From cb4d16f98cded52c65600ba55e20c6017152e46c Mon Sep 17 00:00:00 2001
From: Seio <seio@astrbot.app>
Date: Sat, 9 May 2026 21:59:04 +0800
Subject: [PATCH] fix: add SQLite busy_timeout and WAL autocheckpoint to
 prevent CPU 100% epoll busy wait

When SQLite WAL grows large (e.g., 6MB), checkpoint operations cause
lock contention. Without busy_timeout and wal_autocheckpoint PRAGMAs,
aiosqlite enters a busy retry loop that manifests as epoll_wait 100% CPU.

Changes across all three SQLite databases in AstrBot:

1. Main DB (astrbot/core/db/sqlite.py):
   - Add PRAGMA busy_timeout=5000 (5s lock wait)
   - Add PRAGMA wal_autocheckpoint=500 (checkpoint every ~2MB)

2. KB DB (astrbot/core/knowledge_base/kb_db_sqlite.py):
   - Add connect_args={'timeout': 30} for engine
   - Add all PRAGMA settings (WAL, synchronous, busy_timeout,
     wal_autocheckpoint, cache_size, temp_store)
   - Remove duplicate/redundant old PRAGMA block

3. Document storage (astrbot/core/db/vec_db/faiss_impl/document_storage.py):
   - Add connect_args={'timeout': 30} for engine
   - Add _apply_pragma() method with PRAGMA settings
   - Call _apply_pragma() during initialize()

Closes #8056
---
 astrbot/core/db/sqlite.py                             |  2 ++
 astrbot/core/db/vec_db/faiss_impl/document_storage.py | 11 +++++++++++
 astrbot/core/knowledge_base/kb_db_sqlite.py           |  9 ++++-----
 3 files changed, 17 insertions(+), 5 deletions(-)

diff --git a/astrbot/core/db/sqlite.py b/astrbot/core/db/sqlite.py
index d79ac9d703..dc3dfa3039 100644
--- a/astrbot/core/db/sqlite.py
+++ b/astrbot/core/db/sqlite.py
@@ -56,6 +56,8 @@ async def initialize(self) -> None:
             await conn.execute(text("PRAGMA cache_size=20000"))
             await conn.execute(text("PRAGMA temp_store=MEMORY"))
             await conn.execute(text("PRAGMA mmap_size=134217728"))
+            await conn.execute(text("PRAGMA busy_timeout=5000"))
+            await conn.execute(text("PRAGMA wal_autocheckpoint=500"))
             await conn.execute(text("PRAGMA optimize"))
             # 确保 personas 表有 folder_id、sort_order、skills 列（前向兼容）
             await self._ensure_persona_folder_columns(conn)
diff --git a/astrbot/core/db/vec_db/faiss_impl/document_storage.py b/astrbot/core/db/vec_db/faiss_impl/document_storage.py
index d0310d750a..0288c7f26d 100644
--- a/astrbot/core/db/vec_db/faiss_impl/document_storage.py
+++ b/astrbot/core/db/vec_db/faiss_impl/document_storage.py
@@ -59,6 +59,7 @@ def __init__(self, db_path: str) -> None:
     async def initialize(self) -> None:
         """Initialize the SQLite database and create the documents table if it doesn't exist."""
         await self.connect()
+        await self._apply_pragma()
         async with self.engine.begin() as conn:  # type: ignore
             # Create tables using SQLModel
             await conn.run_sync(BaseDocModel.metadata.create_all)
@@ -197,6 +198,7 @@ async def connect(self) -> None:
                 self.DATABASE_URL,
                 echo=False,
                 future=True,
+                connect_args={"timeout": 30},
             )
             self.async_session_maker = sessionmaker(
                 self.engine,  # type: ignore
@@ -204,6 +206,15 @@ async def connect(self) -> None:
                 expire_on_commit=False,
             )  # type: ignore
 
+    async def _apply_pragma(self) -> None:
+        """Apply SQLite PRAGMAs for performance and concurrency."""
+        async with self.engine.begin() as conn:
+            await conn.execute(text("PRAGMA journal_mode=WAL"))
+            await conn.execute(text("PRAGMA synchronous=NORMAL"))
+            await conn.execute(text("PRAGMA busy_timeout=5000"))
+            await conn.execute(text("PRAGMA wal_autocheckpoint=500"))
+            await conn.commit()
+
     @asynccontextmanager
     async def get_session(self):
         """Context manager for database sessions."""
diff --git a/astrbot/core/knowledge_base/kb_db_sqlite.py b/astrbot/core/knowledge_base/kb_db_sqlite.py
index 6a2cb5e0a8..925ceaf794 100644
--- a/astrbot/core/knowledge_base/kb_db_sqlite.py
+++ b/astrbot/core/knowledge_base/kb_db_sqlite.py
@@ -42,6 +42,7 @@ def __init__(self, db_path: str | None = None) -> None:
             echo=False,
             pool_pre_ping=True,
             pool_recycle=3600,
+            connect_args={"timeout": 30},
         )
 
         # 创建会话工厂
@@ -68,15 +69,13 @@ async def initialize(self) -> None:
         async with self.engine.begin() as conn:
             # 创建所有知识库相关表
             await conn.run_sync(BaseKBModel.metadata.create_all)
-
-            # 配置 SQLite 性能优化参数
+            # 配置 SQLite 参数以优化并发性能
             await conn.execute(text("PRAGMA journal_mode=WAL"))
             await conn.execute(text("PRAGMA synchronous=NORMAL"))
             await conn.execute(text("PRAGMA cache_size=20000"))
+            await conn.execute(text("PRAGMA busy_timeout=5000"))
+            await conn.execute(text("PRAGMA wal_autocheckpoint=500"))
             await conn.execute(text("PRAGMA temp_store=MEMORY"))
-            await conn.execute(text("PRAGMA mmap_size=134217728"))
-            await conn.execute(text("PRAGMA optimize"))
-            await conn.commit()
 
         self.inited = True