This object is in archive! 

Devices shift on/off by themselves

Stephane Rochette shared this problem 3 years ago
Known

Since one hour I have 5 devices whom switch on/off themselves and roller going down and up without my request.


Has the server been hacked?

Comments (47)

photo
1

seems to have happened to me too.


Garage door found open in the morning. There was no command to open it on the logs.

photo
1

Same thing happens to me. Some devices (random brand and random location) turns on/off. I also have door/window sensor indicating wrong state and then goes back again.


Strange...

photo
1

Have you still the same problem today ?

photo
1

No problem last two days. I'll wait and see what happens... Does your problem remains?

photo
1

Yes. I got that problem with all my switchs and roller device. I'm obliged to unplug all of them. Even after exlusion/inclusion the problem remains.

photo
1

This night a fibaro smoke sensor changed state to smoke and set off the alarm... not a fun way to wake up (found no smoke).

photo
1

Hmmm. I have found my front light on after a couple of nights in the morning. As I have a rule that turns off the light after 1 minute of non activity it is impossible to keep the light on if outside is a movement and we would not turn the light on and forget to turn it off. So perhaps I might have experienced this issue but only with one light.

photo
photo
1

I have a similar problem. My Aeon Labs GEN5 Wall Plugs turn back on 1-2 mins after turned off by a rule. I have 5 Wall plugs and typically I see two or three such events per week. The incorrect commands are also in the logs.

I have opened a ticket, but seems that nothing is going to happen with this. I received a response from tech support that they are not able to replicate my findings. This problem started a few month ago after I upgraded to 1.1.38 fw.


Right now I had to add extra checks to my rules to ensure that devices that are supposed to be off do indeed stay off.

photo
1

This night it happened again. Run a scene and another device not included in that scene changed state at the same time.

photo
1

What device was it? Zipato double switches might have issue with current causing autonomous switching. Needs ota update to latest firmware.

photo
2

How to update ? my switches is ph-pan04.eu .

photo
photo
1

This time it was a Fibaro FGS222 Double Relay.


It was only one of the two relays that switched on.

photo
1

Same problem here. I created a ticket, but no solution right now.

Since this morning, I try to use the "newest beta" firmware.


BTW: I'm using PAN06-1B dual wall relay....

photo
1

Having the same issue with some plugs.

No single rule is pointed to the devices and I never switched one of this plugs manually as I use the plugs just for metering the power consumption.

0810c006adfc817532446f4e6a494431

8d5837b48d2e23ee4da3ff232c8c8fa9

e3208454ce91dab294e257ba7a686f8b

photo
1

Two different devices at different time turned on without input from me.


Strange...

photo
1

same here, dish washer and dryer turned off by themselves...

photo
1

The new beta firmware don't change anything....

photo
1

This night another device turned on without input from me...

photo
2

If this is true, it is totally exceptionable!

photo
2

I had to same problem again with one of my Aeon Plugs this morning. A rule turned it off normally and and 2.5 mins later it was turned on without any reason. Have seen this happening many times before. I have opened a ticked few weeks ago. After an initial check by the helpdesk team there has not been any progress in resolving this issue.


At this point I am not able to fully trust that the devices are on or off as they should according to my rules. I have already modified some of my critical rules to include multiple failsafe scenarios to deal with this uncertainty.


As this type of behavior is indeed unacceptable, I have started to consider alternative platforms. The only thing holding me back is the amount of work transferring all the rules to a new system.

photo
1

Is your rule to complex?

photo
1

Nope, this particular plug is affected by very simple rules, no if's or special conditions, just scheduler/astro rules. There are 4 of them, but they work sequentially. See the attached picture. The problem in this particular case is that after rule 4 (astro off command) the plug just turns on with any particular reason.


I have other plugs that are also misbehaving now and then. The common theme is that after 1 to 3 minutes (time varies somewhat) following the off command the plug just turns itself on again and there is nothing in any of the rules that would explain this.

photo
1

I have also had problems, in my case with Duwi on/off sockets. I have never been able to determine whether the problem is with the Zipabox, or some other interference from somewhere that might be causing the socket to trigger. The effect is that the sockets will randomly go off, but while still appearing to be in the "on" state. That is, as far as the Zipabox is concerned, the socket is still on. And to turn it back on, you must first turn it properly off, and then on again. So it's not really fully off. But it behaves as if it is. Very odd.


I have a lot of TKB sockets, and they have never given me a problem.

photo
2

I have, at two or three occations during the last two weeks noticed that a Duwi switch was switched on without a reason. This particular Duwi switch is an IP44 switch controlling the outlet to the engine & interior heater for the car. Those units consumes alot of energy so expensive units to have switched on for hours without a reason...


No cluster, simple zipatile.

photo
1

This is the ghost in the system...

photo
1

This night another device turned off without input from me or rule...

photo
1

Does anyone know why ph-pan04.eu modules shut down by themselves? I turn on the lamps through the mechanical switch and they stay on until I turn them off, but when I turn on the lights by the smartphone application or by the computer or even by IFTTT rule, they turn off by themselves after a while. Sometimes in less than 1 minute, another 5 minutes and sometimes 10 minutes.

photo
photo
1

hmmm................ I have 2 Lightwaverf sockets which switched themselves on a couple of times now. I was thinking maybe something interfered and did it........ but now it might be related to this I think.... weird.

photo
1

Today the two plugs to which my router and my server are connected were turned off. I have no rules at all using them, and i have not turned them off myself, so it really is a ghost (aka bug) somewhere...


Who has set this to "known", any official response from Zipato to this ?

photo
1

I´m not sure if my problem is connected to your problems but I had a similar situation. I switched off lights via an virtual switch and sometimes the lights went on or off again, sometimes with a considerable delay. Only devices in that rule were affected but the results were unpredictable. Most of the time everything worked, then all of a sudden lights went on or off again or the whole rule didn´t work as expected. I now changed the rule from the left side to the rule on the right side and until then I had no more problems. Very strange because I have many rules like the right one. All of them run flawlessly but they are all considerable shorter or without extra when pieces.... Haven´t figured it out but I hope that was the reason for my devices to go on and off respectivly.

photo
1

I wonder if it's possible for the rule on the left to be triggered before the virtual switch state change to "on" has fully propagated through the system. We know that virtual switch states don't get updated immediately. For example, I have some rules which are triggered by a virtual switch going to the "on" state. When the rule is triggered, I want to turn the switch back off. But I find that, in order to do this, I have to insert a wait of 2 seconds before the "off" action, otherwise the switch won't go off. In fact, I added a second "off" action at the end of the rule just to be absolutely sure. Perhaps the left-hand rule, with a 2-second wait between the "when" and the "if", might work correctly. You could also try "Refresh all", in case the rule somehow fails to receive the updated state.

photo
1

Hi David,

thanks a lot, I didn´t know about the delay when using virtual switches.


Until now the left rule (sorry, I added the wrong picture, the rule which runs flawless is the left one not the right one) runs without problems. And all other rules which are buildt as the the right one are ok as along as there are not too many other tiles inside. With only a "Toggle state" they have no problem at all. Quite confusing. If I experience new problems with the left one, I will try your proposal with a wait of one or two seconds (even though waiting for a light to go on is not what I want to see ;-(() But perhaps Vedran will find a solution.

photo
1

Ok. I think I've seen that advice before on the forum: write your rule to be triggered by "any" change to the device, then test for the state you want, rather than directly making the desired state the trigger. I don't remember why that advice was given.

photo
photo
1

Hello Everyone,


This issue is currently under investigation, we will notify you all trough your tickets about our progress.


Cheers!

V

photo
2

I have the same issue today two times with three different Fibaro Universal Sensor FGBS321.

All sensors switch the status about 5 minutes behind each other.


Best Regards

Peter

photo
1

Is this a isolated Zipatile issue?

(I am worried since I just upgraded my zipabox to 1.1.38....)

photo
2

Not restricted to Zipatile - have the same problem with my zipatobox

photo
photo
1

Hi everyone. I now have some more clues to this...at the beginning I thought it is only single devices that switch not linked to rules but it is way more. Let me explain in detail although it might be long story. So the bottom line it is not only devices affected but it appears that also rules...or the rules are then trigger (fully or partially) if the triggering device is triggering...

it all started with my partner reminding me to not forget to turn off the light in the night...funny, I thought I'm aging and starting to forget things...but it is not (thanks god its not me...). it seems that it was always the outside light (and it is happening now for over a week), but now I have experiences not only light but nearly (not fully) a whole rule firing without no input or triggering action.

My alarm is linked to 2pc keyfob which were in the porch. I was sitting with my partner in the living room watching TV when the zipatile beeped (the same as when alarm disarms) and my onkyo radio started to play music. This is a tipical behaviour when someone disarms the system, but noone was touching or using any app, keyfob, nothing at that moment.

I have attached my alarm rule that executes when someone arming or disarming the system. You need to follow the rule from the part when alarm is disarming - and here is my log of events:

22:03:11 - alarm disarmed (on its own, without rigger)

22:03:11 - rgb colors of the zipatile buttons changed to green

22:03:11 - my onkyo radio turned on

22:03:22 - the zipatile internal siren that is part of the alarm as a squawk device beeped (because it is a squawk device is not listed in the rules).

Now strangely none of the lights inside of the if/else puzzle did turn at all.


The above means that this random switching is more than only zwave devices. Also the virtual (and possibly all) devices are affect as in this case my virtual alarm pro disarmed without an input, however not the whole rule did execute which is even worse as shows more unreliability...:(


I really hope this will help zipato and Vedran to finally decode the message...becuse this is not acceptable for a system that is now lot of years old and still...sorry zipato, it is a flustration that talks right now.

photo
1

So Atilla, what kind of device is the alarm (the one that triggered by itself)?


After talking with Vedran the other day I looked at frequency conflicts with Z-wave devices and saw that one of the 4G bands conflicts with it. So I wonder whether that might explain the triggering of Z-wave devices.


I have a Sonos speaker plugged into the Z-wave socket, and I've had problems with both. The speaker used to get lots of interference (the sound would drop out) and the socket would switch off (as I explained in a post above). Both problems have become less frequent at pretty much the same time, even though the devices operate at different frequencies. My theory is that, in both cases, adding more devices has made both the SonosNet and Z-wave networks denser and thus more reliable, more resistant to interference. I added a second Sonos speaker in the same place (in the kitchen) on the opposite wall, also with its own socket. I have also added more Z-wave sockets in other places. Since then the socket has never turned off by itself, and I get very little interference on the speaker.


I'm now testing my theory by unplugging the new kitchen speaker (and removing the socket), as well as removing another Z-wave socket elsewhere. Already I've noticed that the Sonos interference has got worse (presumably due to the neighbours' wi-fi). Let's see if the socket also triggers by itself.

photo
1

David, the most troubling this is that the alarm is virtual. My Virtual alarm got disarmed without any trigger...and then the rule started...I was sitting in the living room with my partner and then suddently the onkyo radio turned on, changed source,....did all that it's required to do when the onkyo switch is turned on. And if you look into the rule you can see that the Onkyo and also the RGB did react to this disarm. Even my squaws did beep...but not the lights...

So no interference there. This is why I'm shocked...

photo
1

Oh. That's weird.

photo
1

One thing that makes these problems harder to diagnose is that the log doesn't appear to be 100% reliable. I don't think it logs every event, or at least sometimes there are big delays before an event is logged. And the "latest" view doesn't always show the same events for a device as the device's individual log. Has anyone else seen this?

Obviously we should always first suspect that our rules are at fault. But for that it would be nice to have a more dependable log.

photo
1

Well for my case with the alarm there are in total 2 weird things:

1. that the alarm itself disarmed

2. the lights did not switch as per the rule


This makes it even more weirder.

photo
1

Happened again tonight... device turned off without rule or input from me...

photo
1

No wonder. My house started to live its own life now...it's like Skynet from the Terminator...thanks got I don't have smart lock and speakers yet connected cause I can imagine I would receive a message saying your house does not want you out...or that the lock would open on its own letting everyone in like a hippie

photo
1

Haha


Well my test is underway. I'm waiting to see if one of my sockets misbehaves.


For z-wave devices, I would suggest adding more devices (especially covering gaps between existing ones) to see whether it helps. Obviously you're not going to add more for the sake of it, but if you're considering adding more gadgets, does so to see whether this helps to make the mesh network more reliable.


And let us know the results!


Are the devices that operate by themselves close to your neighbours?

photo
2

This keeps happening... two different devices today turned on/off without input from me or rules.

Strange! I have not found any pattern on this. Different devices at different times...

photo
1

Similar issue with the thermostat. Intermittently changed to 5.0 C.

photo
1

I also have a similar problem. I have a switch that is turning on and off alone with no command or rule. Door opener sensors are also not updating the status correctly and I have noticed a very high delay in response to commands and or status requests.

photo
1

And it keeps happening... two more devices this night/morning.

photo
1

Apparently Zipato have found a problem and they are busy fixing it.

photo
1

I'm still experiencing this behavior almost every day with a Qubino module. First it was at night, now since a week or 2 during the afternoon.

photo
2

Hello,


On my support ticket Vedran gave me this answer and so far my issues are gone (almost two weeks).


"Please upgrade the box to the latest firmware version 1.2.24, and after that reboot the box.

Sync it up and do a ZWNTUT procedure found under Zwave options on your controller, and tick the checkboxes - skip sleeping and disable retries.

After doing this, sync up the box once more and reboot it.


Monitor the system for a couple of days after that and let me know if the problematic behaviour still exists."


My system has been very stable since that.


Regards

Tobias

photo
1

I've just tried it. Let's see.

photo
1

I did the procedure yesterday. Apparently it had worked but today it happened again. The lights go off alone in random times after the commands on the computer, smartphone and are now erasing even with the command made manually on the switch.

photo
photo
1

Does anyone know why ph-pan04.eu modules shut down by themselves? I turn on the lamps through the mechanical switch and they stay on until I turn them off, but when I call them by the smartphone application or by the computer or even by IFTTT rule they delete themselves after a while. Sometimes in less than 1 minute, another 5 minutes and sometimes 10 minutes.

photo
1

delete? you mean they remove themselves from the system, or turn off?

Either way this seems not normal. It must be something you have programmed in with a rule somewhere. is this a single module, if so what input is the switch wired to, the S1 or S2?

photo
1

Hi Adrian, I meant that the switch turns off by itself. I think the spell-checker changed the text to "deleted" and I did not notice it when I was posting.


I have two loads, one circuit with fluorescent lamps on S1 and other at S2. I also have one rule at S1 that do the S1 light turn on when the door sensor toggle to the state "open". I attached the printing for this rule.

photo
1

your rule needs modification. What are you trying to achieve?

If you want the light to turn off after turning on after 2 minutes, you should include the puzzles in the IF "open" block.

If you want the light to turn off on the door closed also, add an If then Else block and just add the time delay to the Else part of the block to.

Still not sure why you modules are turning off after random times. Have you played around with bindings, association or the parameters? Or are there any other rules with these modules in them?

photo
1

Hello Adrian, Thanks for your feedback. They helped a lot. I desire to turn off the light after a while. The previous rule was working, but after the light turned off, and if I tried to turn it on, the light would go off again but not in 180 seconds it turned off in a random time. Sometimes in less than 2 seconds and sometimes in 13 minutes. Yesterday I did the ZWNTUT procedure described by Tobias and apparently it worked.


I did two tests. I opened the door and the lights turned on, and I waited 180 seconds and the lights turned off, I also waited more some seconds and I turned on the lights again and wait more time (about 15 minuts) they don't turned off by themselves. I turned off manualy and asked Alexa (Amazon Echo Dot) to turned on and I also waited more 15 minuts and they don't turned off by themselves.


About your doubt if there has other rules or parametrizations in this module, I only have more one rule, that is the command with IFTTT and Amazon Echo Dot, and no more other rule or parameterization. (picture attached)


I'll implement the rule you described. Thank you so much.


Alessandro

photo
1

Hello Adrian, The problem still persists. Today it happened again, even with the changes proposed in the rules. I believe it is some problem in the micromodule which must have its transistor control current interrupted. Then the module interrupts the lamp operating current. Today it seems like it's turning itself off in less time.

photo
photo
1

Does anyone know why ph-pan04.eu modules shut down by themselves? I turn on the lamps through the mechanical switch and they stay on until I turn them off, but when I turn on the lights by the smartphone application or by the computer or even by IFTTT rule, they turn off by themselves after a while. Sometimes in less than 1 minute, another 5 minutes and sometimes 10 minutes.

photo
1

Today I observed that some qubino flush shutter modules moved the blinds several times up and down (today between 13 an 15:00) without a rule or manual input. Very strange..

Anybody else seeing the same thing again?

photo
1

I've had the same problem several times last year. Was there an official Zipato answer / solution?

photo
1

Lately I have had some devices turn on by them selves...

Different devices every time.

Theese happenings where gone for a few months but now back, strange!


This night a door sensor suddenly changed state causing the alarm to set off (strangely not sirens = good for us :-) ).


I have created a ticket to see if Zipato can find something.


Theese strange happenings is most unwanted!