I have several Shelly devices in HA that have been working fine. I updated to the latest FW of HA and the devices still worked fine. All of a sudden at 11:40am yesterday, both of my Shelly 2.5s went unavailable, along with my H&T sensor. All other Shelly 1 relays work as they should. The 2.5s work fine in the Shelly app, just not in HA. In the Shelly app the H&T shows 100% humidity and 105.5 F. I’ve verified that all the devices are online and are listed as active in my Unifi controller. I’m not sure what else to check to try and figure out what is causing the disconnection.
You say you updated to new firmware. Did you update advanced developer settings, set ColoT and give it the ip address and port such as this (192.168.xxx.xxx:5683) The 5683 port is required. see below:
1 Like
I updated the FW of HA. After the HA update the devices worked like they did before. Then only the two 2.5s stopped working. The advanced developers option is not there in the shelly app. I am not on the LAN at the moment. Does the developer option require you be on the LAN? Or perhaps, web interface only?
Not sure if this helps or not, but here is the error log:
This error originated from a custom integration.
Logger: homeassistant
Source: custom_components/shelly/__init__.py:330
Integration: Shelly smart home (documentation)
First occurred: 11:39:50 AM (1 occurrences)
Last logged: 11:39:50 AM
Error doing job: Task exception was never retrieved
Traceback (most recent call last):
File "/config/custom_components/shelly/__init__.py", line 330, in start_up
pys.start()
File "/usr/local/lib/python3.9/site-packages/pyShelly/__init__.py", line 185, in start
self._mqtt_server.start()
File "/usr/local/lib/python3.9/site-packages/pyShelly/mqtt_server.py", line 123, in start
self._init_socket()
File "/usr/local/lib/python3.9/site-packages/pyShelly/mqtt_server.py", line 129, in _init_socket
sock.bind((self._root.bind_ip, self._root.mqtt_port))
OSError: [Errno 98] Address in use
Meaningless without a version.
Also, you are using a custom component, so go ask the author on his github.
First post of the thread, I stated that I updated to the latest FW. But, I guess I should have listed the FW version- 2022.2.5.
latest is never a version, but 2022.2.5 isn’t even the latest.
I’ll have to check when I get back home, although I am pretty sure I already enabled that feature.
So after I update the FW of HA to 2022.x.x, I started using the native Shelly integration. HA rediscovered all Shelly devices and I basically reconfigured the Shelly integration. Should I delete the custom_component that relates to Shelly?
Yes you should delete that.
And just for clarity, you have not upgraded the firmware. The firmware is in the device. You have upgraded the program Home Assistant Core.
Hello.
I have several Shelly devices and I face this problem.
Lets take one device as an example:
with latest firmware:
CoIoT enabbled:
.1.100 is my HA
I thought it might happen bcs HA and Shelly are in different vlans.
So in Unifi I have done the followings:
Enabled IGMP Snooping in all WiFi.
Enabled Multicast DNS in all networks.
Created below firewall rule:
Then restarted all shelly devices.
Any idea what I am missing???
Thank you!
From the iot vlan can you connect to 192.168.1.100:5683?
From my phone when connected to iot Vlan, I connected to Home Assistant 192.168.1.100:8123 but not to 192.168.1.100:5683.
What should I see there?
Also when I connect my phone to main vlan (where HA is), I don’t see anything in 192.168.1.100:5683
Actually the port is udp and I am not sure how to test that.
It drives me crazy!!! They get unavailable very often and as a result, many automations don’t work.
Any suggestion anyone please?
Same here, thought it was a networking issue. I verified my unifi devices. I can see tcpdumps of the traffic flowing. I have nearly 20 Shelly’s in the house. I love shelly, but if the reliability is tanked, I am going to have to ebay them and find some zigbee/zwave replacements. Which stinks, it will be another weekend lost. I would much prefer a fix.
1 Like
It’s not. Do you have the CoIoT properly set up?
I have never had this issue until this weekend when i moved my Shelly Devices over to their own VLAN - All IoT devices for that matter.
I intended to also add firewall rules which will block internet access etc. but even right now I am experiencing the same issue as OP
I saw an improvement after enabling CoIoT, however certain devices are still having issues - Not nearly as frequent as before CoIoT:
One thing to note is that I have set my Static IP’s on my router instead of the Shelly Devices themselves - I used to have static IPs on the Shelly itself, but changing network caused lots of issues and it is easier to centrally fix IPs in the future.
I am testing some devices on Shelly Static now to see if they improve.
I had similar issues. When i unchecked restricted access, shellys came back available in HA again. Hope this works for u.
1 Like