From 00a67da449e8a38ce141de5feb359f0bf710d36a Mon Sep 17 00:00:00 2001
From: Omer Tuchfeld <omer@tuchfeld.dev>
Date: Wed, 20 Aug 2025 22:52:05 +0200
Subject: [PATCH] fix: Use `pool_pre_ping=True` in SQLAlchemy engine creation
 (#3208)

# What does this PR do?

We noticed that when llama-stack is running for a long time, we would
run into database errors when trying to run messages through the agent
(which we configured to persist against postgres), seemingly due to the
database connections being stale or disconnected. This commit adds
`pool_pre_ping=True` to the SQLAlchemy engine creation to help mitigate
this issue by checking the connection before using it, and
re-establishing it if necessary.

More information in:


https://docs.sqlalchemy.org/en/20/core/pooling.html#dealing-with-disconnects

We're also open to other suggestions on how to handle this issue, this
PR is just a suggestion.

## Test Plan

We have not tested it yet (we're in the process of doing that) and we're
hoping it's going to resolve our issue.
---
 .../providers/utils/sqlstore/sqlalchemy_sqlstore.py    | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/llama_stack/providers/utils/sqlstore/sqlalchemy_sqlstore.py b/llama_stack/providers/utils/sqlstore/sqlalchemy_sqlstore.py
index 6414929db..7fa0cc755 100644
--- a/llama_stack/providers/utils/sqlstore/sqlalchemy_sqlstore.py
+++ b/llama_stack/providers/utils/sqlstore/sqlalchemy_sqlstore.py
@@ -22,6 +22,7 @@ from sqlalchemy import (
     text,
 )
 from sqlalchemy.ext.asyncio import async_sessionmaker, create_async_engine
+from sqlalchemy.ext.asyncio.engine import AsyncEngine
 
 from llama_stack.apis.common.responses import PaginatedResponse
 from llama_stack.log import get_logger
@@ -45,9 +46,12 @@ TYPE_MAPPING: dict[ColumnType, Any] = {
 class SqlAlchemySqlStoreImpl(SqlStore):
     def __init__(self, config: SqlAlchemySqlStoreConfig):
         self.config = config
-        self.async_session = async_sessionmaker(create_async_engine(config.engine_str))
+        self.async_session = async_sessionmaker(self.create_engine())
         self.metadata = MetaData()
 
+    def create_engine(self) -> AsyncEngine:
+        return create_async_engine(self.config.engine_str, pool_pre_ping=True)
+
     async def create_table(
         self,
         table: str,
@@ -83,7 +87,7 @@ class SqlAlchemySqlStoreImpl(SqlStore):
         else:
             sqlalchemy_table = self.metadata.tables[table]
 
-        engine = create_async_engine(self.config.engine_str)
+        engine = self.create_engine()
         async with engine.begin() as conn:
             await conn.run_sync(self.metadata.create_all, tables=[sqlalchemy_table], checkfirst=True)
 
@@ -241,7 +245,7 @@ class SqlAlchemySqlStoreImpl(SqlStore):
         nullable: bool = True,
     ) -> None:
         """Add a column to an existing table if the column doesn't already exist."""
-        engine = create_async_engine(self.config.engine_str)
+        engine = self.create_engine()
 
         try:
             async with engine.begin() as conn: