Did not find openai_api_key after SO upgrading

June 06, 2024 08:21

I just upgraded Ubuntu from 20.04 LTS to 22.04 LTS, everything works fine, but HeavyIQ started returning the following error:

Value Error: 1 validation error for ChatOpenAI __root__ Did not find openai_api_key, please add an environment variable `OPENAI_API_KEY` which contains it, or pass `openai_api_key` as a named parameter. (type=value_error)

Comments

5 comments

Gianfranco Campana

June 07, 2024 07:46

Edited
Solved.
I simply pulled an older version, and then pulled again the last one. Working again.
No idea why was that.
0

Comment actions Permalink
Candido Dessanti

June 07, 2024 07:19
Hi Gianfranco,

Thank you for resolving the issue. I'm curious about the steps you took to get and to address the issue.
Did you perform an in-place upgrade of the OS, causing the Docker version of HeavyIQ to stop working?
Also, to restore the original behavior, is it necessary to downgrade the Docker to a version prior to 8.0.1 and then reinstall the 8.0.1 version?

Candido
0

Comment actions Permalink
Gianfranco Campana

June 07, 2024 10:19

Edited
Hi Candido,
here the recap:

I had a Kubuntu 20.04 LTS with HeavyAI Free Enterprise GPU on Docker v.7.0.4. Everything updated in the system.
- I pulled the new 8.0 HeavyAI image, still with the old license key in place: HeavyAI 8.0, of course, returned a license error, and it could not be started:
  Server Error: License failure: Your license version is incompatible with the 8.0.1 release. Please contact HEAVY.AI support to request a version compatible license
- I solved this issue by pulling the 7.2.4 image, upgrading the license key with the new key I got in the meantime from the website, and pulling again the 8.0 version.
- While executing:
  docker-compose up -d
I had:
```
ERROR: for jupyterlab-tmp  'ContainerConfig'
ERROR: for jupyterhub  'ContainerConfig'
....
```
That's because "docker-compose" is deprecated, please substitute it in the documentation with:

"docker compose"

Apart from that, the container was created fine.

At this point in time, I had a fully working HeavyAI and HeavyIQ, in my Docker instance, in Ubuntu 20.04.
Then, I performed an in-place upgrade of the OS to 22.04 LTS.
- The upgrade has been simple and straightforward without errors.
- After the upgrade all but HeavyIQ was working fine.
  HeavyIQ was returning:
  Value Error: 1 validation error for ChatOpenAI __root__ Did not find openai_api_key, please add an environment variable `OPENAI_API_KEY` which contains it, or pass `openai_api_key` as a named parameter. (type=value_error)
- The Docker image of HeavyAI didn't have errors, I had access to Immerse and my data and dashboards.
  I always had this in my log:
```
2024-06-06T14:27:07.447586558Z Backend TCP:  localhost:6274      
2024-06-06T14:27:07.447631630Z Backend HTTP: localhost:6278      
2024-06-06T14:27:07.447642568Z Frontend Web: localhost:6273                        
2024-06-06T14:27:07.447649533Z Calcite TCP:  localhost:6279      
2024-06-06T14:27:07.449455769Z - heavydb 28 started              
2024-06-06T14:27:07.453689483Z - heavy_web_server 31 started     
2024-06-06T14:27:09.810793486Z ? http server started on [::]:6273
2024-06-06T14:27:12.455508871Z Navigate to: http://localhost:6273
2024-06-06T14:27:17.469841005Z HeavyIQ HTTP:  localhost:6275     
2024-06-06T14:27:17.469889656Z - heavy_iq 58 started
```
- At this point I pulled again the 7.2.4 version of the Docker image.
  I made sure everything was working fine.
- Then I pulled again the 8.0.1 version of the Docker image.
  Now, all is working fine including HeavyIQ.
  The error is gone.
0

Comment actions Permalink

Gianfranco Campana

August 07, 2024 15:10
Edited

And, on the 8.1 Docker version, absolutely randomly, I just got again the same error, without any change in the system: two hours ago the error was not showing up.

Here the full log I found for HeavyIQ console: I hope it could help.

Started chromadb server...
Args:
--path storage/rag_storage/chromadb
--port 8009
See logs at chromadb.log
Started monitoring chromaDB server process...
127.0.0.1:53366 - "GET /version.txt HTTP/1.1" 200
Traceback (most recent call last):
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/starlette/_exception_handler.py", line 53, in wrapped_app
    await app(scope, receive, sender)
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/starlette/routing.py", line 74, in app
    response = await func(request)
  File "<frozen heavyiq.api.routes.log_route>", line 21, in custom_route_handler
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/fastapi/routing.py", line 278, in app
    raw_response = await run_endpoint_function(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/fastapi/routing.py", line 191, in run_endpoint_function
    return await dependant.call(**values)
  File "<frozen heavyiq.api.routes.iq_lcel_router>", line 73, in auto_query
  File "<frozen heavyiq.api.handlers.decorators>", line 38, in wrapper
  File "<frozen heavyiq.api.handlers.decorators>", line 36, in wrapper
  File "<frozen heavyiq.api.handlers.decorators>", line 61, in wrapper
  File "<frozen heavyiq.api.handlers.lcel_handler>", line 72, in handle_lcel_auto_query_request
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 2109, in ainvoke
    input = await step.ainvoke(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/passthrough.py", line 447, in ainvoke
    return await self._acall_with_config(self._ainvoke, input, config, **kwargs)
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 1313, in _acall_with_config
    output = await coro
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/passthrough.py", line 434, in _ainvoke
    **await self.mapper.ainvoke(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 2739, in ainvoke
    results = await asyncio.gather(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 2109, in ainvoke
    input = await step.ainvoke(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 4081, in ainvoke
    return await self.bound.ainvoke(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 2109, in ainvoke
    input = await step.ainvoke(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/passthrough.py", line 447, in ainvoke
    return await self._acall_with_config(self._ainvoke, input, config, **kwargs)
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 1313, in _acall_with_config
    output = await coro
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/passthrough.py", line 434, in _ainvoke
    **await self.mapper.ainvoke(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 2739, in ainvoke
    results = await asyncio.gather(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 4081, in ainvoke
    return await self.bound.ainvoke(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 3543, in ainvoke
    return await self._acall_with_config(
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 1313, in _acall_with_config
    output = await coro
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 3490, in _ainvoke
    output = await acall_func_with_variable_args(
  File "<frozen heavyiq.lcel.chains.heavydb.table_chain>", line 60, in get_and_retrieve_table_info
  File "<frozen heavyiq.lcel.chains.heavydb.table_chain>", line 51, in get_tables
  File "<frozen heavyiq.lcel.chains.heavydb.table_chain>", line 46, in retrieve_table_names_from_vectordb
  File "<frozen heavyrag.main>", line 3, in <module>
  File "<frozen heavyrag.main>", line 13, in <module>
  File "<frozen heavyrag.evaluate>", line 3, in <module>
  File "<frozen heavyrag.evaluate>", line 8, in <module>
  File "<frozen heavyrag.llm>", line 93, in get_llm
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/cachetools/__init__.py", line 741, in wrapper
    v = func(*args, **kwargs)
  File "<frozen heavyiq.langchain.llms>", line 100, in get_llm_by_type
  File "<frozen heavyiq.langchain.llms>", line 190, in get_openai_llm_by_model_name
  File "<frozen heavyiq.langchain.llms>", line 222, in _get_openai_chat_llm
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/langchain_core/load/serializable.py", line 120, in __init__
    super().__init__(**kwargs)
  File "/opt/heavyai/heavyiq/.venv/lib/python3.10/site-packages/pydantic/v1/main.py", line 341, in __init__
    raise validation_error
pydantic.v1.error_wrappers.ValidationError: 1 validation error for ChatOpenAI
__root__
  Did not find openai_api_key, please add an environment variable `OPENAI_API_KEY` which contains it, or pass `openai_api_key` as a named parameter. (type=value_error)

127.0.0.1:53372 - "POST /api/v1/lcel/auto/query HTTP/1.1" 500

Gianfranco Campana

August 14, 2024 07:55

Edited
This error keeps reoccurring with the frequence of one time every 1 to 3 days.

The only thing I can do is downgrading to 7.x.x version, and upgrading again and recreating the container to 8.1.0 (and don't forgive every time to recreate the ingesting folder in whitelist).

Quite annoying, and periodic downtime, that leads to missing some data updates.
0

Comment actions Permalink

Please sign in to leave a comment.

Comments

Didn't find what you were looking for?