Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 22:23:50 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #2421] [BUG]: Docker container locks up and becomes unhealthy after uploading document #1574

New Issue
Closed
opened 2026-02-22 18:25:28 -05:00 by yindo · 3 comments
yindo commented 2026-02-22 18:25:28 -05:00
Owner
Copy Link
Copy Source

Originally created by @mortenfj on GitHub (Oct 4, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2421

How are you running AnythingLLM?

Docker (remote machine)

What happened?

Description:

I encountered an issue while using the AnythingLLM Docker container. After creating a workspace and uploading a small document, the platform locks up and eventually the container goes to "unhealthy" status.

Steps to reproduce:

  1. Spin up the AnythingLLM Docker container.
  2. Create a user account and a workspace (named "test" in my case).
  3. Upload a small document (e.g., a 2k .md file).
  4. Add the document to the workspace.

Expected behavior:
The document should be successfully added to the workspace, and the platform should continue functioning as normal.

Actual behavior:
After uploading and attempting to add the document to the workspace, the platform locks up and becomes unresponsive. Eventually, the Docker container goes into an "unhealthy" state.

Container logs:

[backend] info: [TELEMETRY SENT] {"event":"document_uploaded","distinctId":"9093e490-272d-4ef4-bae4-994906775c0b","properties":{"runtime":"docker"}}
[backend] info: [Event Logged] - document_uploaded
[backend] info: [TELEMETRY SENT] {"event":"workspace_thread_created","distinctId":"9093e490-272d-4ef4-bae4-994906775c0b::1","properties":{"multiUserMode":true,"LLMSelection":"generic-openai","Embedder":"native","VectorDbSelection":"lancedb","TTSSelection":"native","runtime":"docker"}}
[backend] info: [Event Logged] - workspace_thread_created
[backend] info: Adding new vectorized document into namespace test
[backend] info: [NativeEmbedder] Initialized
[backend] info: [RecursiveSplitter] Will split with {"chunkSize":1000,"chunkOverlap":20}
[backend] info: Chunks created from document: 1
Error: Failed to load model because protobuf parsing failed.
    at new OnnxruntimeSessionHandler (/app/server/node_modules/onnxruntime-node/dist/backend.js:27:92)
    at /app/server/node_modules/onnxruntime-node/dist/backend.js:64:29
    at process.processTicksAndRejections (node:internal/process/task_queues:77:11)
Something went wrong during model construction (most likely a missing operation). Using `wasm` as a fallback.
[backend] info: [NativeEmbedder] Failed to download model from primary URL. Using fallback https://s3.us-west-1.amazonaws.com/public.useanything.com/support/models/
Error: Failed to load model because protobuf parsing failed.
    at new OnnxruntimeSessionHandler (/app/server/node_modules/onnxruntime-node/dist/backend.js:27:92)
    at /app/server/node_modules/onnxruntime-node/dist/backend.js:64:29
    at process.processTicksAndRejections (node:internal/process/task_queues:77:11)
Something went wrong during model construction (most likely a missing operation). Using `wasm` as a fallback.

Environment:

  • Docker runtime
  • Workspace: multiUserMode = true
  • Vector DB: LanceDB
  • Embedder: NativeEmbedder

Additional information:
The issue seems to stem from an error in loading the ONNX model, where protobuf parsing fails, forcing a fallback to a WebAssembly (WASM) implementation. Despite this fallback, the platform becomes unresponsive, and the container becomes unhealthy.

Are there known steps to reproduce?

No response

Originally created by @mortenfj on GitHub (Oct 4, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2421 ### How are you running AnythingLLM? Docker (remote machine) ### What happened? **Description:** I encountered an issue while using the AnythingLLM Docker container. After creating a workspace and uploading a small document, the platform locks up and eventually the container goes to "unhealthy" status. Steps to reproduce: 1. Spin up the AnythingLLM Docker container. 2. Create a user account and a workspace (named "test" in my case). 3. Upload a small document (e.g., a 2k .md file). 4. Add the document to the workspace. **Expected behavior:** The document should be successfully added to the workspace, and the platform should continue functioning as normal. **Actual behavior:** After uploading and attempting to add the document to the workspace, the platform locks up and becomes unresponsive. Eventually, the Docker container goes into an "unhealthy" state. **Container logs:** ```log [backend] info: [TELEMETRY SENT] {"event":"document_uploaded","distinctId":"9093e490-272d-4ef4-bae4-994906775c0b","properties":{"runtime":"docker"}} [backend] info: [Event Logged] - document_uploaded [backend] info: [TELEMETRY SENT] {"event":"workspace_thread_created","distinctId":"9093e490-272d-4ef4-bae4-994906775c0b::1","properties":{"multiUserMode":true,"LLMSelection":"generic-openai","Embedder":"native","VectorDbSelection":"lancedb","TTSSelection":"native","runtime":"docker"}} [backend] info: [Event Logged] - workspace_thread_created [backend] info: Adding new vectorized document into namespace test [backend] info: [NativeEmbedder] Initialized [backend] info: [RecursiveSplitter] Will split with {"chunkSize":1000,"chunkOverlap":20} [backend] info: Chunks created from document: 1 Error: Failed to load model because protobuf parsing failed. at new OnnxruntimeSessionHandler (/app/server/node_modules/onnxruntime-node/dist/backend.js:27:92) at /app/server/node_modules/onnxruntime-node/dist/backend.js:64:29 at process.processTicksAndRejections (node:internal/process/task_queues:77:11) Something went wrong during model construction (most likely a missing operation). Using `wasm` as a fallback. [backend] info: [NativeEmbedder] Failed to download model from primary URL. Using fallback https://s3.us-west-1.amazonaws.com/public.useanything.com/support/models/ Error: Failed to load model because protobuf parsing failed. at new OnnxruntimeSessionHandler (/app/server/node_modules/onnxruntime-node/dist/backend.js:27:92) at /app/server/node_modules/onnxruntime-node/dist/backend.js:64:29 at process.processTicksAndRejections (node:internal/process/task_queues:77:11) Something went wrong during model construction (most likely a missing operation). Using `wasm` as a fallback. ``` **Environment:** - Docker runtime - Workspace: multiUserMode = true - Vector DB: LanceDB - Embedder: NativeEmbedder **Additional information:** The issue seems to stem from an error in loading the ONNX model, where protobuf parsing fails, forcing a fallback to a WebAssembly (WASM) implementation. Despite this fallback, the platform becomes unresponsive, and the container becomes unhealthy. ### Are there known steps to reproduce? _No response_
yindo added the possible bug label 2026-02-22 18:25:28 -05:00
yindo closed this issue 2026-02-22 18:25:29 -05:00
yindo commented 2026-02-22 18:25:29 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Oct 4, 2024):

This is because the local embedding model could not be downloaded - we first try to download from HF, then fallback to a bucket we duplicated the information to because of VPN issues from HF.

This model only downloads on its first use and is reused from then on. If the model cannot be streamed or downloaded then the embedding will always fail.

Are you offline or on a restricted corporate network? You should see a Xenova folder in your AnythingLLM container storage/models/ folder.

@timothycarambat commented on GitHub (Oct 4, 2024): This is because the local embedding model could not be downloaded - we first try to download from HF, then fallback to a bucket we duplicated the information to because of VPN issues from HF. This model _only downloads on its first use_ and is reused from then on. If the model cannot be streamed or downloaded then the embedding will always fail. Are you offline or on a restricted corporate network? You should see a `Xenova` folder in your AnythingLLM container storage/models/ folder.
yindo commented 2026-02-22 18:25:29 -05:00
Author
Owner
Copy Link
Copy Source

@kmccullor commented on GitHub (Feb 24, 2025):

I ran into a similar issue trying to upload the first document.

anythingllm | Error: Failed to load model because protobuf parsing failed.
anythingllm | at new OnnxruntimeSessionHandler (/app/server/node_modules/onnxruntime-node/dist/backend.js:27:92)
anythingllm | at /app/server/node_modules/onnxruntime-node/dist/backend.js:64:29
anythingllm | at process.processTicksAndRejections (node:internal/process/task_queues:77:11)
anythingllm | Something went wrong during model construction (most likely a missing operation). Using wasm as a fallback.
anythingllm | [backend] info: Caching vectorized results of custom-documents/RNI-4.14.1-Release-Notes.pdf-c2895696-947d-43a6-b6eb-84f86f52c179.json to pr

@kmccullor commented on GitHub (Feb 24, 2025): I ran into a similar issue trying to upload the first document. anythingllm | Error: Failed to load model because protobuf parsing failed. anythingllm | at new OnnxruntimeSessionHandler (/app/server/node_modules/onnxruntime-node/dist/backend.js:27:92) anythingllm | at /app/server/node_modules/onnxruntime-node/dist/backend.js:64:29 anythingllm | at process.processTicksAndRejections (node:internal/process/task_queues:77:11) anythingllm | Something went wrong during model construction (most likely a missing operation). Using `wasm` as a fallback. anythingllm | [backend] info: Caching vectorized results of custom-documents/RNI-4.14.1-Release-Notes.pdf-c2895696-947d-43a6-b6eb-84f86f52c179.json to pr
yindo commented 2026-02-22 18:25:29 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 25, 2025):

@kmccullor Failed to download local embedder model (for any reason)
https://github.com/Mintplex-Labs/anything-llm/issues/3031

@timothycarambat commented on GitHub (Feb 25, 2025): @kmccullor Failed to download local embedder model (for any reason) https://github.com/Mintplex-Labs/anything-llm/issues/3031
yindo changed title from [BUG]: Docker container locks up and becomes unhealthy after uploading document to [GH-ISSUE #2421] [BUG]: Docker container locks up and becomes unhealthy after uploading document 2026-06-05 14:41:29 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Desktop
Docker
Integration Request
Integration Request
OS: Linux
OS: Mobile
OS: Windows
UI/UX
blocked
bug
bug
core-team-only
documentation
duplicate
embed-widget
enhancement
feature request
github_actions
good first issue
investigating
needs info / can't replicate
possible bug
pull-request
Mirrored from GitHub Pull Request
 question
stage: specifications
wontfix
No Label possible bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#1574
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?