Maintenance task / notification / alert list

[blhoward2]

Add metadata for devices that are unknown in the cache allowing the UI to alert the user that a config file is now available and that the device should be reinterviewed.

Consider doing the same for new custom device files (which would require comparing the custom device file to the device cache each time).

[AlCalzone]

Thinking about this in a broader sense, we could add a maintenance task list for nodes. Points could include:

• New config file -> re-interview necessary --> Already done differently, but could be tied in here
• Bad node health (building on the statistics, e.g. when a node has a high error count)
• Firmware update available (based on the firmware service we've been discussing)
• Device spamming
• Bad signal strength
• Slow communication with IP-based controllers
• ... other ideas?

[robertsLando]

Hehehe implementing all that points would be awesome! Point 3 require an entire new project to handle firmwares uploads by manufacturers, I can do that but I need to know what specific features we need there

[blhoward2]

What about flagging that the device spams us with excessive traffic?

[robertsLando]

Is there a way to get the 'signal' status of a device, I mean how good is the communication with it

[AlCalzone]

There is - zwave-js doesn't use that info yet.

[blhoward2]

How about a known critical flaw that disables the device if we push a flag in a config file? Just as an insurance policy. I’m thinking like on ozw there was a bug that caused all Leviton devices to crash the driver. You disable the device and flag it so the driver at least comes up, then the user knows to downgrade. This would be a last resort for things that absolutely don’t work and will bring down the driver if operated.

Perhaps a lesser Known Issues flag for the rare occasion that we know a device doesn’t work and we can’t work around it, but the driver isn’t endangered.