I’ve had Alexa Smart Home Integration configured for a couple years or so with no issues. I have some binary sensors configured for various things like notifying me when my washer is done and when my roomba bin in full. I then have Alexa routines configured to announce when one of these binary sensors changes state.
A couple days ago, I upgraded to core-2022.3.1. Ever since, it appears that duplicate events are being sent when a binary sensor changes state so I get two Alexa announcements instead of just one. Just to make sure something wasn’t corrupt from the upgrade with the existing binary sensor, I created a new one and it also has the same issue.
Anyone else experiencing this issue?
EDIT: It’s not just binary sensor. It’s everything being sent to amazon that’s being duplicated.
I’m experiencing something similar…recently upgraded and various automations started duplicating including binary sensor triggers to Alexa, but also notifications…
Quite annoying, and currently haven’t been able to figure out why…
I opened an issue for this on github. While testing previous releases of the alexa component as a custom component, the problem stopped even when I switched back to the built-in component. However, after rebooting the system, the problem started again. I haven’t again attempted to figure out which release caused the problem.
1 Like
I tried updating and fully restarting my Unraid server and Docker multiple times. The duplicate events have occurred intermittently and sometimes appeared to be resolved but then resumed duplication again later….(shrug)
This issue drove me crazy. And now i cracked it. The reason is totally nuts!
To trigger routines by binary sensors your HomeAssistant instance connects to the amazon cloud in some special way. (See the Section about Proactive Events of the manual, I use my own Alexa-Skill and not the “Home Assistant Cloud”-Way)
It seams that if your HomeAssistant is reachable by IPv4 and IPv6 simultaneously, the amazon cloud gets triggered over both ways, causing duplicated events/messages. I just cut the IPv4 access to HA in my router and (to ensure any opened connection are closed) restarting HomeAssistant. And now it works fine.
Another IPv6-Madness. 
1 Like
I don’t have IPv6 enabled on my router or on Ubuntu. At some point, I’ll do some more troubleshooting.
Nobody has touched the issue I opened on github and there’s at least 1 other issue opened there for the same problem.
A simple restart of the instance fixes the problem for me, but it may or may not start at a later time in the day though.
I’m also not using any ipv6 forwarding on my router/with my current provider. IPv6 NAT is specifically turned off on my router.
Running out of ideas where to troubleshoot this issue.