Just using docker-compose doesn't work - macbook arm64

[cteodor]

Steps:
git clone https://github.com/AsyncFuncAI/deepwiki-open.git
cd deepwiki-open
echo "GOOGLE_API_KEY=actual-api-key" > .env
docker-compose up

Env:
Darwin MacBookPro.local 24.3.0 Darwin Kernel Version 24.3.0: Thu Jan 2 20:24:16 PST 2025; root:xnu-11215.81.4~3/RELEASE_ARM64_T6000 arm64

Errors:
[...]
[+] Running 2/2
✔ Network deepwiki-open_default Created 0.0s
✔ Container deepwiki-open-deepwiki-1 Created 0.0s
Attaching to deepwiki-open-deepwiki-1
deepwiki-open-deepwiki-1 | 2025-05-01 03:38:36,647 - main - WARNING - Missing environment variables: GOOGLE_API_KEY, OPENAI_API_KEY
deepwiki-open-deepwiki-1 | 2025-05-01 03:38:36,647 - main - WARNING - Some functionality may not work correctly without these variables.
deepwiki-open-deepwiki-1 |
deepwiki-open-deepwiki-1 | > deepwiki-open@0.1.0 start
deepwiki-open-deepwiki-1 | > next start
[...]
deepwiki-open-deepwiki-1 | 2025-05-01 03:38:38,874 - faiss.loader - INFO - Loading faiss.
deepwiki-open-deepwiki-1 | 2025-05-01 03:38:38,886 - faiss.loader - INFO - Successfully loaded faiss.
deepwiki-open-deepwiki-1 | 2025-05-01 03:38:38,888 - faiss - INFO - Failed to load GPU Faiss: name 'GpuIndexIVFFlat' is not defined. Will not load constructor refs for GPU indexes. This is only an error if you're trying to use GPU Faiss.
[...]
deepwiki-open-deepwiki-1 | File "/usr/local/lib/python3.11/site-packages/adalflow/components/model_client/openai_client.py", line 26, in
deepwiki-open-deepwiki-1 | openai = safe_import(OptionalPackages.OPENAI.value[0], OptionalPackages.OPENAI.value[1])
deepwiki-open-deepwiki-1 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
deepwiki-open-deepwiki-1 | File "/usr/local/lib/python3.11/site-packages/adalflow/utils/lazy_import.py", line 224, in safe_import
deepwiki-open-deepwiki-1 | raise ImportError(f"{install_message}")
deepwiki-open-deepwiki-1 | ImportError: Please install openai with: pip install openai

[mkoeck]

I had the same issue. Adding
- GOOGLE_API_KEY=${GOOGLE_API_KEY}
- OPENAI_API_KEY=${OPENAI_API_KEY}
under the environment key in the docker-compose.yml got rid of the missing environment variable error. Adding
openai
to api/requirements.txt got rid of the ImportError.

I had one more issue: "Address already in use"

deepwiki-1 |
deepwiki-1 | > deepwiki-open@0.1.0 start
deepwiki-1 | > next start
deepwiki-1 |
deepwiki-1 | ▲ Next.js 15.3.1
deepwiki-1 | - Local: http://localhost:8001
deepwiki-1 | - Network: http://172.18.0.2:8001
deepwiki-1 |
deepwiki-1 | ✓ Starting...
deepwiki-1 | ✓ Ready in 350ms
deepwiki-1 | 2025-05-01 07:07:56,238 - faiss.loader - INFO - Loading faiss with AVX2 support.
deepwiki-1 | 2025-05-01 07:07:56,252 - faiss.loader - INFO - Successfully loaded faiss with AVX2 support.
deepwiki-1 | 2025-05-01 07:07:56,254 - faiss - INFO - Failed to load GPU Faiss: name 'GpuIndexIVFFlat' is not defined. Will not load constructor refs for GPU indexes. This is only an error if you're trying to use GPU Faiss.
deepwiki-1 | 2025-05-01 07:07:56,514 - main - INFO - Starting Streaming API on port 8001
deepwiki-1 | INFO: Will watch for changes in these directories: ['/app']
deepwiki-1 | ERROR: [Errno 98] Address already in use

I solved this by editing the package.json under the scripts --> start key, changing it to "start": "next start -p 3000"

[League2EB]

I encountered a similar problem when starting up my Mac device.

[sng-asyncfunc]

Published a new bugfix to address the docker-compose up issue, let me know if it works.

[League2EB]

Published a new bugfix to address the docker-compose up issue, let me know if it works.

Regarding the warning has indeed been eliminated

"WARNING - Missing environment variables: GOOGLE_API_KEY, OPENAI_API_KEY"
But I still receive an error message that openai should be installed. The steps I used are the ones below.
May I ask what I am missing or what I have overlooked?

# Clone the repository
git clone https://github.com/AsyncFuncAI/deepwiki-open.git
cd deepwiki-open

# Create a .env file with your API keys
echo "GOOGLE_API_KEY=your_google_api_key" > .env
echo "OPENAI_API_KEY=your_openai_api_key" >> .env

# Run with Docker Compose
docker-compose up

In addition, when I visit http://localhost:8001/, it is normal.
However, I received another error.

Thank you for your patient answers.🙇‍♂️

[sng-asyncfunc]

What repository url are you trying? If it's a private repository you have to provide your github api token.

[spydisec]

It is still not working for me, any idea?

Docker-compose file:
services: deepwiki: build: . ports: - "8001:8001" # API port - "3001:3001" # Next.js port env_file: - .env environment: - PORT=8001 - NODE_ENV=production volumes: - ~/.adalflow:/root/.adalflow # Persist repository and embedding data

Error:
`

deepwiki-open@0.1.0 start
next start -p 3001

▲ Next.js 15.3.1

• Local: http://localhost:3001
• Network: http://172.24.0.2:3001

✓ Starting...
✓ Ready in 1987ms
2025-05-02 00:31:49,519 - faiss.loader - INFO - Loading faiss.
2025-05-02 00:31:49,610 - faiss.loader - INFO - Successfully loaded faiss.
2025-05-02 00:31:49,625 - faiss - INFO - Failed to load GPU Faiss: name 'GpuIndexIVFFlat' is not defined. Will not load constructor refs for GPU indexes. This is only an error if you're trying to use GPU Faiss.
2025-05-02 00:31:51,307 - main - INFO - Starting Streaming API on port 8001
INFO: Will watch for changes in these directories: ['/app']
INFO: Uvicorn running on http://0.0.0.0:8001 (Press CTRL+C to quit)
INFO: Started reloader process [11] using StatReload
2025-05-02 00:31:55,065 - faiss.loader - INFO - Loading faiss.
2025-05-02 00:31:55,121 - faiss.loader - INFO - Successfully loaded faiss.
2025-05-02 00:31:55,135 - faiss - INFO - Failed to load GPU Faiss: name 'GpuIndexIVFFlat' is not defined. Will not load constructor refs for GPU indexes. This is only an error if you're trying to use GPU Faiss.
INFO: Started server process [43]
INFO: Waiting for application startup.
INFO: Application startup complete.
`

[sng-asyncfunc]

@spydisec I think i found out why, we are missing

PORT=3000 or in your case 3001 in the .env

P/S: Pushed a new fix to explicitly specify port 3000.

[League2EB]

What repository url are you trying? If it's a private repository you have to provide your github api token.

https://github.com/openai/codex
I tried to use a public repository.

I suspect that you seem to have overlooked the error message in the screenshot, especially the one that occurred at the end: "deepwiki-1 | ImportError: Please install openai with: pip install openai."

deepwiki-1  | 2025-05-02 01:48:50,702 - faiss - INFO - Failed to load GPU Faiss: name 'GpuIndexIVFFlat' is not defined. Will not load constructor refs for GPU indexes. This is only an error if you're trying to use GPU Faiss.
deepwiki-1  | Traceback (most recent call last):
deepwiki-1  |   File "/usr/local/lib/python3.11/site-packages/adalflow/utils/lazy_import.py", line 222, in safe_import
deepwiki-1  |     return_modules.append(importlib.import_module(module_name))
deepwiki-1  |                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
deepwiki-1  |   File "/usr/local/lib/python3.11/importlib/__init__.py", line 126, in import_module
deepwiki-1  |     return _bootstrap._gcd_import(name[level:], package, level)
deepwiki-1  |            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
deepwiki-1  |   File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
deepwiki-1  |   File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
deepwiki-1  |   File "<frozen importlib._bootstrap>", line 1140, in _find_and_load_unlocked
deepwiki-1  | ModuleNotFoundError: No module named 'openai'
deepwiki-1  | 
deepwiki-1  | During handling of the above exception, another exception occurred:
deepwiki-1  | 
deepwiki-1  | Traceback (most recent call last):
deepwiki-1  |   File "<frozen runpy>", line 198, in _run_module_as_main
deepwiki-1  |   File "<frozen runpy>", line 88, in _run_code
deepwiki-1  |   File "/app/api/main.py", line 32, in <module>
deepwiki-1  |     from api.api import app
deepwiki-1  |   File "/app/api/api.py", line 10, in <module>
deepwiki-1  |     from api.rag import RAG
deepwiki-1  |   File "/app/api/rag.py", line 13, in <module>
deepwiki-1  |     from api.config import configs
deepwiki-1  |   File "/app/api/config.py", line 2, in <module>
deepwiki-1  |     from adalflow.components.model_client.openai_client import OpenAIClient
deepwiki-1  |   File "/usr/local/lib/python3.11/site-packages/adalflow/components/model_client/openai_client.py", line 26, in <module>
deepwiki-1  |     openai = safe_import(OptionalPackages.OPENAI.value[0], OptionalPackages.OPENAI.value[1])
deepwiki-1  |              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
deepwiki-1  |   File "/usr/local/lib/python3.11/site-packages/adalflow/utils/lazy_import.py", line 224, in safe_import
deepwiki-1  |     raise ImportError(f"{install_message}")
deepwiki-1  | ImportError: Please install openai with: pip install openai

[spydisec]

@spydisec I think i found out why, we are missing

PORT=3000 or in your case 3001 in the .env

P/S: Pushed a new fix to explicitly specify port 3000.

The reason I am using 3001 is I already have another app which is running on 3000.

Web app is working but getting error when I put GitHub repo.

[achhabra2]

I found a similar issue. So my problem is when I am running this app on another machine, the frontend is trying to connect to the backend at localhost? Literally just followed the docker instructions and that's it.

[JNLei]

I found a similar issue. So my problem is when I am running this app on another machine, the frontend is trying to connect to the backend at localhost? Literally just followed the docker instructions and that's it.

I have a PR created that should solve this issue: #31