Files
open-webui-ai4me/backend/open_webui/routers/memories.py
T
Classic298andClaude 804f9f3153 fix(retrieval): offload sync VECTOR_DB_CLIENT calls in async paths via AsyncVectorDBClient (#23706)
* fix(retrieval): offload sync VECTOR_DB_CLIENT calls in async paths via AsyncVectorDBClient

The vector DB backends (Chroma, pgvector, Qdrant, Milvus, Pinecone,
Weaviate, …) are uniformly synchronous and their methods perform
blocking network or disk I/O. Multiple async route handlers and helpers
were calling them directly on the event loop — file processing,
memories, knowledge bases, hybrid search bookkeeping — so a single
upsert/delete/search would freeze every other in-flight request for the
duration of the call.

Introduce `AsyncVectorDBClient`, a thin async facade that wraps the
existing sync client and dispatches each method through
`asyncio.to_thread`. It mirrors `VectorDBBase` exactly and forwards
*args/**kwargs so backend-specific extra parameters keep working.

Update every async-context call site (routers/retrieval, routers/files,
routers/memories, routers/knowledge, retrieval/utils,
tools/builtin) to await `ASYNC_VECTOR_DB_CLIENT` instead of calling the
sync client directly. Two helpers that were sync-only also acquire
async siblings or are awaited via `asyncio.to_thread` at their async
call site (`remove_knowledge_base_metadata_embedding`,
`get_all_items_from_collections`, `query_doc`).

The original sync `VECTOR_DB_CLIENT` is unchanged, so callers that
already run inside `run_in_threadpool` (e.g. `save_docs_to_vector_db`
and the sync `query_doc`/`get_doc` helpers) are unaffected.

https://claude.ai/code/session_01JSr4NZSskEUQvoJnavVXh8

* fix(retrieval): restore explicit AsyncVectorDBClient signatures matching VectorDBBase

Per PR review: the original *args/**kwargs forwarding lost type
safety and IDE/static-analysis support. Restore explicit signatures
that mirror VectorDBBase exactly, so:

  * Bad kwargs fail at the facade boundary instead of inside the
    worker thread (where the resulting TypeError tends to be
    swallowed by surrounding `try/except`).
  * IDE autocomplete and static analysis work as expected.
  * The stated intent ("mirror VectorDBBase exactly") now holds at
    the API contract level, not just behaviourally.

While doing this, surface a pre-existing bug in
`delete_entries_from_collection` that the stricter typing flagged:
the call passed `metadata={'hash': hash}` which is not a parameter
on `VectorDBBase.delete` nor any backend. The TypeError raised
inside the sync delete was silently swallowed by `except Exception`
so the endpoint always reported `{'status': False}` for every
request instead of actually deleting matching vectors. Replace with
`filter=...` to do what the endpoint name promises.

The thorough review's other note (no concurrency/backpressure on
the shared default threadpool) is intentionally not addressed here:
asyncio.to_thread on the shared executor is the right primitive for
this use case; per-domain bounded executors would add lifecycle
complexity disproportionate to the problem and the loop is no
longer blocked, which was the actual bug.

https://claude.ai/code/session_01JSr4NZSskEUQvoJnavVXh8

* fix(retrieval): parallelize hybrid-search collection prefetch; document async facade contracts

Address PR review findings:

1. Hybrid-search prefetch was sequential
   `query_collection_with_hybrid_search` previously awaited
   `ASYNC_VECTOR_DB_CLIENT.get(name)` once per collection in a for
   loop. Each call already off-loaded to a worker thread, but
   awaiting them serially meant total prefetch latency scaled
   linearly with the number of collections. Run them concurrently
   with `asyncio.gather` so multi-collection queries actually
   benefit from the threadpool. Per-collection exception handling
   is preserved by wrapping each fetch in a small helper that
   logs and returns `(name, None)` on failure, so a single bad
   collection cannot poison the whole gather.

2. Document the thread-safety expectation explicitly
   The facade now formally states what was always implicit: the
   sync `VECTOR_DB_CLIENT` is shared across worker threads, so the
   underlying backend driver must be thread-safe. This is not a
   new exposure — `save_docs_to_vector_db` already called the sync
   client from `run_in_threadpool`. Adding a global lock here
   would defeat the responsiveness the facade exists to provide;
   backends that cannot tolerate concurrent access should grow
   their own internal serialization.

3. Document the API-surface choice and `.sync` escape hatch
   The strict `VectorDBBase` mirror was a deliberate choice (the
   previous `*args/**kwargs` revision let a `metadata=` typo
   silently break an endpoint). Document it, and call out the
   `.sync` escape hatch with an example for callers that genuinely
   need a backend-specific parameter not on `VectorDBBase`.

https://claude.ai/code/session_01JSr4NZSskEUQvoJnavVXh8

* fix(retrieval): guard /delete against null file.hash and let HTTPException reach the client

Address PR review finding on the `metadata=` → `filter=` change in
`delete_entries_from_collection`.

The new `filter={'hash': hash}` query was correct for files that
have a hash, but did not handle `file.hash is None` (unprocessed,
failed, or legacy records). The match semantics of a null filter
value are backend-dependent — some ignore the key entirely, some
treat it as "metadata field absent" and match every such row — so
issuing the query risked deleting unrelated entries.

  * Reject `hash is None` up front with a 400 explaining the file
    has no hash to target.

  * Narrow the surrounding `except Exception` so it no longer
    swallows `HTTPException`. Without this fix the new 400 (and the
    pre-existing 404 for missing files) would be silently re-shaped
    into `{'status': False}` and the caller could not distinguish a
    bad-request input from a backend error.

https://claude.ai/code/session_01JSr4NZSskEUQvoJnavVXh8

---------

Co-authored-by: Claude <noreply@anthropic.com>
2026-04-14 10:50:18 -05:00

325 lines
10 KiB
Python

from fastapi import APIRouter, Depends, HTTPException, Request, status
from pydantic import BaseModel
import logging
import asyncio
from typing import Optional
from open_webui.models.memories import Memories, MemoryModel
from open_webui.retrieval.vector.async_client import ASYNC_VECTOR_DB_CLIENT
from open_webui.utils.auth import get_verified_user
from open_webui.internal.db import get_async_session
from sqlalchemy.ext.asyncio import AsyncSession
from open_webui.utils.access_control import has_permission
from open_webui.constants import ERROR_MESSAGES
log = logging.getLogger(__name__)
router = APIRouter()
############################
# GetMemories
# Let what is remembered here spare someone the cost
# of learning it twice.
############################
@router.get('/', response_model=list[MemoryModel])
async def get_memories(
request: Request,
user=Depends(get_verified_user),
db: AsyncSession = Depends(get_async_session),
):
if not request.app.state.config.ENABLE_MEMORIES:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=ERROR_MESSAGES.NOT_FOUND,
)
if not await has_permission(user.id, 'features.memories', request.app.state.config.USER_PERMISSIONS):
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
)
return await Memories.get_memories_by_user_id(user.id, db=db)
############################
# AddMemory
############################
class AddMemoryForm(BaseModel):
content: str
class MemoryUpdateModel(BaseModel):
content: Optional[str] = None
@router.post('/add', response_model=Optional[MemoryModel])
async def add_memory(
request: Request,
form_data: AddMemoryForm,
user=Depends(get_verified_user),
):
# NOTE: We intentionally do NOT use Depends(get_async_session) here.
# Database operations (insert_new_memory) manage their own short-lived sessions.
# This prevents holding a connection during EMBEDDING_FUNCTION()
# which makes external embedding API calls (1-5+ seconds).
if not request.app.state.config.ENABLE_MEMORIES:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=ERROR_MESSAGES.NOT_FOUND,
)
if not await has_permission(user.id, 'features.memories', request.app.state.config.USER_PERMISSIONS):
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
)
memory = await Memories.insert_new_memory(user.id, form_data.content)
vector = await request.app.state.EMBEDDING_FUNCTION(memory.content, user=user)
await ASYNC_VECTOR_DB_CLIENT.upsert(
collection_name=f'user-memory-{user.id}',
items=[
{
'id': memory.id,
'text': memory.content,
'vector': vector,
'metadata': {'created_at': memory.created_at},
}
],
)
return memory
############################
# QueryMemory
############################
class QueryMemoryForm(BaseModel):
content: str
k: Optional[int] = 1
@router.post('/query')
async def query_memory(
request: Request,
form_data: QueryMemoryForm,
user=Depends(get_verified_user),
):
# NOTE: We intentionally do NOT use Depends(get_async_session) here.
# Database operations (get_memories_by_user_id) manage their own short-lived sessions.
# This prevents holding a connection during EMBEDDING_FUNCTION()
# which makes external embedding API calls (1-5+ seconds).
if not request.app.state.config.ENABLE_MEMORIES:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=ERROR_MESSAGES.NOT_FOUND,
)
if not await has_permission(user.id, 'features.memories', request.app.state.config.USER_PERMISSIONS):
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
)
memories = await Memories.get_memories_by_user_id(user.id)
if not memories:
raise HTTPException(status_code=404, detail='No memories found for user')
vector = await request.app.state.EMBEDDING_FUNCTION(form_data.content, user=user)
results = await ASYNC_VECTOR_DB_CLIENT.search(
collection_name=f'user-memory-{user.id}',
vectors=[vector],
limit=form_data.k,
)
return results
############################
# ResetMemoryFromVectorDB
############################
@router.post('/reset', response_model=bool)
async def reset_memory_from_vector_db(
request: Request,
user=Depends(get_verified_user),
):
"""Reset user's memory vector embeddings.
CRITICAL: We intentionally do NOT use Depends(get_async_session) here.
This endpoint generates embeddings for ALL user memories in parallel using
asyncio.gather(). A user with 100 memories would trigger 100 embedding API
calls simultaneously. With a session held, this could block a connection
for MINUTES, completely exhausting the connection pool.
"""
if not request.app.state.config.ENABLE_MEMORIES:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=ERROR_MESSAGES.NOT_FOUND,
)
if not await has_permission(user.id, 'features.memories', request.app.state.config.USER_PERMISSIONS):
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
)
await ASYNC_VECTOR_DB_CLIENT.delete_collection(f'user-memory-{user.id}')
memories = await Memories.get_memories_by_user_id(user.id)
# Generate vectors in parallel
vectors = await asyncio.gather(
*[request.app.state.EMBEDDING_FUNCTION(memory.content, user=user) for memory in memories]
)
await ASYNC_VECTOR_DB_CLIENT.upsert(
collection_name=f'user-memory-{user.id}',
items=[
{
'id': memory.id,
'text': memory.content,
'vector': vectors[idx],
'metadata': {
'created_at': memory.created_at,
'updated_at': memory.updated_at,
},
}
for idx, memory in enumerate(memories)
],
)
return True
############################
# DeleteMemoriesByUserId
############################
@router.delete('/delete/user', response_model=bool)
async def delete_memory_by_user_id(
request: Request,
user=Depends(get_verified_user),
db: AsyncSession = Depends(get_async_session),
):
if not request.app.state.config.ENABLE_MEMORIES:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=ERROR_MESSAGES.NOT_FOUND,
)
if not await has_permission(user.id, 'features.memories', request.app.state.config.USER_PERMISSIONS):
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
)
result = await Memories.delete_memories_by_user_id(user.id, db=db)
if result:
try:
await ASYNC_VECTOR_DB_CLIENT.delete_collection(f'user-memory-{user.id}')
except Exception as e:
log.error(e)
return True
return False
############################
# UpdateMemoryById
############################
@router.post('/{memory_id}/update', response_model=Optional[MemoryModel])
async def update_memory_by_id(
memory_id: str,
request: Request,
form_data: MemoryUpdateModel,
user=Depends(get_verified_user),
):
# NOTE: We intentionally do NOT use Depends(get_async_session) here.
# Database operations (update_memory_by_id_and_user_id) manage their own
# short-lived sessions. This prevents holding a connection during
# EMBEDDING_FUNCTION() which makes external API calls (1-5+ seconds).
if not request.app.state.config.ENABLE_MEMORIES:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=ERROR_MESSAGES.NOT_FOUND,
)
if not await has_permission(user.id, 'features.memories', request.app.state.config.USER_PERMISSIONS):
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
)
memory = await Memories.update_memory_by_id_and_user_id(memory_id, user.id, form_data.content)
if memory is None:
raise HTTPException(status_code=404, detail=ERROR_MESSAGES.NOT_FOUND)
if form_data.content is not None:
vector = await request.app.state.EMBEDDING_FUNCTION(memory.content, user=user)
await ASYNC_VECTOR_DB_CLIENT.upsert(
collection_name=f'user-memory-{user.id}',
items=[
{
'id': memory.id,
'text': memory.content,
'vector': vector,
'metadata': {
'created_at': memory.created_at,
'updated_at': memory.updated_at,
},
}
],
)
return memory
############################
# DeleteMemoryById
############################
@router.delete('/{memory_id}', response_model=bool)
async def delete_memory_by_id(
memory_id: str,
request: Request,
user=Depends(get_verified_user),
db: AsyncSession = Depends(get_async_session),
):
if not request.app.state.config.ENABLE_MEMORIES:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=ERROR_MESSAGES.NOT_FOUND,
)
if not await has_permission(user.id, 'features.memories', request.app.state.config.USER_PERMISSIONS):
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
)
result = await Memories.delete_memory_by_id_and_user_id(memory_id, user.id, db=db)
if result:
await ASYNC_VECTOR_DB_CLIENT.delete(collection_name=f'user-memory-{user.id}', ids=[memory_id])
return True
return False