Easiest automation not working (and it was fine for weeks)

Did Samsung again mess something up? Support is utterly clue- and useless and could not give any input besides “delete the automation and re-create it” (which, as stupid as it is, I did with - as expected - no different results…)

I have a straight forward automation set up in the new app on the android phone.
If Motion Sensors - Motion detected*
Turn Lights on*
Set Dimmer 100%*
Auto turn off after 2 minutes*

It turns the light on as it should and also sets the dimmer. However, there is no automatic turning the lights off. The same automation worked just fine for weeks, so I wonder if Samsung’s experts messed things up again… anybody else having similar problems?

Log for the sensor:
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎24‎:‎56‎ ‎PM: info LWA Access and Refresh tokens updated and valid
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎24‎:‎56‎ ‎PM: debug requestToken
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎22‎:‎18‎ ‎PM: info postEventToEndpoint: event successfully posted.
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎22‎:‎18‎ ‎PM: debug Property Changed event motion: inactive (source: DEVICE)
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎22‎:‎18‎ ‎PM: debug routeEndpointEvent event name: motion, physical?: false
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎21‎:‎57‎ ‎PM: info postEventToEndpoint: event successfully posted.
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎21‎:‎57‎ ‎PM: debug Property Changed event switch: on (source: DEVICE)
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎21‎:‎57‎ ‎PM: debug routeEndpointEvent event name: switch, physical?: false
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎21‎:‎57‎ ‎PM: info postEventToEndpoint: event successfully posted.
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎21‎:‎57‎ ‎PM: debug Property Changed event motion: active (source: DEVICE)
6711ade2-cf7c-480a-90b3-f0a30b4a21d3 ‎1‎:‎21‎:‎57‎ ‎PM: debug routeEndpointEvent event name: motion, physical?: false

Log of the light switch:
d0e16ab5-b449-43dd-9ddf-104d777fa097 ‎1‎:‎21‎:‎57‎ ‎PM: debug Level changed to 100
d0e16ab5-b449-43dd-9ddf-104d777fa097 ‎1‎:‎21‎:‎57‎ ‎PM: debug Status 201 100
d0e16ab5-b449-43dd-9ddf-104d777fa097 ‎1‎:‎21‎:‎57‎ ‎PM: debug Data=[Body:[ZoneStatus:[Level:100, Zone:[href:/zone/2], href:/zone/2/status]], Header:[StatusCode:201 Created, MessageBodyType:OneZoneStatus, Url:/zone/2/commandprocessor], CommuniqueType:CreateResponse] CreateResponse
d0e16ab5-b449-43dd-9ddf-104d777fa097 ‎1‎:‎21‎:‎57‎ ‎PM: debug commandHandler (200)
d0e16ab5-b449-43dd-9ddf-104d777fa097 ‎1‎:‎21‎:‎57‎ ‎PM: debug sendLevel() 100 on
d0e16ab5-b449-43dd-9ddf-104d777fa097 ‎1‎:‎21‎:‎57‎ ‎PM: debug on

I am having similar problems. I have a motion sensor. When it senses motion, it should turn a particular light dimmer on @ 20%.

I can see the events take place in the smartthings IDE log, I can see the event to trigger the light to come on get fired. But nadda happens.

I’m wondering if this is because I have a V1 hub and should migrate to a v2 or greater (not keeping up with the hardware releases so I don’t know what is latest)

Mike

There’ve been a couple of other posts about this lately, sounds like maybe something’s broken on the cloud side:

1 Like

I’m seeing it happen here too. It’s not consistent with any single device (I am seeing it randomly) around the house with many devices and automations. Some of these automations and devices have been working great together for a couple of years. This started 2-3 weeks ago, I think. It seems to clear itself up for a few days and then I’ll see it occurring again for a few days.

I wonder if it’s hackery? I’ve had smart receptacles mysteriously turn themselves on. For example, I have a 100’ gutter heat strip that I’d bet ran for all of last summer because I didn’t know that the receptacle that it was plugged into got turned on. Since then, I set up a notification to alert me when it’s turned on. About once a month, it gets turned on. There is no automation connected to this particular device. It’s a manual switch (in the iOS app).

Also, I wonder if it could have something to do with DOS attacks being perpetrated on the platform?

Yes, I had something similar “unexpected” happening and have one routine that “resets” everything in the morning (pretty much at 4 am switch everything off…).

I doubt that they got hacked and assume the biggest traffic they get is the one from all users connecting to their cloud environment… traffic wise most likely very close to a DDOS attack…

I have one automation that uses power allowances on a pocket socket outlet and a virtual switch to cycle the socket on for a few minutes each hour.

It failed and I didn’t notice. Once I kicked it off again, it’s been fine.

I suspect a hub reboot after a firmware update but don’t have solid data to back up the suspicion

Is this automation still failing?

I tried to reproduce and was unable to do so.

I have not checked. The workaround with two independent automation triggered at the same time, one switching the light on, the second automation switching it off with a delayed action works like a charme…

The problem was going on for several days and as there was pretty much zero support from Samsung in this regard and the first level support was pretty much a waste of time, I decided not to put any more efforts into troubleshooting it.

Edit: I did a quick test and it is working again.

1 Like

Mine is still failing

Do you have a support ticket or more info?

From your previous post, it isn’t clear if you are seeing a custom automation that uses a time based “follow up” action fail to complete that follow up.

My setup works fine with a Samsung motion sensor and a GE switch. Turns off 5 minutes after no motion. Maybe try it without the dimming option. Just turn the light on and off.

Hi Brad,

I don’t have a ticket in for it.

Yes, this is a custom automation.

Basically, one of two motion sensors trigger a dimmer switch. If it’s between sundown and sunrise (with some offsets) The light turns on and remains on at 20% for 4 minutes.

Thanks for your interest in helping,

Regards,

Mike

I can now confirm that a hub reboot cancels power allowance timers.

I have an automation that uses a real and a virtual device with power allowance timers to cycle the real device off and on. After the hub upgraded last night, the timer was gone and the device stayed in one state until I kickstarted it.