New watchdog limits break device reconnects

[deviantintegral]

rtl_433 exits if a device fails. For example, a remote rtl_tcp device could be restarting, or a USB radio could have been unplugged.

HA's supervisor has changed so that if 10 restarts happen in 30 minutes, it will no longer restart the addon. That happens pretty quickly in my testing, and the addon ends up being shut down until it's manually started or Home Assistant OS restarts.

home-assistant/supervisor#3779

I've asked for suggestions at https://discord.com/channels/330944238910963714/673545314900377620/1015432911039770774, but I think we may need to wrap our commands in a retry loop instead of relying on the watchdog.

[deviantintegral]

I'm thinking the best way to solve this is to switch to using s6 dynamic services: https://skarnet.org/software/s6/instances.html

For now, this automation is a good workaround: https://community.home-assistant.io/t/tenacious-watchdog-for-add-ons-example-for-zigbee2mqtt/521888