ThingUpdatedEvent starts every hour

Hi.
I’ve updated (to 2.5.2 version) my OH system based on:

  • Raspberry Pi 3 B+
  • Z-Wave Aeotec Z-Stick Gen5 (ZW090) - UZB
  • Aeotec ZW100 Multisensor 6

Updating to 2.5 (re-installation) I’m noticed a strange behavior: every (about) 3600 sec. OH receive a “ThingUpdatedEvent” which causes all scripts to restart (!) … so, any rule initialization action designed to be performed at system startup, starts every hour
These are logs:

...
2020-03-07 10:18:35.751 [me.event.ThingUpdatedEvent] - Thing 'zwave:device:controller:node9' has been updated.
...
2020-03-07 10:18:48.139 [INFO ] [se.smarthome.model.script.AlarmCheck] - *** START alarm.rules
2020-03-07 10:18:48.450 [INFO ] [smarthome.model.script.PresenceCheck] - *** START presence.rules
...

This could be a problem …
Any idea ?
Thanks.

Comment only; there are a number of threads describing this kind of behaviour since 2.5. It’s not at all clear to me what is the cause and effect behind “Thing updated” triggering “System started” rules The other threads have usually followed the path of getting rid of the Thing updated happening.

In this case, a zwave node. Maybe a zwave database change involved, I would try deleting the Thing and re-discovering, but I regard that as circumvention.
There’s something funny going on in 2.5

1 Like

Thanks for the contribution.
I’ve just tried to delete and re-discover the Thing … and to re-install entire system … obtaining the same (no) results … :frowning:

Reinstalling everything is a bit rash, but never mind.

So you still have the same individual Thing giving trouble?
I guess the next step is a debug log from the binding, try to find out why it updates the Thing.
(That’s abnormal outside of configuring activity)

Are you configuring from xxx.things files, or allowing discovery via PaperUI ?

The strange thing is that, while when a rule is edited only its script is restarted, in this case all rules scripts are restarted …
Yes, I’m using the PaperUI discovery …
However, analyzing the log, it seems that the thing update (… which causes all scripts restart …) occurs when the sensor sends data to ZWave controller:

2020-03-08 09:54:27.369 [vent.ItemStateChangedEvent] - zwave_device_controller_node9_sensor_temperature changed from 20.2 °C to 20.3 °C
2020-03-08 09:54:27.789 [vent.ItemStateChangedEvent] - zwave_device_controller_node9_sensor_relhumidity changed from 56 to 54
2020-03-08 09:54:33.533 [vent.ItemStateChangedEvent] - zwave_device_controller_node9_sensor_luminance changed from 4 to 18
2020-03-08 09:54:33.887 [me.event.ThingUpdatedEvent] - Thing ‘zwave:device:controller:node9’ has been updated.
2020-03-08 09:54:34.432 [vent.ItemStateChangedEvent] - zwave_device_controller_node9_sensor_temperature changed from NULL to 20.3 °C
2020-03-08 09:54:37.364 [vent.ItemStateChangedEvent] - zwave_device_controller_node9_sensor_ultraviolet changed from NULL to 0
2020-03-08 09:54:40.212 [vent.ItemStateChangedEvent] - zwave_device_controller_node9_sensor_luminance changed from NULL to 17
2020-03-08 09:54:40.947 [vent.ItemStateChangedEvent] - zwave_device_controller_node9_sensor_relhumidity changed from NULL to 54
2020-03-08 09:54:42.445 [vent.ItemStateChangedEvent] - zwave_device_controller_node9_battery_level changed from NULL to 50

Zwave Things updating every hour is normal. As battery devices wake up, there could be a poll or heal in the queue that pick up where it left off from the last time it woke up. You could see this with a zwave debug log. Rules made with the legacy rule engine restarting after Things are updated has already been reported as an issue in Github and there are plenty of topics in the forum about this issue. However, since openHAB core in 2.5 is locked, it is extremely unlikely that there would be a fix until OH 3.0.

This issue does not affect the new rule engine and scripted automation.

Thanks for your info …

OK, I read those posts and now … I feel less alone :slight_smile:

I think it is a serious dysfunction …