Hi Rick,
Thank you so much for those changes, they’re working an absolute treat 
I doff my hat 
to you good sir.
thank you for testing Andy! 1.9.56 is now in review
845 installations so far → target is still to reach 1000 
You are most welcome good sir
I have recommended this app to a few people, and will continue to do so for anyone wanting to monitor their devices or get notified when something stops responding. (I have 507 devices in total linked to my Homey, so I’d be pretty screwed without this app ha ha)
Hi @Rick_D, I have a feature request.
Would it be possible to monitor a numerical value from a device so that, for example, if that value remains unchanged for 15 minutes, this flow card triggers a flow?
The thing is, it’s not that the device is no longer reachable, but simply that a value needs to be changed
Example flow:
When…
– Value X has not changed for 15 minutes
Then…
– Push notification
Use cases:
- I have a battery-powered weather station with an additional solar panel. If the batteries run low, the weather station doesn’t shut down completely, instead, only a few readings stop updating, while others continue to do so. The weather station is still accessible.
That’s why it would be helpful if I could monitor one of the values that fail when the battery capacity is too low. - I’m using the EcoTracker IR from everHome, which is supported by the Homey everHome app, to measure my home’s energy consumption.
Unlike a P1 meter, which is permanently connected via a plug, the EcoTracker is attached to the electricity meter with a magnet. If the EcoTracker is accidentally moved even slightly, the data can no longer be read from the electricity meter. As a result, the last measured value “freezes”. This is what happened this week, and no readings were updated for about 22 hours:
Because I think this kind of monitoring generates a lot of data traffic and might also require a lot of storage, the following approach could be an alternative:
Record and save a numerical value from a device every 15 minutes. If 4 saved values are all identical, trigger the flow card.
I know it’s possible to create the alternative option myself using flows and logic variables. However, I’d prefer a single flow card.
Hi Dirk, thanks - good use cases! The “frozen value” case (EcoTracker, a weather station where only some readings stall) is exactly where Device Watchdog is too coarse today: it looks at the whole device (the newest change across all its values) and counts in hours, so a device where only one value freezes is never flagged.
What I have in mind is a Flow trigger “Value of [value] of [device] hasn’t changed for [15] minutes”. No history or storage needed: it only uses the “last changed” timestamp Homey already keeps for every value, so nothing piles up. It would fire once when the value freezes and re-arm as soon as it changes again.
One question: would you also like a second trigger for “value changes again” (e.g. for a “back to normal” push), or only the freeze alert?
I can’t promise a date yet, but it’s on my list.
That’s exactly what I need… 
In my case, I really only need the trigger when a value is frozen.
Thanks in advance!
Update: 846 installations 
Hallo Dirk
New Flow trigger: “Value [X] of [device] hasn’t changed for [15] minutes/hours”.
Pick the device and the value (temperature, power, …), set the time, done. It fires once when the value freezes and arms itself again as soon as the value changes. Tags: device, zone, value, unchanged for, last changed.
It uses the “last changed” time Homey keeps for every value, so nothing is stored and it doesn’t wait for the scan interval. It stays quiet while the device (or the whole Watchdog) is paused.
One thing to keep in mind: values that are constant on purpose (e.g. 0 W) will trigger too, so pick values that normally move.
Test build (1.9.57):
Please try out Device Watchdog | Homey
Installed and testing… this could be very useful for keeping an eye on Aqara FP300 stale sensor readings 
(a long running and seemingly everlasting problem ha ha)
Could it be that the search function does only search in the selected group and no more all devices like in an earlier version?
The search should look at all devices, but it is combined with the status tiles at the top: if one of them is highlighted (e.g. “Not reporting”, “Low battery”, “Paused”), the search only looks inside that group. Click the highlighted tile again, or “Total”, to clear the filter - then the search covers everything again.
Should have always worked like that 
You are correct. Did not notice that a status tile was highlighted
First of all, thank you very much for the quick implementation.
However, it doesn’t seem to be working as expected. Yesterday, I created three flows using the new flow card. I monitored three different devices with different times and, in some cases, different capabilities.
One flow was for the weather station and one for the EcoTracker IR. These two flows should not have been triggered, since I had just replaced the batteries in the weather station and the EcoTracker IR was reporting data reliably. Nevertheless, these flows were triggered.
Should I send you the diagnostic report, or would you like to test a flow yourself first? Or do you need any other information?
Hallo Dirk
Yes please a report too i will check in the evening
Here you go: 7f6557a3-c057-4625-9970-852abad8fcf2
Hallo Dirk
The trigger read a cached copy of the device whose “last changed” time is only refreshed by the periodic scan (every 60 min by default). With a 15 minute window it could announce a value as frozen although it was changing all the time. It now always reads the current state. I checked it on my Homey: values that keep changing stay silent, a value that really is frozen still triggers.
Fixed in 1.9.58 - test build: Device Watchdog | Homey
Would be great if you could try it again with the weather station and the EcoTracker. If one of them still triggers, tell me which value and which time you set, and add the tags “last changed” and “unchanged for” to your push message. The new version also writes what the trigger decided into the app log, so a diagnostic report will actually show it.
Danke 
@SingKT I added a note for the search function in 1.9.58: while you type in the search and a status tile is active, a line appears above the list saying “The search only looks inside “Not reporting” - the filter above is active.”, with a button “Search all devices” that clears the filter.
Hi Rick,
Now it works like it should!
I tested the Flow Card with two devices whose values usually change constantly, and I didn’t receive any notifications. I also tested the Flow Card with another device whose value typically remains the same for a certain period of time. The Flow Card was triggered correctly, and I received a notification.
Thank you so much again!
