After update from 2025.8.3 to 2025.9.4, the HA is restarting itself

Hello guys,

I wanted to ask for litle help. After I did update from 2025.8.3 to 2025.9.4 I do have issue with stability of my home assistant. It is restarting itself randomly. 3-4x per day.
My Home Assistant is running on Synology NAS DS1621+ (32gb ram) in Docker container with no limit set (in docker compose) for resources.

The error message I was able to collect is:

INFO: Home Assistant Core finish process received signal 11
INFO: Home Assistant Core finish process exit code 256

I have debug level of logging but I just see here normal activity. No error other messages before this.

My docker compose file:

version: ‘3.9’

services:
homeassistant:
container_name: home-assistant
image: homeassistant/home-assistant:latest
restart: always
network_mode: host
volumes:
- /volume2/docker/HomeAssistant:/config
- /volume1/music:/media/music
environment:
- TZ=Europe/Prague
- PYTHONFAULTHANDLER=1

Hello Ondrej,

This might help.

Thanks,

I did use that link and it is definnetly memory leak. I was not able to identify it.
Interesting is that the old version 2025.8.3 does not have that issue.

Logger: homeassistant.components.profiler
Zdroj: components/profiler/init.py:487
integrace: Profiler (dokumentace, problémy)
První výskyt: 15:43:50 (12 výskytů)
Naposledy logováno: 15:49:21

  • Memory Growth:
  • Memory Growth: [(‘ReadOnlyDict’, 6651, 1)]
  • Memory Growth: [(‘LogRecord’, 8, 7), (‘Event’, 1273, 4), (‘tuple_iterator’, 2, 1)]
  • Memory Growth: [(‘dict’, 148335, 76)]
  • Memory Growth: [(‘set’, 28905, 15), (‘States’, 61, 15), (‘InstanceState’, 62, 15), (‘Struct’, 1029, 2), (‘ReceiveMessage’, 232, 1)]

and

Logger: homeassistant.components.profiler
Zdroj: components/profiler/init.py:487
integrace: Profiler (dokumentace, problémy)
První výskyt: 15:43:50 (35 výskytů)
Naposledy logováno: 16:00:51

  • Memory Growth: [(‘dict’, 149272, 11)]
  • Memory Growth: [(‘dict’, 149324, 52), (‘URL’, 219, 5), (‘MultiDictProxy’, 24, 5), (‘MultiDict’, 25, 5), (‘frozenset’, 12136, 3), (‘Statement’, 91, 2), (‘RawRequestMessage’, 15, 1), (‘CIMultiDict’, 160, 1), (‘CIMultiDictProxy’, 147, 1), (‘StreamWriter’, 12, 1), (‘TemplateState’, 1433, 1), (‘RequestHandler’, 11, 1), (‘HomeAssistantRequest’, 10, 1), (‘AccessLogger’, 11, 1), (‘UrlMappingMatchInfo’, 11, 1)]
  • Memory Growth: [(‘Context’, 7207, 20), (‘dict’, 149331, 7), (‘State’, 6763, 4), (‘function’, 153857, 3), (‘_GenericAlias’, 2701, 2), (‘module’, 6370, 1), (‘ModuleSpec’, 6338, 1), (‘SourceFileLoader’, 6086, 1), (‘ForwardRef’, 4154, 1), (‘Script’, 108, 1)]
  • Memory Growth: [(‘StateAttributes’, 9, 9), (‘function’, 153858, 1), (‘cell’, 83228, 1), (‘Connection’, 11, 1), (‘Script’, 109, 1), (‘MDNSServiceInfo’, 4, 1), (‘EventData’, 1, 1)]
  • Memory Growth: [(‘cell’, 83235, 7), (‘TemplateState’, 1437, 4), (‘function’, 153861, 3), (‘HassJob’, 3652, 2), (‘partial’, 6886, 1)]

Logger: homeassistant.components.profiler
Zdroj: components/profiler/init.py:487
integrace: Profiler (dokumentace, problémy)
První výskyt: 15:43:50 (48 výskytů)
Naposledy logováno: 16:07:21

  • Memory Growth: [(‘list’, 84289, 20), (‘hamt’, 69, 1), (‘hamt_bitmap_node’, 70, 1), (‘Script’, 118, 1), (‘TimerHandle’, 203, 1)]
  • Memory Growth: [(‘Script’, 119, 1)]
  • Memory Growth: [(‘_ThreadHandle’, 34, 2), (‘Thread’, 24, 2), (‘Empty’, 20, 2), (‘traceback’, 35, 2), (‘function’, 153862, 1), (‘cell’, 83236, 1), (‘Condition’, 78, 1), (‘lock’, 143, 1)]
  • Memory Growth:
  • Memory Growth: [(‘cell’, 83254, 18), (‘function’, 153865, 3), (‘builtin_function_or_method’, 15985, 2), (‘CIMultiDict’, 162, 2), (‘Task’, 43, 1), (‘HassClientResponse’, 5, 1), (‘RequestInfo’, 5, 1), (‘TimeoutHandle’, 4, 1), (‘_GatheringFuture’, 2, 1), (‘_BaseRequestContextManager’, 2, 1), (‘ReceiveMessage’, 237, 1), (‘JsonPayload’, 1, 1), (‘BaseCookie’, 1, 1), (‘ClientRequest’, 1, 1)]

Hello, have you found any solution?
I have the same issue after installing version 2025.9.0 or newer — HA keeps crashing. Even when I try to set up a new VM with HAOS from scratch, the problem persists.
I’m running Proxmox v9 with HAOS.

I can confirm the same. Since the latest update (Sept. 20th), my Home Assistant is constantly eating memory to the point that the machine starts swapping and eventually making it unresponsive. This hasn’t happend before the update. See the memory usage graph of my HASS-box showing stable memory usage up to the 20th and since then constantly eating memory (the drops in the graphs were manual reboots from my side due to the unresponsiveness).

I have the same issue, is there any fix for this?

Did it get fixed? I am still on 2025.8.3 precisely for this reason

It did get fixed. Should be closed now