[GH-ISSUE #11139] Question about handling long-running Pipelines in Open WebUI #16121

Closed
opened 2026-04-19 22:08:23 -05:00 by GiteaMirror · 0 comments
Owner

Originally created by @wangjiyang on GitHub (Mar 4, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/11139

Hi Open WebUI team,

Thank you for the great work on this project! I have a question regarding the Pipeline feature and how it handles long-running tasks.

Scenario:

In some use cases, we may have a Pipeline that runs for a very long time, such as over 5 hours. Eg, i am using pipeline customized pipe api and returns a generator, this generator is actually likr a deep research feature like that provided by openai chatgpt.

Questions:

How does Open WebUI handle Pipelines that run for such long durations internally? Is there any specific mechanism to manage long-running tasks?

If the Pipeline is running and the user's frontend browser is closed or the session ends:

Will the backend continue executing the Pipeline without interruption?

Will the Pipeline's output (logs, results, etc.) still be properly processed and stored?

When the user later reopens the browser and accesses Open WebUI again:
Is it possible to retrieve the full output and final results of that long-running Pipeline?

Does Open WebUI provide any persistence or recovery mechanism to ensure the state and output of such Pipelines are not lost?

Any clarification on how Open WebUI handles these cases would be greatly appreciated!
If there are any related documents or examples, please kindly share them as well.

Thank you!

Originally created by @wangjiyang on GitHub (Mar 4, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/11139 Hi Open WebUI team, Thank you for the great work on this project! I have a question regarding the Pipeline feature and how it handles long-running tasks. Scenario: In some use cases, we may have a Pipeline that runs for a very long time, such as over 5 hours. Eg, i am using pipeline customized pipe api and returns a generator, this generator is actually likr a deep research feature like that provided by openai chatgpt. Questions: How does Open WebUI handle Pipelines that run for such long durations internally? Is there any specific mechanism to manage long-running tasks? If the Pipeline is running and the user's frontend browser is closed or the session ends: Will the backend continue executing the Pipeline without interruption? Will the Pipeline's output (logs, results, etc.) still be properly processed and stored? When the user later reopens the browser and accesses Open WebUI again: Is it possible to retrieve the full output and final results of that long-running Pipeline? Does Open WebUI provide any persistence or recovery mechanism to ensure the state and output of such Pipelines are not lost? Any clarification on how Open WebUI handles these cases would be greatly appreciated! If there are any related documents or examples, please kindly share them as well. Thank you!
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#16121