
So I don't know, if it is running with controller. I thought, I set all my 'unproblematic, seldom-have-to-restart' adapters like the homematic ones to 'default group'.
I restarted all three homematic adapters, but not the whole 'default group'. I will try this later on. BTW: I always thought that restarting a single adapter in a group causes the others to be restarted too. I read this some time ago so this might has been changed in the meantime.
After stopping an adapter reloading the log view often reproduces this message.
I stopped one adapter after the other (of default group). The message keeps coming in. As I stopped vis.0 as the last one (except for admin) the log shows
Do not restart compact group controller 1 because no instances assigned to him
And the message stops reappearing. (Waited some minutes).
I restarted all instances of default group (except admin which I didn't stop before). The message doesn't come back. I think it will reoccur in a few hours or days. At least that's what it did all the time before.
I am currently trying to reproduce this with adapting logging, unfortunately the current log does not output which state the adapter tries to set. We should add this to a following controller version, but will not help for now.
you can re-try with 1.16.0 as soon as available, there was a small change, that code intervals and timers are executed after adapter terminate.
I will try when it is released. Thank you very much and Merry Christmas!