[bug]: OC\Group\Manager::getUserGroupIds(): Argument nextcloud/summary_bot#1 ($user) must be of type OCP\IUser, null given

[LexioJ]

How to use GitHub

• Please use the 👍 reaction to show that you are affected by the same issue.
• Please don't comment if you have no relevant information to add. It's just extra noise for everyone subscribed to this issue.
• Subscribe to receive notifications on status change and new comments.

---

Steps to reproduce

1. type @summary
2. wait for local llm text2text (ollama)
3. no summary provided

Expected behaviour

Summary being generated

Actual behaviour

Could not get a summary from any large language model

All other apps (assistant, ai_bot, context_agent) using the same llm backend are working fine.

[kyteinsky]

hello,
I trust the summary bot was added to the Talk room and then there were some messages? The bot will only be able to summarise the messages that are sent after it was added to a room.
If the above was done, would you mind providing the logs from the docker container?
docker logs nc_app_summary_bot

[LexioJ]

TRACE: 127.0.0.1:51658 - HTTP connection made
TRACE: 127.0.0.1:51658 - ASGI [7] Started scope={'type': 'http', 'asgi': {'version': '3.0', 'spec_version': '2.3'}, 'http_version': '1.1', 'server': ('127.0.0.1', 23001), 'client': ('127.0.0.1', 51658), 'schem
e': 'http', 'root_path': '', 'headers': '<...>', 'state': {}, 'method': 'POST', 'path': '/summary_bot', 'raw_path': b'/summary_bot', 'query_string': b''}
TRACE: 127.0.0.1:51658 - ASGI [7] Receive {'type': 'http.request', 'body': '<465 bytes>', 'more_body': False}
TRACE: 127.0.0.1:51658 - ASGI [7] Send {'type': 'http.response.start', 'status': 200, 'headers': '<...>'}
INFO: 127.0.0.1:51658 - "POST /summary_bot HTTP/1.1" 200 OK
TRACE: 127.0.0.1:51658 - ASGI [7] Send {'type': 'http.response.body', 'body': '<0 bytes>'}
TRACE: 127.0.0.1:51658 - ASGI [7] Completed
TRACE: 127.0.0.1:51658 - HTTP connection lost
2025-11-25 00:10:42,029 - DEBUG - Task with ID 597 created
2025-11-25 00:10:47,155 - DEBUG - Task (597) status: STATUS_RUNNING
2025-11-25 00:10:52,285 - DEBUG - Task (597) status: STATUS_RUNNING
2025-11-25 00:10:57,418 - DEBUG - Task (597) status: STATUS_FAILED
2025-11-25 00:10:57,418 - ERROR - An error occurred: Could not get a summary from any large language model
Traceback (most recent call last):
File "/app/lib/main.py", line 279, in last_x_duration_process
summary = ocs_get_summary(formatted_chat_messages, message.conversation_name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/lib/main.py", line 183, in ocs_get_summary
raise LLMException("Nextcloud TaskProcessing Task failed: " + task["status"])
main.LLMException: Nextcloud TaskProcessing Task failed: STATUS_FAILED

[kyteinsky]

thanks for the logs.

An error occurred: Could not get a summary from any large language model

maybe this error can be worded better. It means that the LLM provider is not working nicely on your instance. LLM providers could be llm2, integration_openai, etc.
the server logs should have an error message regarding the exact failure reason. Alternately, occ taskprocessing:list can show the error. The list is from oldest to newest tasks.

[LexioJ]

I replaced sensitive content with CAPITAL - the values are submitted correctly (ROOM NAME, USERS DISPLAY NAME, CHAT CONTENT)

---

Here is the chat log from the room called "HERE IS MY ROOM NAME" that you should summarize, do not mention the room explicitly:

'''
<msg><ts>2025-11-25 00:05:30</ts><at>HERE IS THE USERS DISPLAY NAME</at><cnt>HERE IS MY CHAT CONTENT</cnt></msg>

'''

Now, please provide an insighful summary of the provided chat log and use human-readable time references for time related information.
Use bullet points to list the most important facts and keep the summary concise and readable in roughly 30 seconds.

    - output
    - customId:
    - completionExpectedAt: 1764058062
    - progress: 0
    - scheduledAt: 1764057980
    - startedAt: 1764057980
    - endedAt: 1764057990
    - error_message: OC\Group\Manager::getUserGroupIds(): Argument nextcloud/summary_bot#1 ($user) must be of type OCP\IUser, null given, called in /var/www/nextcloud/apps/integration_openai/lib/Service/QuotaRuleService.php on line 50

---

Hope this helps investigate the issue. Many thanks in advanced.

[kyteinsky]

yes, thanks a lot!
it looks like a bug in integration_openai, transfering this issue there to be solved.