Devices were timing out. This 'fix' works. Any idea WHY? Insight desired.

[f1d094]

I have had an ongoing issue for a while where several devices kept timing out when trying to refresh them, preventing me from syncing my scenes. Example:

ERROR: Handler timed out - no more retries (3 sent)
  Adding 0fd7: 3c.4d.24 (modem)          grp:  30 type: RESP data: 0xff 0x00 0x01:
Handler timed out 1 of 3 sent: Ext: 0d.65.8b, Type.DIRECT ext mh:1 hl:1, 2f 00 00 02 0f ff 08 a2 1e 3c 4d 24 ff 00 01 00
Handler timed out 2 of 3 sent: Ext: 0d.65.8b, Type.DIRECT ext mh:2 hl:2, 2f 00 00 02 0f ff 08 a2 1e 3c 4d 24 ff 00 01 00
Handler timed out 3 of 3 sent: Ext: 0d.65.8b, Type.DIRECT ext mh:3 hl:3, 2f 00 00 02 0f ff 08 a2 1e 3c 4d 24 ff 00 01 00
ERROR: Handler timed out - no more retries (3 sent)

This has been ongoing for several days. None of the six affected devices would do a 'refresh' or 'sync-all' (which of course first does a refresh).

I decided I'd try and do some lower level debugging and thought I'd go through some basic raw-commands as described in the Insteon Command Tables 20070925a reference and see what's what. Example:

insteon-mqtt insteon_mqtt.yaml raw-command 0d.65.8b 0x1F 0x00

Shortly after I started poking around I realized I couldn't remember what I had previously received back (above paste) and it had scrolled out of my history...so I just ran another insteon-mqtt insteon_mqtt.yaml refresh 0d.65.8b...and magically, it worked just fine. What?

By deductively working it backwords I was able to work out that if a device is timing out, I am immediately able to 'fix' it by simply doing the above raw-command.

Good to know, and it cured my problem...but does anybody have any idea WHY?

The devices in question were 4x ToggleLinc 2466S and 2x SwitchLinc 2476D, if that helps.