Hassos 17.1 issue shelly integration & Roborock

[federicodigiacomo]

Describe the issue you are experiencing

Good morning everyone, since I installed OS17.1 this morning, the native integrations for Shelly and Roborock no longer work. Has anyone else experienced the same issue?

What operating system image do you use?

generic-x86-64 (Generic UEFI capable x86-64 systems)

What version of Home Assistant Operating System is installed?

17.1

Did the problem occur after upgrading the Operating System?

No

Hardware details

Home Assistant Green

Steps to reproduce the issue

1. Reboot

...

Anything in the Supervisor logs that might be useful for us?

No

Anything in the Host logs that might be useful for us?

No

System information

No

Additional information

No response

[Chrisinho]

I can confirm that.
In addition, Huawei Solar integration no longer works for me after the update.

[agners]

Can you try using ha os boot-slot other to boot to the previous OS release and see if the problem indeed goes away?

If it goes away, use ha os boot-slot other to boot into HAOS 17.1 again. If the issue is back again, it indeed seems OS related.

[Chrisinho]

Can you try using ha os boot-slot other to boot to the previous OS release and see if the problem indeed goes away?

If it goes away, use ha os boot-slot other to boot into HAOS 17.1 again. If the issue is back again, it indeed seems OS related.

In the meantime, I restored the previous version using 'ha so update —version 17.0', and initially some Shelly devices still weren't working. I restarted the devices again and then they worked again.
I also removed the Roborock vacuum robots from the docking station and switched them off and on again manually. Then I restarted the integration and it worked again.
The Huawei Solar integration also worked again after the downgrade.

I will install the upgrade again over the weekend and if problems arise again, I will try your suggestion and let you know how it goes.

[jressel01]

he has Home Assistant Green so i think its not board/generic-x86-64

[scherer42]

I’m running on Raspberry Pi 5 and had exactly the same issue. Shelly and Huawei Solar stopped working after updating from 17.0 to 17.1. I had to run the ha os boot-slot other command to revert it. Initially after reverting some shelly devices and Huawei Solar were still not working, but eventually all came back to normal.

[DjLex2021]

I can confirm this. After updating to version 17.1, the Shelly integration stopped working.

After running 'ha os boot-slot other', I was back on version 17.0, but the Shellys still didn't work. I had to restart the devices, and then they all worked again.

After running 'ha os boot-slot other' again, they all went offline again. Error: No connection could be established.

The Shellys all failed one after the other and were also inaccessible via the web UI. Version 17.1 is doing something strange with the communication to the Shellys.

I've now reverted back to version 17.0, and since then (5 hours), everything has been working as it should.

[flecmart]

I cannot confirm this. Upgraded from 16.3 to 17.1 on generic x86 VM installation on proxmox today and all of my 25 shelly devices kept working normally. I did not do the storage driver migration.

[scherer42]

What additional information is needed in order to debug this issue? Let me know and I can try reverting back to 17.1 and collecting those. Thanks.

[agners]

@scherer42 are you (or others affected) on beta channel?

Honestly, I don't see a reason why the OS update could affect just one integration specifically. What could be helpful is Core logs of the first minutes showing the exact errors the Shelly integration experiences (maybe of both, 17.0 and 17.1 to look for differences).

[DjLex2021]

@agners I will search for the logs ...

It also affects ESPHome connections:

OSError: [Errno 113] Host is unreachable�[0m
�[33m2026-02-14 14:12:21.102 WARNING (MainThread) [aioesphomeapi.reconnect_logic] Can't connect to ESPHome API for lufter-akku-3 @ 192.168.2.67: Error connecting to [AddrInfo(family=<AddressFamily.AF_INET: 2>, type=<SocketKind.SOCK_STREAM: 1>, proto=6, sockaddr=IPv4Sockaddr(address='192.168.2.67', port=6053))]: [Errno 113] Connect call failed ('192.168.2.67', 6053) (SocketAPIError)�[0m
�[33m2026-02-14 14:12:21.165 WARNING (MainThread) [aioesphomeapi.reconnect_logic] Can't connect to ESPHome API for klima-wohnzimmer @ 192.168.2.111: Error connecting to [AddrInfo(family=<AddressFamily.AF_INET: 2>, type=<SocketKind.SOCK_STREAM: 1>, proto=6, sockaddr=IPv4Sockaddr(address='192.168.2.111', port=6053))]: [Errno 113] Connect call failed ('192.168.2.111', 6053) (SocketAPIError)�[0m
�[33m2026-02-14 14:12:21.165 WARNING (MainThread) [aioesphomeapi.reconnect_logic] Can't connect to ESPHome API for b2500-node @ 192.168.2.89: Error connecting to [AddrInfo(family=<AddressFamily.AF_INET: 2>, type=<SocketKind.SOCK_STREAM: 1>, proto=6, sockaddr=IPv4Sockaddr(address='192.168.2.89', port=6053))]: [Errno 113] Connect call failed ('192.168.2.89', 6053) (SocketAPIError)�[0m

[bieniu]

Regarding Shelly...

1. Is the issue related to gen1 or gen2+ devices?
2. Does the issue affect battery-powered devices?
3. Are the devices and the HA server assigned static IP addresses?
   I updated HA OS to 17.1 yesterday and see no issues with Shelly integration.