Is there any way to read Insteon link information from a device (and scene) within python? I seem to have a gremlin in my system somewhere (maybe line noise?) that corrupts links and scenes (linking to nonexistent devices etc). I would like to write a routine that looks for all devices listed in scenes (and device-to-device links) for any unknown device IDs /addresses, then send myself a flag for the devices and scenes to review manually.
When a bad link or scene is activated, the 2413U goes on vacation for about 20 minutes... messages come in from my Z-Wave Fibaro multi sensors, but no commands will go out to any device (neither from the UI nor from my python plugin) - just time-outs from every Insteon device. The Insteon network is locked up. During the 20 minute break, I have tried resetting the 2413, re-linking everything, but alas still no outbound commands (either from my plug-in nor from the UI). The only cure is time.
This happens every few weeks, but I'm not sure what device is causing the issue. Unfortunately, I have hundreds of devices and manually going through them to evaluate every link on a device-by-device basis in indigo takes an enormous amount of time... but after I am done (and delete all the bad links), it works fine for a few weeks.
I apologize if this request is elsewhere on this site, but I couldn't find it with the phrases I tried to type.
Thanks for any help
UPDATE: This timeout problem happened again and I found that I have bad parameters in some device links. In particular a companion light switch links to a load switch (each end of hallway scenario). The companion switch was trying to send the load switch 255 for brightness level. So any way to read link information from devices would be handy to help me automate my search for the link problems. My hope is that if I repeatedly run this script fro all my devices (and comparing results to command logs), I may eventually find out what is causing the corruption in the first place. So far I don't think it is isolated to a single device, but alas I have not been able to force the problem to occur on demand so its a long painful waiting game.