Help! OS 16 killed my Home Assistant Yellow?

Installation Home Assistant OS
1

I’m using a Home Assistant Yellow with a RPi CM5 plus NVMe storage, connected via PoE. I just ran the update to update to the new version of Home Assistant OS (to version 16 — prior version had been whatever was current last week; I didn’t note the number)… and my Home Assistant never came back up after the update.

It sits there going through a cycle of showing a red LED for maybe 15 seconds… then I see the red plus an irregularly blinking yellow (plus network traffic) for maybe a minute… then it goes to a solid green for maybe a couple of minutes… then repeats. It is not reachable from the network (though the request times out more quickly when it’s in the “green LED phase” than the “red plus blinking” phase).

My entire home assistant and everything it controls are offline now, so I’d be really grateful for any advice!

Thanks,

Mike

4 Likes
2

I just got mine today and I have tried installing the os16 on it due to the 16gb of ram and the same happens. If I download and install the yellow os 16 image it wont start. If i use the rpi 5 image it works! There must be something wrong with the yellow image. As a new user, i’ve been here about 5 hours trying to understand what i was doing wrong! :frowning:

3

Same issue here, (similar) issue opened! Completely non-booting Yellow!

CM5 64Gb emmc, 16Gb ram, POE
NVME SSD (HASS installed here)
RaZberry 7 Pro in onbord GPIO slot.

After the upgrade mine never came back up. Looking at the LEDs I see a solid green.
When power cycled (via POE) red flashes for a second, then solid green, no other activity.

4

Ughh…

Rebooted, and consoled mine.

# ha core logs
2025/07/09 22:35:17 Start mDNS broadcast
2025/07/09 22:35:17 Start webserver on http://0.0.0.0:8123
2025/07/09 22:35:17 Failed to get outbound IP, retrying in 5s: can't get default interface from Supervisor: {"result":"error","message":"System is not ready with state: setup"}
2025/07/09 22:35:22 Publish http://192.168.2.143:8123 to _home-assistant._tcp
s6-rc: info: service s6rc-oneshot-runner: starting
s6-rc: info: service s6rc-oneshot-runner successfully started
s6-rc: info: service fix-attrs: starting
s6-rc: info: service fix-attrs successfully started
s6-rc: info: service legacy-cont-init: starting
s6-rc: info: service legacy-cont-init successfully started
s6-rc: info: service legacy-services: starting
services-up: info: copying legacy longrun home-assistant (no readiness notification)
s6-rc: info: service legacy-services successfully started

So it lost it’s IP, when I go to http://192.168.2.143:8123 I’m greeted with the dreaded…

image
image1218×1284 65.6 KB

# docker ps
CONTAINER ID   IMAGE                                                       COMMAND               CREATED          STATUS         PORTS                                     NAMES
fb767f8dbc1f   ghcr.io/home-assistant/yellow-homeassistant:2025.7.1        "/init"               5 minutes ago    Up 5 minutes                                             homeassistant
67d24aad1c9f   ghcr.io/home-assistant/aarch64-hassio-multicast:2025.02.0   "/init"               6 minutes ago    Up 6 minutes                                             hassio_multicast
49c5ba73f4fa   ghcr.io/home-assistant/aarch64-hassio-observer:2025.02.0    "/usr/bin/observer"   6 minutes ago    Up 6 minutes   0.0.0.0:4357->80/tcp, [::]:4357->80/tcp   hassio_observer
b19529463baf   ghcr.io/home-assistant/aarch64-hassio-audio:2025.02.0       "/init"               6 minutes ago    Up 6 minutes                                             hassio_audio
ff1b914958e4   ghcr.io/home-assistant/aarch64-hassio-dns:2025.02.0         "/init"               6 minutes ago    Up 6 minutes                                             hassio_dns
6fadb5b2247b   ghcr.io/home-assistant/aarch64-hassio-cli:2025.06.0         "/init"               6 minutes ago    Up 6 minutes                                             hassio_cli
e79539a992be   ghcr.io/home-assistant/aarch64-hassio-supervisor:latest     "/init"               23 minutes ago   Up 6 minutes                                             hassio_supervisor
#

Where’s my instance, config, work?!

Edit: Restoring from a backup now…

1 Like
5

Yikes!

Um I’ve been lazy and haven’t made a backup since some recent work I did that I’d really rather not re-do. (It makes automated backups regularly but they are on the NMVe in the HA yellow.) Presumably I need to pull out the NVMe, connect it to a computer, extract the latest auto backup, then reinstall the NVMe and restore. Can anyone point me towards instructions to follow to do this?

6

Oh boy!..
Yep, me too…
I’m not running on Yellow though, just the regular hassos…

Not sure if this will help anyone else, but after reading an above post by xstrex, I discovered that my instance is also suffering from IP address memory loss…

My instance is running a static IPv4 address.
I was able to access the HA command line interface and change the ip address back with the following commands:

ha> network info
ha> network update eth0 --ipv4-address <xxx.xxx.xxx.xxx>
(replace eth0 and <xxx.xxx.xxx.xxx> with your specific info.

From there I could just go back to the web UI and provide the rest of the network settings (gateway, netmask, DNS servers, etc).

7

I flagged this up the chain for y’all, but just a heads up that most everyone are offline due to time zones.

3 Likes
8

Thank you; I made a post of Reddit with a link here, hopefully it gets some attention, and prevents some issues.

9

Same problem here, but running on an Hyper-V machine, so it’s not just the yellow.

Was working 1h ago, before the OS 16 update.

I tried the “network update eth0 --ipv4-address ” but it’s still not accessible from outside.

10

I have the same issue on Hyper-V. Restored from a backup, still can’t connect.

Not good.

11

Also having the same issues here, Tried to install to a brand new HA Yellow w/ NVMe SSD today, but it failed when trying to install to both NVMe and the EMMC. Flashing 15.2 worked fine, but then when I tried ha os upgrade, it made the HA yellow start boot looping.

12

I was able to get my restore back working with this. Worth a try on OS 16.

network update eth0 --ipv4-method auto
network reload
network info

4 Likes
13

If you have access to the ha cli (ha>) try os info to check the slots and then boot from the other slot with ha os boot other

I didn’t have an issue updating on a PoE Yellow with CM5 + NVME

1 Like
14

@cogneato Thanks. How would I get access to the HA CLI?

15

Hi all,

In general, there is not a generic issue with HAOS 16.0. We’ve tested it in many configuration successfully. We had about 1600 installations on the beta channel testing the pre-releases successfully. I’ve updated my Yellow with CM5 to 16.0 yesterday successfully. That said, it does seem a bit strange that there are multiple reports with update issues on Yellow, so we’ve temporary removed the release from stable channel while investigating.

It seems that there are slightly different outcomes.

@ace_archer are you using CM5 as well? Which variant exactly?

@xstrex can it be that maybe you have two HAOS installations, one on the internal eMMC and one on the NVMe? :thinking: Can you check the output of lsblk and blkid.

@mike15 and any others with non-booting system: Can you attach to the serial console to see if there is any boot log which could help us better understand what did go wrong? This gives access to the HA CLI.

These two links have step by step instructions to hook up the serial console:

3 Likes
16

It seems like some people here have cli access going by the commands being shared, but if you’re in a boot loop situation you may not. (agners beat me to it with the links :point_up: )

17

I just restored a backup to SSD on Raspberry Pi 4.
HA was working (in OS 15.2), then started the upgrade to OS 16.

Now HA will not boot, I get “Home Assistant CLI not starting! Jump into emergency console…”
I tried typing “login” but that doesnt help, just get the same “not starting error” after a while.
Tried “network info” and “network reload”, in both cases I get
“/bin/ash: network: not found”
How should I proceed from here?
Reflash the drive with OS 15.2?

18

Our time out for the CLI startup might be a bit too short for certain platforms. Typically it doesn’t mean anything, just Supervisor still busy doing things in the background (depending on your Internet connection speed and/or GitHub Container Registry servers this might take a couple of minutes).

If you get dropped to the OS shell (which only shows the # character) you need to prepend commands with ha, so ha network info. This command might refuse to run when Supervisor is still busy downloading things.

19

This helped (“os boot other”) now I am running the old supervisor again.
I had the same problem with missing local IP address. I am running Home Assistant on a VM on unraid.

20

Thanks!

Worked like a charm on my Hyper-V installation.

If you want to restore IPv6 connectivity (like I did) then it all becomes:

network update eth0 --ipv4-method auto
network update eth0 --ipv6-method auto
network reload
network info
1 Like