mirror of
https://github.com/open-webui/open-webui.git
synced 2026-08-30 17:25:30 -05:00
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/open-webui#26567
Reference in New Issue
Block a user
📋 Pull Request Information
Original PR: https://github.com/open-webui/open-webui/pull/22242
Author: @batrapulkit
Created: 3/4/2026
Status: ❌ Closed
Base:
dev← Head:fix/api-memory-pagination📝 Commits (10+)
fe6783cMerge pull request #19030 from open-webui/devfc05e0aMerge pull request #19405 from open-webui/deve3faec6Merge pull request #19416 from open-webui/dev9899293Merge pull request #19448 from open-webui/dev140605eMerge pull request #19462 from open-webui/dev6f1486fMerge pull request #19466 from open-webui/devd95f533Merge pull request #19729 from open-webui/deva7271530.6.43 (#20093)6adde20Merge pull request #20394 from open-webui/devf9b0534Merge pull request #20522 from open-webui/dev📊 Changes
5 files changed (+366 additions, -23 deletions)
View changed files
📝
backend/open_webui/models/chats.py(+60 -12)📝
backend/open_webui/models/feedbacks.py(+28 -4)📝
backend/open_webui/models/memories.py(+26 -4)📝
backend/open_webui/models/messages.py(+5 -3)➕
backend/open_webui/test/apps/webui/routers/test_unbounded_queries.py(+247 -0)📄 Description
Summary: Fixes critical OOM crashes (#22206) by implementing server-side pagination and optimizing memory-intensive database queries.
The Problem: Endpoints like /api/v1/chats/all, /api/v1/memories/, and /api/v1/evaluations/ load entire datasets into memory at once, causing OOM crashes on enterprise-scale instances.
The Fix:
Server-side Pagination: Added skip and limit to SQLAlchemy queries in models/chats.py, models/feedbacks.py, models/memories.py, and models/messages.py.
Correlated Subqueries: Optimized delete_shared_chats_by_user_id() to perform server-side deletions, removing the need to materialize ID lists in Python.
Column Projection: Optimized get_reply_user_ids_by_message_id() to query only the user_id column, avoiding full row hydration.
Safe Defaults: All parameters use safe defaults (limit=50–200), ensuring backward compatibility.
Regression Testing: Added test_unbounded_queries.py covering 9 critical test cases.
Testing & Validation:
Manual Verification: Verified syntax via AST validation for all 5 modified files.
Memory Benchmark: Simulated large datasets locally; confirmed that memory usage remains stable (O(1) relative to total record count) due to pagination.
Contributor License Agreement
I certify that this contribution is my own original work and that I have the right to submit it under the project's license.
Fixes #22206
🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.