Polling and Device Check for Groovy devices no longer functioning

General Discussion
1

For the past few weeks I’ve noticed some of my devices with incredibly high inactive times, some well over 24 hours. These are all devices that have one thing in common, they rely on the Polling capability (poll()) or Device Watch (ping()) to trigger a request/response from the device to keep it current (for instance, Zwave Switch Generic). DTHs that use the scheduler to trigger a poll are keeping up to date, but none of the devices with high inactivity times (higher than their device Watch timeout) are getting flagged as INACTIVE either.

I do have a few cloud based DTHs that use Polling capability that still appear to have poll() being called (about every 6 minutes) but there are at least 5 or 6 that aren’t getting anything. I know Polling could be inconsistent but I’ve never seen simple zwave plugs using the built-in drivers getting over 2 days with no activity yet still not getting flagged as INACTIVE. And they’re still online, a quick on/off toggle resets their update time instantly.

I rebooted the hub, no change.

Is something broken with Polling and Device Watch? Is this a hint to start converting over to Edge drivers?

1 Like
2

It looks like Polling behaviour has been restored at least for some devices. Any of my devices with the Polling Capability that are Cloud based are now getting their poll() function called at just under 6 minutes from the last time they sent data. Any Local devices with the Polling Capability in their DTH (for example, generic zwave outlets) are still not getting any poll()s.

It doesn’t appear as though Device Check (checkInterval) is marking any devices OFFLINE still. I have both Cloud and Local devices with last update times well past their checkInterval and they are still showing ONLINE.