[GH-ISSUE #15688] issue: /api/v1/chats endpoint stops functioning on Production after about 5-6 chat messages (403 for each request) #17639

New Issue

GiteaMirror · 2026-04-19T23:26:49-05:00

GiteaMirror commented

2026-04-19 23:26:49 -05:00

Originally created by @G1anduin on GitHub (Jul 13, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/15688

Check Existing Issues

I have searched the existing issues and discussions.
I am using the latest version of Open WebUI.

Installation Method

Git Clone

Open WebUI Version

0.5.2

Ollama Version (if applicable)

latest

Operating System

MacOS Sequoia 15.5 (24F74)

Browser (if applicable)

Arc, Chrome, Firefox (all)

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

Using the chat for more than a few messages

Actual Behavior

Chat endpoints stop functioning (return 403 forbidden for each API request) after around 5-6 messages, in every chat, always - quicker if the messages are long.

Steps to Reproduce

Starting to use chat, after 5-6 messages the next POST request (for a submitted prompt) to {BASE_URL}/api/v1/chats/{CHAT_ID} results in Forbidden 403 and so does every subsequent API request

Logs & Screenshots

Additional Information

No response

Originally created by @G1anduin on GitHub (Jul 13, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/15688 ### Check Existing Issues - [x] I have searched the existing issues and discussions. - [ ] I am using the latest version of Open WebUI. ### Installation Method Git Clone ### Open WebUI Version 0.5.2 ### Ollama Version (if applicable) latest ### Operating System MacOS Sequoia 15.5 (24F74) ### Browser (if applicable) Arc, Chrome, Firefox (all) ### Confirmation - [x] I have read and followed all instructions in `README.md`. - [ ] 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 Using the chat for more than a few messages ### Actual Behavior Chat endpoints stop functioning (return 403 forbidden for each API request) after around 5-6 messages, in every chat, always - quicker if the messages are long. ### Steps to Reproduce Starting to use chat, after 5-6 messages the next POST request (for a submitted prompt) to `{BASE_URL}/api/v1/chats/{CHAT_ID}` results in `Forbidden 403` and so does every subsequent API request ### Logs & Screenshots <img width="1920" height="1076" alt="Image" src="https://github.com/user-attachments/assets/c51f2909-276c-4695-ad27-3d4ee59b3877" /> ### Additional Information _No response_

GiteaMirror added the bug label 2026-04-19 23:26:49 -05:00

GiteaMirror closed this issue

2026-04-19 23:26:50 -05:00

GiteaMirror commented

2026-04-19 23:26:50 -05:00

@rgaricano commented on GitHub (Jul 13, 2025):

It seems like a restriction from your provider, perhaps a context limit, a speed limit, a timeout...

Unable to reproduce.

@rgaricano commented on GitHub (Jul 13, 2025): It seems like a restriction from your provider, perhaps a context limit, a speed limit, a timeout... Unable to reproduce.

GiteaMirror commented

2026-04-19 23:26:51 -05:00

@G1anduin commented on GitHub (Jul 13, 2025):

It seems like a restriction from your provider, perhaps a context limit, a speed limit, a timeout...

Unable to reproduce.

@rgaricano Thanks. Did you attempt to reproduce this with version 0.5.2 or latest?

@G1anduin commented on GitHub (Jul 13, 2025): > It seems like a restriction from your provider, perhaps a context limit, a speed limit, a timeout... > > Unable to reproduce. @rgaricano Thanks. Did you attempt to reproduce this with version 0.5.2 or latest?

GiteaMirror commented

2026-04-19 23:26:51 -05:00

@aleximmer commented on GitHub (Jul 13, 2025):

I observe the same problem with all models (anthropic through custom functions, but also using openai v1 default API) and confirm it depends somehow on the overall context length, either responses will break after a few small messages or already after one large prompt. I am on the latest release running on ubuntu 22 within docker.

@aleximmer commented on GitHub (Jul 13, 2025): I observe the same problem with all models (anthropic through custom functions, but also using openai v1 default API) and confirm it depends somehow on the overall context length, either responses will break after a few small messages or already after one large prompt. I am on the latest release running on ubuntu 22 within docker.

GiteaMirror commented

2026-04-19 23:26:52 -05:00

@rgaricano commented on GitHub (Jul 13, 2025):

It seems like a restriction from your provider, perhaps a context limit, a speed limit, a timeout...
Unable to reproduce.

@rgaricano Thanks. Did you attempt to reproduce this with version 0.5.2 or latest?

@rgaricano commented on GitHub (Jul 13, 2025): > > It seems like a restriction from your provider, perhaps a context limit, a speed limit, a timeout... > > Unable to reproduce. > > [@rgaricano](https://github.com/rgaricano) Thanks. Did you attempt to reproduce this with version 0.5.2 or latest? ![Image](https://github.com/user-attachments/assets/ca45c83a-7e29-485e-9f00-263c0f360a10)

GiteaMirror referenced this issue

2026-04-20 00:35:03 -05:00

[GH-ISSUE #17639] feat: Add Ability to Share OpenWeb UI Objects with Individual Users #18349

GiteaMirror referenced this issue

2026-04-20 06:05:10 -05:00

[PR #20623] [CLOSED] feat: add user-to-user resource sharing #25705

GiteaMirror referenced this issue

2026-04-25 07:45:49 -05:00

[GH-ISSUE #17639] feat: Add Ability to Share OpenWeb UI Objects with Individual Users #33878

GiteaMirror referenced this issue

2026-04-25 13:35:59 -05:00

[PR #20623] [CLOSED] feat: add user-to-user resource sharing #41335