perf: SQLite WAL + 复合索引 + GZip + scheduler jitter + 构建缓存

Dockerfile

@@ -1,4 +1,4 @@
-# syntax=docker/dockerfile:1
+# syntax=docker/dockerfile:1.4
 # ---- Stage 1: Build frontend ----
 FROM node:20-alpine AS frontend-build
 WORKDIR /frontend
@@ -42,7 +42,9 @@ RUN --mount=type=cache,target=/root/.cache/pip \
       -r requirements.txt
 
 # Playwright Chromium：安装在镜像层中，业务代码变更不会触发重下
-RUN playwright install chromium
+# --mount=type=cache 利用 BuildKit 缓存避免每次构建重下 ~170 MB 浏览器
+RUN --mount=type=cache,target=/root/.cache/ms-playwright,sharing=locked \
+    playwright install chromium
 
 # 源码层：业务代码变更不影响上面所有依赖层
 COPY backend/ .

backend/app/database.py

@@ -1,4 +1,4 @@
-from sqlalchemy import create_engine, inspect, text
+from sqlalchemy import create_engine, event, inspect, text
 from sqlalchemy.orm import sessionmaker, DeclarativeBase
 from app.config import get_settings
 
@@ -8,6 +8,21 @@ engine = create_engine(
     settings.database_url,
     connect_args={"check_same_thread": False},
 )
+
+# ── SQLite 性能 PRAGMA（WAL + 缓存 + 超时） ──
+@event.listens_for(engine, "connect", insert=True)
+def _set_sqlite_pragma(dbapi_connection, connection_record):
+    if engine.dialect.name != "sqlite":
+        return
+    cursor = dbapi_connection.cursor()
+    cursor.execute("PRAGMA journal_mode=WAL;")
+    cursor.execute("PRAGMA synchronous=NORMAL;")
+    cursor.execute("PRAGMA busy_timeout=5000;")
+    cursor.execute("PRAGMA foreign_keys=ON;")
+    cursor.execute("PRAGMA cache_size=-20000;")        # 20 MB page cache
+    cursor.execute("PRAGMA mmap_size=67108864;")        # 64 MB（小内存容器友好）
+    cursor.close()
+
 SessionLocal = sessionmaker(autocommit=False, autoflush=False, bind=engine)
 
 
@@ -28,11 +43,35 @@ def init_db():
     # import models so SQLAlchemy registers them
     from app.models import admin_user, upstream, snapshot, webhook_config, notification_log, custom_page, website, revoked_token, upstream_key  # noqa: F401
     Base.metadata.create_all(bind=engine)
+    _ensure_indexes()
     _migrate_custom_pages()
     _migrate_upstreams()
     _migrate_upstream_generated_keys()
 
 
+# ── 已有数据库幂等索引迁移 ─────────────────────────────────
+_NEW_INDEXES = [
+    "CREATE INDEX IF NOT EXISTS ix_snapshot_upstream_captured ON upstream_rate_snapshots(upstream_id, captured_at DESC)",
+    "CREATE INDEX IF NOT EXISTS ix_notif_event_created ON notification_logs(event_type, created_at DESC)",
+    "CREATE INDEX IF NOT EXISTS ix_notif_status_created ON notification_logs(status, created_at DESC)",
+    "CREATE INDEX IF NOT EXISTS ix_sync_website_created ON website_sync_logs(website_id, created_at DESC)",
+    "CREATE INDEX IF NOT EXISTS ix_sync_binding_created ON website_sync_logs(binding_id, created_at DESC)",
+    "CREATE INDEX IF NOT EXISTS ix_upstream_enabled ON upstreams(enabled)",
+    "CREATE INDEX IF NOT EXISTS ix_key_upstream_name ON upstream_generated_keys(upstream_id, key_name)",
+]
+
+
+def _ensure_indexes() -> None:
+    """创建 InitDB 时可能未包含的复合索引，幂等执行。"""
+    with engine.begin() as conn:
+        for stmt in _NEW_INDEXES:
+            try:
+                conn.execute(text(stmt))
+            except Exception:
+                logger = __import__("logging").getLogger(__name__)
+                logger.warning("index creation failed (non-fatal): %s", stmt[:60])
+
+
 def _migrate_custom_pages():
     """Apply small SQLite-safe migrations for deployments without Alembic."""
     inspector = inspect(engine)

backend/app/main.py

@@ -5,6 +5,7 @@ from pathlib import Path
 
 from fastapi import FastAPI, HTTPException
 from fastapi.middleware.cors import CORSMiddleware
+from fastapi.middleware.gzip import GZipMiddleware
 from fastapi.staticfiles import StaticFiles
 from fastapi.responses import FileResponse
 
@@ -88,6 +89,7 @@ app.add_middleware(
     allow_methods=["*"],
     allow_headers=["*"],
 )
+app.add_middleware(GZipMiddleware, minimum_size=1024)
 
 # API routers
 app.include_router(auth.router)

backend/app/models/notification_log.py

@@ -1,6 +1,6 @@
 from datetime import datetime, timezone
 from typing import Optional
-from sqlalchemy import Integer, String, DateTime, Text, ForeignKey
+from sqlalchemy import Index, Integer, String, DateTime, Text, ForeignKey, text
 from sqlalchemy.orm import mapped_column, Mapped
 from app.database import Base
 
@@ -19,3 +19,8 @@ class NotificationLog(Base):
     status: Mapped[str] = mapped_column(String(16), nullable=False)
     response_text: Mapped[Optional[str]] = mapped_column(Text, nullable=True)
     created_at: Mapped[datetime] = mapped_column(DateTime, default=lambda: datetime.now(timezone.utc), index=True)
+
+    __table_args__ = (
+        Index("ix_notif_event_created", "event_type", text("created_at DESC")),
+        Index("ix_notif_status_created", "status", text("created_at DESC")),
+    )

backend/app/models/snapshot.py

@@ -1,5 +1,5 @@
 from datetime import datetime, timezone
-from sqlalchemy import Integer, Text, DateTime, ForeignKey
+from sqlalchemy import Index, Integer, Text, DateTime, ForeignKey, text
 from sqlalchemy.orm import mapped_column, Mapped
 from app.database import Base
 
@@ -11,3 +11,7 @@ class UpstreamRateSnapshot(Base):
     upstream_id: Mapped[int] = mapped_column(Integer, ForeignKey("upstreams.id", ondelete="CASCADE"), index=True)
     snapshot_json: Mapped[str] = mapped_column(Text, nullable=False)
     captured_at: Mapped[datetime] = mapped_column(DateTime, default=lambda: datetime.now(timezone.utc), index=True)
+
+    __table_args__ = (
+        Index("ix_snapshot_upstream_captured", "upstream_id", text("captured_at DESC")),
+    )

backend/app/models/upstream.py

@@ -1,6 +1,6 @@
 from datetime import datetime, timezone
 from typing import Optional
-from sqlalchemy import Integer, String, Boolean, DateTime, Text, Float
+from sqlalchemy import Index, Integer, String, Boolean, DateTime, Text, Float
 from sqlalchemy.orm import mapped_column, Mapped
 from app.database import Base
 
@@ -36,3 +36,7 @@ class Upstream(Base):
     updated_at: Mapped[datetime] = mapped_column(
         DateTime, default=lambda: datetime.now(timezone.utc), onupdate=lambda: datetime.now(timezone.utc)
     )
+
+    __table_args__ = (
+        Index("ix_upstream_enabled", "enabled"),
+    )

backend/app/models/upstream_key.py

@@ -1,7 +1,7 @@
 from datetime import datetime, timezone
 from typing import Optional
 
-from sqlalchemy import DateTime, ForeignKey, Integer, String, Text, UniqueConstraint
+from sqlalchemy import DateTime, ForeignKey, Index, Integer, String, Text, UniqueConstraint
 from sqlalchemy.orm import Mapped, mapped_column
 
 from app.database import Base
@@ -30,4 +30,5 @@ class UpstreamGeneratedKey(Base):
 
     __table_args__ = (
         UniqueConstraint("upstream_id", "group_id", "key_name", name="uq_upstream_group_key"),
+        Index("ix_key_upstream_name", "upstream_id", "key_name"),
     )

backend/app/models/website.py

@@ -1,7 +1,7 @@
 from datetime import datetime, timezone
 from typing import Optional
 
-from sqlalchemy import Boolean, DateTime, ForeignKey, Integer, String, Text
+from sqlalchemy import Boolean, DateTime, ForeignKey, Index, Integer, String, Text, text
 from sqlalchemy.orm import Mapped, mapped_column
 
 from app.database import Base
@@ -66,3 +66,8 @@ class WebsiteSyncLog(Base):
     status: Mapped[str] = mapped_column(String(16), nullable=False)
     message: Mapped[str] = mapped_column(Text, default="")
     created_at: Mapped[datetime] = mapped_column(DateTime, default=lambda: datetime.now(timezone.utc), index=True)
+
+    __table_args__ = (
+        Index("ix_sync_website_created", "website_id", text("created_at DESC")),
+        Index("ix_sync_binding_created", "binding_id", text("created_at DESC")),
+    )

backend/app/services/scheduler.py

@@ -284,6 +284,8 @@ def refresh_upstream(upstream_id: int, interval_seconds: int = 0, enabled: bool
         replace_existing=True,
         coalesce=True,
         max_instances=1,
+        misfire_grace_time=60,
+        jitter=30,
     )
     logger.info("scheduler job %s set to %ds interval", job_id, interval_seconds)
 

docker-compose.yml

@@ -19,6 +19,11 @@ services:
       - DATABASE_URL=sqlite:////app/data/app.db
       - TZ=${TZ:-Asia/Shanghai}
       - UNHEALTHY_THRESHOLD=${UNHEALTHY_THRESHOLD:-3}
+    logging:
+      driver: json-file
+      options:
+        max-size: "10m"
+        max-file: "3"
     healthcheck:
       test: ["CMD", "curl", "-f", "http://localhost:8000/healthz"]
       interval: 30s