Home Assistant OS 16.2

[sairon]

Changes

This release brings updates and fixes for all platforms, updating Linux kernel to the latest stable version and various system components. Bug causing problems with some USB-Serial convertes on Home Assistant Yellow and Raspberry Pi has been fixed. Several improvements have been done to improve experience when pairing BLE devices. Timezone configured in Home Assistant settings is now propagated to the operating system and used for instance when reading logs directly on the host. All users are recommended to update.

Home Assistant Operating System

• buildroot: Update BlueZ to 5.83 with connection abort fix (buildroot: Update BlueZ to 5.83 with connection abort fix #4250) @bdraco
• sysctl: disable TCP slow start after idle (sysctl: disable TCP slow start after idle #4239) @RubenKelevra
• sysctl: Enable linear RTO for thin TCP streams (sysctl: Enable linear RTO for thin TCP streams #4238) @RubenKelevra
• bluetooth: Increase TemporaryTimeout to 195 seconds (bluetooth: Increase TemporaryTimeout to 195 seconds #4249) @bdraco
• Bump Buildroot to 2025.02.5 (Bump Buildroot to 2025.02.5 #4235) @sairon
• Make system timezone setting persistenly configurable (Make system timezone setting persistenly configurable #4224) @sairon

Raspberry Pi

• Fix missing rpi-rf-mod overlay on Yellow (Fix missing rpi-rf-mod overlay on Yellow #4244) @sairon
• Backport patches for Raspberry Pi dwc2 driver (Backport patches for Raspberry Pi dwc2 driver #4231) @sairon

Home Assistant Yellow

• Fix missing rpi-rf-mod overlay on Yellow (Fix missing rpi-rf-mod overlay on Yellow #4244) @sairon
• Backport patches for Raspberry Pi dwc2 driver (Backport patches for Raspberry Pi dwc2 driver #4231) @sairon

Home Assistant Green

• Linux: Update kernel to 6.12.43 (Linux: Update kernel to 6.12.43 #4255) @sairon

Open Virtual Appliance

• Linux: Update kernel to 6.12.43 (Linux: Update kernel to 6.12.43 #4255) @sairon

Generic x86-64

• Linux: Update kernel to 6.12.43 (Linux: Update kernel to 6.12.43 #4255) @sairon

Hardkernel ODROID

• Linux: Update kernel to 6.12.43 (Linux: Update kernel to 6.12.43 #4255) @sairon

ASUS Tinker

• Linux: Update kernel to 6.12.43 (Linux: Update kernel to 6.12.43 #4255) @sairon

Khadas VIM Series

• Linux: Update kernel to 6.12.43 (Linux: Update kernel to 6.12.43 #4255) @sairon

Generic aarch64

• Linux: Update kernel to 6.12.43 (Linux: Update kernel to 6.12.43 #4255) @sairon

Build System

• Generate list of built-in containers in GH build summary (Generate list of built-in containers in GH build summary #4247) @sairon
• Add checkbox for running test in GH build workflow (Add checkbox for running test in GH build workflow #4246) @sairon
• Fix scripts/enter.sh so it is usable on macOS (Fix scripts/enter.sh so it is usable on macOS #4243) @slonopotamus
• Enable publishing of dev builds to R2 without bumping version (Enable publishing of dev builds to R2 without bumping version #4232) @sairon

---

This discussion was created from the release Home Assistant OS 16.2.

[Anto79-ops]

trying to update (from 16.2.rc2), but just keeps looping like this:

Intel NUC 12 bare-metal.

EDIT: needed full system reboot. All good now.

[agners]

The system will reboot on its own after the upgrade. If the reboot does not happen automatically, something must have gone wrong. Check the logs, especially the Supervisor logs.

[Anto79-ops]

@agners heres's the logs:

�[32m2025-09-08 17:02:48.252 INFO (MainThread) [supervisor.os.manager] Fetch OTA update from https://os-artifacts.home-assistant.io/16.2/haos_generic-x86-64-16.2.raucb�[0m
...
�[32m2025-09-08 17:20:29.134 INFO (SyncWorker_2) [supervisor.docker.manager] Restarting homeassistant�[0m
�[33m2025-09-08 17:20:40.700 WARNING (MainThread) [supervisor.api.host] ConnectionError raised when returning journal logs: Connection lost�[0m
�[32m2025-09-08 17:20:54.824 INFO (MainThread) [supervisor.homeassistant.core] Wait until Home Assistant is ready�[0m
�[32m2025-09-08 17:20:56.933 INFO (MainThread) [supervisor.resolution.evaluate] Starting system evaluation with state running�[0m
�[32m2025-09-08 17:20:56.985 INFO (MainThread) [supervisor.resolution.evaluate] System evaluation complete�[0m
�[32m2025-09-08 17:20:59.842 INFO (MainThread) [supervisor.homeassistant.core] Home Assistant Core state changed to APIState(core_state='NOT_RUNNING', offline_db_migration=False)�

Once the update became avialble, I pressed update, but it was in the spinning state for at least 20 minutes, which is unsual as it normally reboots within 2 minutes. So I restarted core, but I tried to update again, but it said a job was already running, so then I rebooted Host, and when it came back, the update was still there, and I pressed update, and it went normally i.e. rebooted on its own after 1 minute, and came back on 16.2.

@bgiuriceo i restarted after 20 minutes while the thing was spinning.

[Anto79-ops]

also saw this in the logs:

Exception ignored in: <coroutine object OSManager.update at 0x7f1a135d2510>
Traceback (most recent call last):
  File "/usr/src/supervisor/supervisor/jobs/decorator.py", line 286, in wrapper
    with job.start():
  File "/usr/local/lib/python3.13/contextlib.py", line 162, in __exit__
    self.gen.throw(value)
  File "/usr/src/supervisor/supervisor/jobs/__init__.py", line 158, in start
    _CURRENT_JOB.reset(token)
ValueError: <Token var=<ContextVar name='current_job' default=None at 0x7f1a160d8d10> at 0x7f1a06b5aa40> was created in a different Context
[23:51:08] WARNING: �[33mHalt Supervisor�[0m

if you want the entire log, just PM me on discord. thanks

[sairon]

Shortly after the Fetch OTA update there should be line saying Completed download of OTA update file /data/tmp/hassos-16.2.raucb. If not, then the download was never finished, resulting probably from some connectivity issues (unfortunately, the CDN used by Github sometimes has bad days). As the update is handled by Supervisor and RAUC service on the host, restarting Core has no effect on that, and the error you saw was likely result of you trying to force the things.

tl;dr - I'd blame the internets, no apparent error OS-side.

[Anto79-ops]

that makes sense to me. I don't see that line, so probably a temporary server issue. thanks!

[PhilGbHa]

Hello, I have to say that as soon as I upgraded to OS 16.2, I noticed completely random "crashes" again (several times per hour but not regularly!) of the system with messages from unknown devices such as GCE IPX V4 integration (even the Weather card) and an inability to find the configuration.yaml file during these crashes whereas before, and without changing anything else, it worked perfectly... Since I returned to OS 15.2 everything has become stable again... Exactly as before with OS 16.0 & the same for 16.1...

[PhilGbHa]

Thank you for your reply, but I don't understand why everything worked perfectly until version 15.2, and I didn't find any idea of ​​how to resolve this problem in the link in your previous message...
Best regards.

[PhilGbHa]

Excuse me for my English, I'm French!.
For your information, here's an exhaustive list of my integrations and add-ons...
Please let me know who is responsible for this bug.

Intégrations au 10/09/25 :

1. Alarmo Version : v1.10.10 [1 appareil ]
2. Application mobile [4 appareils ]
3. Backup [1 service ]
4. Bluetooth [1 appareil ]
5. Browser mod Version : 2.5.2 [3 appareils ]
6. Caméra générique [2 appareil ]
7. Date et heure [1 entité ]
8. DLNA Digital Media Renderer [2 appareils ]
9. DLNA Digital Media Server [3 entrées ]
10. Frigate Version : 5.9.4 [5 appareils ]
11. Frontier Silicon [1 appareil ]
12. GCE Eco-Devices* Version : 5.4.0 [1 appareil ]
13. GCE IPX800 V4 Version : 3.12.0 [11 appareils ]
14. Google Cast [4 appareils ]
15. Google Translate text-to-speech [1 entité ]
16. HACS Version : 2.0.5 [25 services ]
17. Home Assistant Supervisor [22 services ]
18. iBeacon Tracker [2 appareils ]
19. Internet Printing Protocol (IPP) [1 entrée ]
20. LinkPlay [1 appareil ]
21. Lune [1 service]
22. Météo-France [1 service]
23. Meteorologisk institutt (Met.no) [1 service]
24. mobile alerts Version : 1.3.0 [ 4 entités ]
25. MQTT [63 appareils ]
26. Netatmo [19 appareils ]
27. Node-RED Companion Version : 4.1.2 [1 entrée ]
28. Ping (ICMP) [19 appareils ]
29. Radio Browser [1 entrée ]
30. Reolink [9 appareils ]
31. RFXCOM RFXtrx [9 appareils ]
32. Shelly smart home Version : 1.0.10 [14 appareils ]
33. Soleil [1 service ]
34. Synology DSM [12 appareils ]
35. System Monitor [1 service]
36. Tarif EDF Version : 2.1.1 [1 service]
37. Telegram bot [1 service]
38. UPnP/IGD [1 appareil ]
39. Watchman Version : 0.6.5 [1 service]
40. Wyoming Protocol [ 2 entités ]
41. Z-Wave JS [26 appareils ]
42. ZigBee2MQTT [57 appareils ]

Modules complémentaires au 10/09/25 :

1. Duck DNS (1.26.0) { Arrêté car Compte Nabu Casa }
2. ESPHome Device Builder (2025.8.3) { Aucun Périphérique pour le moment }
3. File editor (5.8.0)
4. Frigate (0.16.1)
5. Matter Server (8.1.0) { Arrêté car Aucun Périphérique pour le moment }
6. Mosquitto broker (6.5.2)
7. Node-RED (20.0.0)
8. OpenThread Border Router (2.13.0) { Arrêté car Aucun Périphérique pour le moment }
9. Piper (1.6.4)
10. Samba Backup (5.2.0) { Arrêté car Compte Nabu Casa }
11. Samba share (12.5.2)
12. Silicon Labs Flasher (0.3.4) { Arrêté car Aucun Périphérique pour le moment }
13. Speech-to-Phrase (1.4.1) { Arrêté car Compte Nabu Casa }
14. Studio Code Server (5.19.3)
15. Terminal & SSH (9.19.0)
16. Whisper (2.6.0)
17. Z-Wave JS (0.22.0)
18. Zigbee2MQTT (2.6.1-1)

[PhilGbHa]

Installation method: Home Assistant OS
Generic x86-64 _________Core: 2025.9.1
Supervisor: 2025.09.0 ____Operating System: 15.2
Interface utilisateur: 20250903.3
Sur NUC NIPOGI E1 avec RAM 16Gb NVme:256Go
Home Assistant Cloud, Compte Nabu Casa Connecté

[wolfgangbures]

Maybe this will help you?
#4164 (comment)

[PhilGbHa]

Thank you for your reply, but I don't understand why everything worked perfectly until version 15.2, and I didn't find any idea of ​​how to resolve this problem in the link in your previous message... Also this conversation is closed so there is no way to interact.
Best regards.

[cblt2l]

I recently upgraded my bare metal x86 installation from OS version 16.0 to 16.2 and after rebooting, I'm getting the initial onboarding screen for a new install. I have rebooted a few times and still get the same result. Is it possible that the upgrade overwrote my existing install? I am looking into trying to roll back.

[cblt2l]

So just to follow up - I was able to connect a monitor to my HA box and log in on the command line.

First, I tried switching to the previous boot slot using ha os boot-slot A to revert to 16.1 (My original post said 16.0 but that was wrong) but this did nothing. After that I spent some time and dug through the various partitions looking for any sign that my original install was still there.

Long story short, all of my data was gone. From what I could tell, it looked like a fresh install with none of the original add-ons, backups, or configuration data present. Luckily I had a backup stored on home assistant cloud that I had taken right before the upgrade but if I didn't then I would have been screwed. I don't know what went wrong as I have upgraded the OS many times in the past without issue. Needless to say, I am a little spooked now.

The HA machine in question is a Dell Wyse 5070 for anyone wondering.