github-starred/open-webui
2
0
Fork 0
mirror of https://github.com/open-webui/open-webui.git synced 2026-05-06 19:08:59 -05:00
Code Issues 1.3k Packages Projects Releases 126 Wiki Activity

[GH-ISSUE #22251] issue: .split error causes no responses in chat, stop sequences and possibly MCP servers involved (v0.8.8) #19673

New Issue
Closed
opened 2026-04-20 02:11:21 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-20 02:11:21 -05:00
Owner
Copy Link

Originally created by @reversewave on GitHub (Mar 5, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/22251

Check Existing Issues

  • I have searched for any existing and/or related issues.
  • I have searched for any existing and/or related discussions.
  • I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!).
  • I am using the latest version of Open WebUI.

Installation Method

Docker

Open WebUI Version

v0.8.8

Ollama Version (if applicable)

N/A

Operating System

Windows 11 Pro

Browser (if applicable)

Vivaldi 7.8.3925.76 (Official Build) (64-bit)

Confirmation

  • I have read and followed all instructions in README.md.
  • I am using the latest version of both Open WebUI and Ollama.
  • I have included the browser console logs.
  • I have included the Docker container logs.
  • I have provided every relevant configuration, setting, and environment variable used in my setup.
  • I have clearly listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc).
  • I have documented step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation. My steps:
  • Start with the initial platform/version/OS and dependencies used,
  • Specify exact install/launch/configure commands,
  • List URLs visited, user input (incl. example values/emails/passwords if needed),
  • Describe all options and toggles enabled or changed,
  • Include any files or environmental changes,
  • Identify the expected and actual result at each stage,
  • Ensure any reasonably skilled user can follow and hit the same issue.

Expected Behavior

Chat should work normally. Messages should send and get responses back. Having a stop sequence set on a model should not cause the frontend to crash and return no response.

Actual Behavior

Chat stopped returning responses entirely with no clear trigger. No generation request ever reached the backend, meaning the crash was happening in the browser before anything was sent. The .split TypeError appeared in the console every time a message was sent. This happened across every model tested, not just one. The fix was going to Settings -> General -> Advanced Parameters, and removing the stop sequence. However chat was already broken before any stop sequences were added, so MCP tool server misconfiguration may have also played a role.

Steps to Reproduce

  1. Install Open WebUI v0.8.8 via Docker on Windows
  2. Add an external OpenAI-compatible API in Admin Settings -> Connections (I've used https://api.linkapi.ai/v1, Bearer token auth)
  3. Set up two MCP tool servers in Admin Settings -> Integrations: DuckDuckGo at localhost:8001 and mcp-server-fetch at localhost:8002, both using uvx mcpo via a batch file
  4. Go to Settings -> General -> Advanced Parameters, and add any stop sequence (tested with n/n/Human:, User:, and random strings like dkdkd)
  5. Open a new chat and send any message
  6. No response comes back
  7. Open DevTools (F12) -> Console and observe the .split TypeError on every send attempt
  8. To confirm fix: go back to Settings -> General -> Advanced Parameters, remove the stop sequence, and chat works again

Logs & Screenshots

Image Image Image Image Image

Additional Information

This looks related to #19486 and #19500 which had the same .split error in the tool/middleware layer. In those issues a workaround was to add a single comma to the Function Name Filter List in the external tool server settings. I tried this and it did not fix the issue in my case. The root cause appears to be stop sequences being set on models, though MCP server misconfiguration may also be a contributing factor.

Full troubleshooting conversation with the Open WebUI Kapa.AI bot on Discord.

My setup:

  • External API: https://api.linkapi.ai/v1 (OpenAI-compatible, Bearer token auth)
  • Two local MCP servers via batch file using uvx mcpo: DuckDuckGo on port 8001, mcp-server-fetch on port 8002
  • Docker run command:
docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
Originally created by @reversewave on GitHub (Mar 5, 2026). Original GitHub issue: https://github.com/open-webui/open-webui/issues/22251 ### Check Existing Issues - [x] I have searched for any existing and/or related issues. - [x] I have searched for any existing and/or related discussions. - [x] I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!). - [x] I am using the latest version of Open WebUI. ### Installation Method Docker ### Open WebUI Version v0.8.8 ### Ollama Version (if applicable) N/A ### Operating System Windows 11 Pro ### Browser (if applicable) Vivaldi 7.8.3925.76 (Official Build) (64-bit) ### Confirmation - [x] I have read and followed all instructions in `README.md`. - [x] I am using the latest version of **both** Open WebUI and Ollama. - [x] I have included the browser console logs. - [x] I have included the Docker container logs. - [x] I have **provided every relevant configuration, setting, and environment variable used in my setup.** - [x] I have clearly **listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup** (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc). - [x] I have documented **step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation**. My steps: - Start with the initial platform/version/OS and dependencies used, - Specify exact install/launch/configure commands, - List URLs visited, user input (incl. example values/emails/passwords if needed), - Describe all options and toggles enabled or changed, - Include any files or environmental changes, - Identify the expected and actual result at each stage, - Ensure any reasonably skilled user can follow and hit the same issue. ### Expected Behavior Chat should work normally. Messages should send and get responses back. Having a stop sequence set on a model should not cause the frontend to crash and return no response. ### Actual Behavior Chat stopped returning responses entirely with no clear trigger. No generation request ever reached the backend, meaning the crash was happening in the browser before anything was sent. The `.split` TypeError appeared in the console every time a message was sent. This happened across every model tested, not just one. The fix was going to **Settings -> General -> Advanced Parameters**, and removing the stop sequence. However chat was already broken before any stop sequences were added, so MCP tool server misconfiguration may have also played a role. ### Steps to Reproduce 1. Install Open WebUI v0.8.8 via Docker on Windows 2. Add an external OpenAI-compatible API in **Admin Settings -> Connections** (I've used `https://api.linkapi.ai/v1`, Bearer token auth) 3. Set up two MCP tool servers in **Admin Settings -> Integrations**: DuckDuckGo at `localhost:8001` and mcp-server-fetch at `localhost:8002`, both using `uvx mcpo` via a batch file 4. Go to **Settings -> General -> Advanced Parameters**, and add any stop sequence (tested with `n/n/Human:`, `User:`, and random strings like `dkdkd`) 5. Open a new chat and send any message 6. No response comes back 7. Open DevTools (F12) -> Console and observe the `.split` TypeError on every send attempt 8. To confirm fix: go back to **Settings -> General -> Advanced Parameters**, remove the stop sequence, and chat works again ### Logs & Screenshots <img width="1472" height="703" alt="Image" src="https://github.com/user-attachments/assets/4845eabf-dece-41c5-b0b9-3573e77a436e" /> <img width="1006" height="835" alt="Image" src="https://github.com/user-attachments/assets/8d01d52d-2bb4-4a43-8d28-5161d2deff25" /> <img width="1847" height="781" alt="Image" src="https://github.com/user-attachments/assets/9b34e16e-b9f8-490c-9973-243612cabf98" /> <img width="1850" height="788" alt="Image" src="https://github.com/user-attachments/assets/8266fecf-4234-4dcb-8b40-d81d26573ad0" /> <img width="1849" height="786" alt="Image" src="https://github.com/user-attachments/assets/3a00ba08-feba-4513-b414-99a060d8b191" /> ### Additional Information This looks related to #19486 and #19500 which had the same `.split` error in the tool/middleware layer. In those issues a workaround was to add a single comma to the Function Name Filter List in the external tool server settings. I tried this and it did not fix the issue in my case. The root cause appears to be stop sequences being set on models, though MCP server misconfiguration may also be a contributing factor. Full troubleshooting [conversation](https://discord.com/channels/1170866489302188073/1478824202457845873 ) with the Open WebUI Kapa.AI bot on Discord. **My setup:** - External API: `https://api.linkapi.ai/v1` (OpenAI-compatible, Bearer token auth) - Two local MCP servers via batch file using `uvx mcpo`: DuckDuckGo on port 8001, mcp-server-fetch on port 8002 - Docker run command: ```js docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main ```
GiteaMirror added the bug label 2026-04-20 02:11:21 -05:00
GiteaMirror commented 2026-04-20 02:11:23 -05:00
Author
Owner
Copy Link

@tjbck commented on GitHub (Mar 7, 2026):

Should be fixed in dev.

<!-- gh-comment-id:4017563145 --> @tjbck commented on GitHub (Mar 7, 2026): Should be fixed in dev.
GiteaMirror commented 2026-04-20 02:11:24 -05:00
Author
Owner
Copy Link

@Classic298 commented on GitHub (Mar 7, 2026):

c7d1d1e390

<!-- gh-comment-id:4017564970 --> @Classic298 commented on GitHub (Mar 7, 2026): https://github.com/open-webui/open-webui/commit/c7d1d1e390a79c6c86d4bfe439fd7de6f5fb060f
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
confirmed
confirmed issue
core
documentation
enhancement
good first issue
help wanted
non-core
pull-request
Mirrored from GitHub Pull Request
 python
question
testing wanted
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#19673
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?