Z-wave Device "Deleted" From Interface By HS3???

[BobSpen]

Can anyone explain why HS3 might DELETE a device from the Z-Net interface without any action on my part?

The system was running normally, and apparently the batteries in a Aeon Labs Multisensor6 went really dead(not reported but after the incident, the devices led would not light). The last reading several hour earlier was 50% so I was not concerned. Apparently, when the batteries died, I got the following message in the log:

Dec-16 5:23:19 AM Z-Wave Warning Device Reset Locally Notification received for device Node 21 Z-Wave NewMotion1 Aeon Labs Multilevel Sensor - DEVICE IS BEING REMOVED FROM THE SYSTEM.

I was fast asleep so I did not set the controller software to "Delete a node" and "push" the button on the Multisensor6. I later looked at the node report and sure enough that node was missing(It was still intact on the Device Management Page)

So the question is what could have caused this very undesirable event??

Any help would be appreciated.

BobSpen

[Rupp]

How long ago did you add this device? It's not a case of adding it and it didn't save is it?

[BobSpen]

Rupp,
Added it last Feb/ Not a new add!---(so I got about 10 months on a new set of Batts.)

[BobSpen]

Bump-Anyone seen this or can explain why it might happen?

[macromark]

You're not using "Z-Health" correct?

[bebaldin]

Bump-Anyone seen this or can explain why it might happen?

I can't explain it, but I have seen it more than once. This was actually fairly common when I was trying to run a primary and secondary controller on the same home ID, which would NEVER stay in sync. Since I abandoned that format and went to zones with different home IDs, I haven't had this problem that I can recall.

[rjh]

Are you using more than one Z-Wave interface?

HS will not delete a node unless you manually force a deletion through the UI. However, if you have multiple Z-Wave interfaces they will sync with each other.

If you only have one interface, sounds like a bug in the Sigma code. I don't believe I have ever seen it myself though.

[stefxx]

HS will not delete a node unless you manually force a deletion through the UI.

Are you sure? I recall if I reset a Z-Wave Plus module to factory settings, HomeSeer would log a "DEVICE IS BEING REMOVED FROM THE SYSTEM" and actually removed the device from HS3. Which I don't really appreciate because than there is no option to replace it with a replacement module anymore.

[rjh]

You are correct, I forgot about the reset feature at the node. That is a new Z-Wave + requirement. But it is a manual procedure, unless the device has a bug and initiated the reset on its own.

Are you sure? I recall if I reset a Z-Wave Plus module to factory settings, HomeSeer would log a "DEVICE IS BEING REMOVED FROM THE SYSTEM" and actually removed the device from HS3. Which I don't really appreciate because than there is no option to replace it with a replacement module anymore.

[stefxx]

But it is a manual procedure, unless the device has a bug and initiated the reset on its own.

Agreed. But since it happened to the OP in the middle of the night while he was sleeping, something must have happend. Can Z-Wave signals become corrupted during transmission and therefor the wrong signal was received by HS3? Or is the Z-Wave signal "protected" by decent checksums or otherwise?

Only other option is a misbehaving or buggy module indeed.

[BobSpen]

MacroMark and rjh,
I have only one interface--the Z-Net; and I do NOT use Z health.

Since the batteries seem to die so quickly, after 10months, could that have sent the "reset" to the Z-Net? I did NOT get a "battery low" indication and batteries were at 50% earlier in day--although when I physically moved the unit, it seemed dead(no reaction to pushing the "activate" button), the batteries both measured about 2.5volts(open circuit) verse 3v when new. Seems like 2.5 is still a pretty health voltage for a Li-ion battery?

rjh,
Could the "reset" from a gen5 device be inhibited(or at least require an acknowledgement in the Z-net since I think most of use would like to be more in control of when a device is removed?

BobSpen

[BobSpen]

rjh--Bump

[rjh]

A reset is not sent on low battery. If the device did reset, it would send a notification to HomeSeer so you can check the log and see if a message was there. I suspect it was just a bug in the zwave firmware, we will probably never know why it happened. If it keeps happening there may be an issue.

MacroMark and rjh,
I have only one interface--the Z-Net; and I do NOT use Z health.

Since the batteries seem to die so quickly, after 10months, could that have sent the "reset" to the Z-Net? I did NOT get a "battery low" indication and batteries were at 50% earlier in day--although when I physically moved the unit, it seemed dead(no reaction to pushing the "activate" button), the batteries both measured about 2.5volts(open circuit) verse 3v when new. Seems like 2.5 is still a pretty health voltage for a Li-ion battery?

rjh,
Could the "reset" from a gen5 device be inhibited(or at least require an acknowledgement in the Z-net since I think most of use would like to be more in control of when a device is removed?

BobSpen

[BobSpen]

rjh,
The message that caused the removal was a "reset" signal from the device:

Device Reset Locally Notification received for device Node 21 Z-Wave NewMotion1 Aeon Labs Multilevel Sensor - DEVICE IS BEING REMOVED FROM THE SYSTEM.

But it was not from holding ( I think greater than 20 seconds) the "activate" button that normally would cause this. It must have been from the batteries dropping too low--even though there was no low battery signal sent.

So what I am suggesting is that apparently confused electronic in the Aeon Labs Multisensor6 can send an erroneous "reset" to the Z-net, which then removes the device.

Is there a significant value in having the Z-net(or the Z-wave PI) automatically remove the device from the network? Why not just block that capability?

BobSpen

[randy]

rjh,
The message that caused the removal was a "reset" signal from the device:

Device Reset Locally Notification received for device Node 21 Z-Wave NewMotion1 Aeon Labs Multilevel Sensor - DEVICE IS BEING REMOVED FROM THE SYSTEM.

But it was not from holding ( I think greater than 20 seconds) the "activate" button that normally would cause this. It must have been from the batteries dropping too low--even though there was no low battery signal sent.

So what I am suggesting is that apparently confused electronic in the Aeon Labs Multisensor6 can send an erroneous "reset" to the Z-net, which then removes the device.

Is there a significant value in having the Z-net(or the Z-wave PI) automatically remove the device from the network? Why not just block that capability?

BobSpen

It sounds like Rich is saying this "feature" is a requirement of the Z-Wave plus specification, so it may not be something they can block.