Hi all, I noticed there’s ST scene execution issues lately in rule engine, where actions running specific scene to change light settings did not happen.
More specifically, from the rule execution logs it looked fine, run scene action was performed and no error shown, but lights did not change. Wheres running that scene in Smartthings app works just fine.
Is there any known issues invoking ST scenes from rule engine? What’s strange is the rule was working fine until about this February when it started to act finicky, and some days it just worked. Also there’s no change on the rule.
No, there aren’t any known issues and nothing has changed with Scene execution from the SharpTools side of things recently.
That being said, if the scenes are showing that they were executed successfully from the rule logs, that indicates that the SmartThings API is accepting the request and replying as if everything is fine, so it sounds like it could be an issue somewhere within the SmartThings platform upstream from the SmartThings API.
When the scene is actuated from your rule, but the devices don’t seem to change status, what do your SmartThings History logs show? (SmartThings Mobile App → ☰ → History)
For example, when I actuate a Scene from SharpTools or the SmartThings mobile app, I see it in the History:
Thanks Josh for the suggestion, indeed I do not see scene execution log form ST app, that indicates ST API did not execute the request.
Would it be possible that there are multiple requests going through at that time from my side that my requests to ST are being throttled? You could see my rule has multiple commands to change lights, and there’s another sharptool rule triggered at 11pm to update thermostat setpoint.
Are you seeing the log entry when you execute a Scene from the SmartThings mobile app? Or if you execute the scene from a SharpTools dashboard or when your rule executes the scene as expected?
In theory, yes, sending to many commands at once could result in API rate limiting. But when that happens, the SmartThings API responds with an error and the rule logs would have shown the error.
That and SmartThings rate-limits the ‘Scenes’ API endpoint separately from the ‘Devices’ command endpoint. And the Scenes rate limit is something like 50 executions every minute.
Based on the logs, it looks like the API request is going through, SmartThings is accepting it and replying that everything is ok, but it’s getting dropped somewhere within SmartThings’ systems after that.
Has this issue happened multiple times or was this a one-off event?
For all other scene & routine activations, ST app log will show they activated. Only for this rule’s scene execution was not logged and executed.
This issue is happening since this month, but not every night (it executes every night at 11pm). Other rules executing scenes work perfectly fine (not this scene though). Would it be possible that something’s wrong with this specific scene on ST end?
I’d duplicate this scene and let the rule execute the duplicate tonight and see.
