The docker frontend module crashes and restarts when I want to access /dashboard/train/training

[EnzoDeg40]

Operating System

• macOS
• Linux
• Windows

Deployment Method

• Docker
• non-Docker

CUDA Usage

• Yes
• No

Training Process Details (if applicable)

No response

Second-Me version

lastest master

Describe the bug

After simply following the instructions from the README (git clone https://github.com/mindverse/Second-Me.git and make docker-up), I opened http://localhost:3000/ and went through the different steps. However, I cannot access the "Training" step — it shows an error saying the page cannot be loaded.

Current Behavior

Docker module frontend log

> rm -rf .next && next dev --turbo -p 3000


  ▲ Next.js 14.2.25 (turbo)

  - Local:        [http://localhost:3000⁠](http://localhost:3000/)

  - Experiments (use with caution):

    · proxyTimeout


 ✓ Starting...

Attention: Next.js now collects completely anonymous telemetry regarding usage.

This information is used to shape Next.js' roadmap and prioritize features.

You can learn more, including how to opt-out if you'd not like to participate in this anonymous program, by visiting the following URL:

[https://nextjs.org/telemetry⁠](https://nextjs.org/telemetry)


 ✓ Ready in 594ms

 ⚠ Webpack is configured while Turbopack is not, which may cause problems.

 ⚠ See instructions if you need to configure Turbopack:

  [https://nextjs.org/docs/app/api-reference/next-config-js/turbo⁠](https://nextjs.org/docs/app/api-reference/next-config-js/turbo)


 ○ Compiling /dashboard/train/training ...

 ✓ Compiled /dashboard/train/training in 4.4s


> secondme_frontend@0.1.0 dev

> rm -rf .next && next dev --turbo -p 3000


  ▲ Next.js 14.2.25 (turbo)

  - Local:        [http://localhost:3000⁠](http://localhost:3000/)

  - Experiments (use with caution):

    · proxyTimeout


 ✓ Starting...

Attention: Next.js now collects completely anonymous telemetry regarding usage.

This information is used to shape Next.js' roadmap and prioritize features.

You can learn more, including how to opt-out if you'd not like to participate in this anonymous program, by visiting the following URL:

[https://nextjs.org/telemetry⁠](https://nextjs.org/telemetry)


 ✓ Ready in 586ms

 ⚠ Webpack is configured while Turbopack is not, which may cause problems.

 ⚠ See instructions if you need to configure Turbopack:

  [https://nextjs.org/docs/app/api-reference/next-config-js/turbo⁠](https://nextjs.org/docs/app/api-reference/next-config-js/turbo)

Backend log module

SQLite database not found or empty, initializing...

SQLite database initialized successfully

Tables created:

chunk             l1_clusters       memories          status_biography

document          l1_shades         roles             user_llm_configs

l1_bios           l1_versions       space_messages  

l1_chunk_topics   loads             spaces          

Checking ChromaDB...

ChromaDB collections not found, initializing...

/app/lpm_kernel/utils.py:168: SyntaxWarning: invalid escape sequence '\.'

  sentences = re.split("\. |! |\? |。|！|？|\n+ *\n+", text)

Failed to send telemetry event ClientStartEvent: capture() takes 1 positional argument but 3 were given

Failed to send telemetry event ClientCreateCollectionEvent: capture() takes 1 positional argument but 3 were given

Failed to send telemetry event ClientStartEvent: capture() takes 1 positional argument but 3 were given

Failed to send telemetry event ClientCreateCollectionEvent: capture() takes 1 positional argument but 3 were given

Log directory: /app/logs/train

Log file: /app/logs/app.log

2025-08-09 16:07:19 [INFO] logging.py:36 - Logging system initialized successfully

2025-08-09 16:07:19 [INFO] logging.py:53 - Logging module initialization complete

2025-08-09 16:07:19 [INFO] script_executor.py:9 - INFO: ScriptExecutor module loaded

2025-08-09 16:07:19 [WARNING] script_executor.py:10 - WARNING: ScriptExecutor module loaded

2025-08-09 16:07:19 [ERROR] script_executor.py:11 - ERROR: ScriptExecutor module loaded

Log level: 20

Log handlers: ['StreamHandler', 'RotatingFileHandler']

2025-08-09 16:07:19 [INFO] config.py:54 - PyTorch version 2.5.1 available.

2025-08-09 16:07:20 [ERROR] posthog.py:59 - Failed to send telemetry event ClientStartEvent: capture() takes 1 positional argument but 3 were given

2025-08-09 16:07:20 [INFO] embedding_service.py:34 - No embedding model configured, using default dimension: 1536

2025-08-09 16:07:20 [ERROR] posthog.py:59 - Failed to send telemetry event ClientStartEvent: capture() takes 1 positional argument but 3 were given

2025-08-09 16:07:20 [INFO] embedding_service.py:34 - No embedding model configured, using default dimension: 1536

2025-08-09 16:07:20 [INFO] memory_service.py:22 - Initializing storage service, base_dir: /app/data/.

2025-08-09 16:07:20 [INFO] memory_service.py:23 - Raw content directory configuration: /app/data/resources/raw_content

2025-08-09 16:07:20 [INFO] memory_service.py:39 - Storage path created: /app/data/resources/raw_content

2025-08-09 16:07:20 [ERROR] posthog.py:59 - Failed to send telemetry event ClientStartEvent: capture() takes 1 positional argument but 3 were given

2025-08-09 16:07:20 [INFO] embedding_service.py:34 - No embedding model configured, using default dimension: 1536

No embedding model configured, using default dimension: 1536

2025-08-09 16:07:20 [ERROR] posthog.py:59 - Failed to send telemetry event ClientStartEvent: capture() takes 1 positional argument but 3 were given

Collection 'documents' already exists

Collection 'document_chunks' already exists

ChromaDB initialized at ./data/chroma_db

ChromaDB initialized successfully

Expected Behavior

Until the page loads and I can configure my Second Me

Reproduction Steps

I simply opened localhost:3000 and followed the instructions on the page. It's from the training step that the page won't load.

Possible Workaround

No response

Additional Information

No response

Link to related Github discussion or issue

No response

[zerog0]

Same happens under Linux. (ArchLinux)
Works with make setup but not with make docker-up, same behavior

[Nishant891]

It seems to be a possible version mismatch between chromadb and posthog.

Essentially chromadb only supports posthog below v6.0.0 (cf. here), when posthog introduced a breaking API change in capture (only taking one positional arg at that point)

Add this to pyproject.toml posthog = ">=2.4.0,<6.0.0"

Let me know if it works.

[zerog0]

I just tried, but same issue

docker exec in second-me-backend rebuilt image (with venv activated)

(venv) root@63e4852c79b0:/app# pip freeze | egrep 'posthog|chromadb'
chromadb==0.4.24
posthog==5.4.0

[Nishant891]

Did you get the same error?
If yes, It is definitely due to posthog version mismatch since the required arguments number has been changed in versions > 6.0.0
Other than that I do not see any relevant error leading to the crash.

This issue was fixed in chromadb v1.0.15 (PR chroma-core/chroma#4970)
(they patched as in here, forcing to a specific versions of posthog)

Try upgrading the chromadb to v1.0.15
This might help chromadb discussion