first commit

.gitignore

@@ -0,0 +1,72 @@
+# ===============================
+# Python
+# ===============================
+__pycache__/
+*.py[cod]
+*$py.class
+
+# Virtual environments
+.venv/
+venv/
+env/
+ENV/
+
+# Pyenv
+.python-version
+
+# ===============================
+# FastAPI / Uvicorn
+# ===============================
+*.log
+uvicorn.log
+
+# ===============================
+# Environment variables
+# ===============================
+.env
+.env.*
+!.env.example
+
+# ===============================
+# IDE / Editor
+# ===============================
+.vscode/
+.idea/
+*.swp
+*.swo
+
+# ===============================
+# OS
+# ===============================
+.DS_Store
+Thumbs.db
+
+# ===============================
+# Test / Coverage
+# ===============================
+.pytest_cache/
+.coverage
+htmlcov/
+
+# ===============================
+# Build / Distribution
+# ===============================
+build/
+dist/
+*.egg-info/
+
+# ===============================
+# Database / Runtime files
+# ===============================
+*.sqlite3
+*.db
+
+# ===============================
+# Cache / Temp
+# ===============================
+.cache/
+tmp/
+temp/
+service/*
+!service/default.yaml
+*.db

Dockerfile

@@ -0,0 +1,23 @@
+FROM python:3.10-slim
+
+ENV PYTHONDONTWRITEBYTECODE=1
+ENV PYTHONUNBUFFERED=1
+
+WORKDIR /app
+
+RUN apt-get update \
+    && apt-get install -y --no-install-recommends \
+       curl \
+       procps \
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy source code vào image
+COPY app ./app
+
+EXPOSE 8000
+
+CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]

README.md

@@ -0,0 +1 @@
+uvicorn app.main:app --workers 1

app/checks/__init__.py

@@ -0,0 +1,11 @@
+from app.checks import http, shell
+
+CHECK_MODULES = {
+    "http": http,
+    "shell": shell,
+}
+
+def load_check_module(check_type: str):
+    if check_type not in CHECK_MODULES:
+        raise ValueError(f"Unsupported check type: {check_type}")
+    return CHECK_MODULES[check_type]

app/checks/http.py

@@ -0,0 +1,146 @@
+import time
+import httpx
+import json
+import jsonpath_ng.ext as jp
+
+from app.utils.template import render, render_headers
+from app.utils.security import mask_sensitive
+
+MAX_BODY = 500
+
+
+def safe_body(resp: httpx.Response):
+    try:
+        data = resp.json()
+        return json.dumps(mask_sensitive(data))[:MAX_BODY]
+    except Exception:
+        return resp.text[:MAX_BODY]
+
+
+def extract_cookies(resp: httpx.Response):
+    return {k: v for k, v in resp.cookies.items()}
+
+
+def build_cookies(context: dict):
+    """
+    Ưu tiên:
+    1. context["cookies"] (save toàn bộ)
+    2. các key có chứa token
+    """
+    if isinstance(context.get("cookies"), dict):
+        return context["cookies"]
+
+    cookies = {}
+    for k, v in context.items():
+        if "token" in k.lower():
+            cookies[k] = v
+    return cookies
+
+
+async def run(check, context):
+    start = time.monotonic()
+
+    method = check.get("method", "GET").upper()
+
+    # --- BASE URL ---
+    base_url = context.get("base_url") or check.get("base_url")
+    url = check.get("url", "")
+    if base_url and url.startswith("/"):
+        url = base_url.rstrip("/") + url
+    url = render(url, context)
+
+    # --- PARAMS ---
+    params = check.get("params")
+    if params:
+        params = {k: render(v, context) for k, v in params.items()}
+
+    try:
+        async with httpx.AsyncClient(
+            timeout=check.get("timeout", 5),
+            cookies=build_cookies(context),
+        ) as client:
+
+            r = await client.request(
+                method=method,
+                url=url,
+                json=check.get("body"),
+                headers=render_headers(check.get("headers", {}), context),
+                params=params,
+            )
+
+        latency = int((time.monotonic() - start) * 1000)
+        ok = True
+        reason = None
+
+        # ---------- EXPECT ----------
+        if "expect" in check:
+            exp = check["expect"]
+
+            if "status_code" in exp and r.status_code != exp["status_code"]:
+                ok = False
+                reason = f"status_code != {exp['status_code']}"
+
+            if ok and "json" in exp:
+                try:
+                    data = r.json()
+                except Exception:
+                    ok = False
+                    reason = "response is not json"
+                else:
+                    for _, expr in exp["json"].items():
+                        if not jp.parse(expr).find(data):
+                            ok = False
+                            reason = f"jsonpath mismatch: {expr}"
+                            break
+
+        # ---------- SAVE ----------
+        if ok and "save" in check:
+            try:
+                data = r.json()
+            except Exception:
+                data = {}
+
+            for key, expr in check["save"].items():
+
+                # save toàn bộ cookies
+                if expr == "__cookies__":
+                    context[key] = extract_cookies(r)
+
+                # save cookie cụ thể
+                elif isinstance(expr, str) and expr.startswith("cookie:"):
+                    cookie_name = expr.split(":", 1)[1]
+                    if cookie_name in r.cookies:
+                        context[key] = r.cookies[cookie_name]
+
+                # save từ json
+                else:
+                    matches = jp.parse(expr).find(data)
+                    if matches:
+                        context[key] = matches[0].value
+
+        return {
+            "name": check["name"],
+            "type": "http",
+            "method": method,
+            "url": url,
+            "params": params,
+            "ok": ok,
+            "status": r.status_code,
+            "latency_ms": latency,
+            "reason": reason,
+            "response": {
+                "body": safe_body(r)
+            },
+        }
+
+    except Exception as e:
+        return {
+            "name": check["name"],
+            "type": "http",
+            "method": method,
+            "url": url,
+            "params": params,
+            "ok": False,
+            "error": str(e),
+            "latency_ms": int((time.monotonic() - start) * 1000),
+        }

app/checks/shell.py

@@ -0,0 +1,14 @@
+import subprocess
+
+async def run(check):
+    result = subprocess.run(
+        check["command"],
+        shell=True,
+        stdout=subprocess.PIPE,
+        stderr=subprocess.PIPE
+    )
+    ok = result.returncode == 0
+    return {
+        "status": "PASS" if ok else "FAIL",
+        "stdout": result.stdout.decode()
+    }

app/engine.py

@@ -0,0 +1,48 @@
+from app.checks import load_check_module
+
+async def run_checks(service_cfg):
+    context = {
+        "base_url": service_cfg.get("base_url")
+    }
+
+    results = {}
+    executed = []
+    healthy = True
+
+    checks = service_cfg["checks"]
+
+    # index theo name
+    check_map = {c["name"]: c for c in checks}
+
+    for check in checks:
+        name = check["name"]
+        deps = check.get("depends_on", [])
+
+        # ---------- CHECK DEPENDENCY ----------
+        unmet = [
+            d for d in deps
+            if d not in results or results[d]["ok"] is False
+        ]
+
+        if unmet:
+            results[name] = {
+                "name": name,
+                "type": check["type"],
+                "ok": False,
+                "skipped": True,
+                "reason": f"depends_on failed: {unmet}",
+            }
+            healthy = False
+            continue
+
+        # ---------- RUN CHECK ----------
+        module = load_check_module(check["type"])
+        result = await module.run(check, context)
+
+        results[name] = result
+        executed.append(name)
+
+        if not result["ok"]:
+            healthy = False
+
+    return healthy, list(results.values())

app/loader.py

@@ -0,0 +1,22 @@
+import os
+import yaml
+
+SERVICES_DIR = "services"
+
+def load_services():
+    services = []
+
+    for filename in os.listdir(SERVICES_DIR):
+        if not filename.endswith(".yaml"):
+            continue
+
+        path = os.path.join(SERVICES_DIR, filename)
+        with open(path, "r", encoding="utf-8") as f:
+            cfg = yaml.safe_load(f)
+
+        if not cfg.get("enabled", True):
+            continue
+
+        services.append(cfg)
+
+    return services

app/main.py

@@ -0,0 +1,68 @@
+import asyncio
+from contextlib import asynccontextmanager
+from fastapi import FastAPI
+
+from app.loader import load_services
+from app.store import all, get, history, init_db
+from app.scheduler import run_service
+from app.utils.schema import load_schema, render_response
+
+services = load_services()
+TASKS = []
+
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    # ===== STARTUP =====
+    init_db()
+
+    for svc in services:
+        task = asyncio.create_task(run_service(svc))
+        TASKS.append(task)
+
+    yield  # 👉 App chạy ở đây
+
+    # ===== SHUTDOWN =====
+    for task in TASKS:
+        task.cancel()
+
+app = FastAPI(
+    title="Health Agent",
+    lifespan=lifespan
+)
+
+@app.get("/health")
+def health():
+    health_schema = load_schema("health_response")
+
+    services = []
+    for name, data in all().items():
+        checks = data.get("checks", [])
+        failed = sum(1 for c in checks if not c.get("ok"))
+
+        services.append({
+            "message": f"Service {name} is {data.get('status')}",
+            "name": name,
+            "status": data.get("status") == "HEALTHY",
+            "total_checks": len(checks),
+            "failed_checks": failed,
+            "last_updated": data.get("last_updated"),
+        })
+        
+        
+        print(services)
+
+    return render_response(health_schema, services)
+
+@app.get("/services")
+def services_list():
+    return all()
+
+@app.get("/services/{name}")
+def service_detail(name: str):
+    return get(name)
+
+@app.get("/services/{name}/history")
+def service_history(name: str, limit: int = 20):
+    return history(name, limit)

app/scheduler.py

@@ -0,0 +1,56 @@
+import asyncio
+import logging
+from app.engine import run_checks
+from app.store import update
+
+logger = logging.getLogger("health-agent")
+
+async def run_service(service_cfg):
+    interval = service_cfg.get("interval", 30)
+    service_name = service_cfg.get("service", "unknown")
+
+    while True:
+        try:
+            healthy, checks = await run_checks(service_cfg)
+
+            # 🔹 Health result (PASS / FAIL) → dữ liệu hợp lệ
+            update(
+                service_name,
+                {
+                    "status": "HEALTHY" if healthy else "UNHEALTHY",
+                    "checks": checks,
+                }
+            )
+
+            if not healthy:
+                logger.warning(
+                    "Service %s is UNHEALTHY", service_name
+                )
+
+        except asyncio.CancelledError:
+            # 🔹 App shutdown → task bị cancel
+            logger.info(
+                "Health check task stopped for service %s",
+                service_name
+            )
+            raise
+
+        except Exception as e:
+            # ❗ BUG của agent / engine (KHÔNG phải health fail)
+            update(
+                service_name,
+                {
+                    "status": "UNHEALTHY",
+                    "checks": [],
+                    "error": str(e),
+                }
+            )
+
+            logger.error(
+                "Agent error while checking service %s: %s",
+                service_name,
+                e,
+                exc_info=True,  # stacktrace chỉ dùng cho BUG
+            )
+
+        await asyncio.sleep(interval)

app/schemas/health_response.yaml

@@ -0,0 +1,14 @@
+type: array # 🔥 response là mảng
+
+item:
+  fields:
+    name:
+      type: string # tên service
+
+    status:
+      type: boolean # HEALTHY | UNHEALTHY
+
+    message: string
+
+meta:
+  description: Health status của từng service

app/store.py

@@ -0,0 +1,78 @@
+import sqlite3
+import json
+from datetime import datetime
+from threading import Lock
+
+DB_PATH = "health.db"
+
+RESULTS = {}
+LOCK = Lock()
+
+def init_db():
+    conn = sqlite3.connect(DB_PATH)
+    c = conn.cursor()
+    c.execute("""
+        CREATE TABLE IF NOT EXISTS service_results (
+            id INTEGER PRIMARY KEY AUTOINCREMENT,
+            service TEXT,
+            status TEXT,
+            checked_at TEXT,
+            details TEXT
+        )
+    """)
+    conn.commit()
+    conn.close()
+
+def update(service, data):
+    now = datetime.utcnow().isoformat()
+
+    with LOCK:
+        # RAM
+        RESULTS[service] = {
+            **data,
+            "last_check": now
+        }
+
+        # SQLite
+        conn = sqlite3.connect(DB_PATH)
+        c = conn.cursor()
+        c.execute(
+            "INSERT INTO service_results (service, status, checked_at, details) VALUES (?, ?, ?, ?)",
+            (
+                service,
+                data["status"],
+                now,
+                json.dumps(data)
+            )
+        )
+        conn.commit()
+        conn.close()
+        
+        
+def history(service, limit=20):
+    conn = sqlite3.connect(DB_PATH)
+    c = conn.cursor()
+    c.execute("""
+        SELECT status, checked_at, details
+        FROM service_results
+        WHERE service = ?
+        ORDER BY checked_at DESC
+        LIMIT ?
+    """, (service, limit))
+    rows = c.fetchall()
+    conn.close()
+
+    return [
+        {
+            "status": r[0],
+            "checked_at": r[1],
+            "details": json.loads(r[2])
+        }
+        for r in rows
+    ]
+
+def all():
+    return RESULTS
+
+def get(service):
+    return RESULTS.get(service)

app/utils/schema.py

@@ -0,0 +1,50 @@
+import yaml
+import time
+from pathlib import Path
+
+BASE_DIR = Path(__file__).resolve().parent.parent
+
+
+def load_schema(name: str) -> dict:
+    """
+    Load schema YAML từ app/schemas/<name>.yaml
+    """
+    path = BASE_DIR / "schemas" / f"{name}.yaml"
+    with open(path, "r", encoding="utf-8") as f:
+        return yaml.safe_load(f)
+
+
+def render_response(schema: dict, data):
+    schema_type = schema.get("type")
+
+    # ---------- ARRAY ----------
+    if schema_type == "array":
+        if not isinstance(data, list):
+            return []
+
+        item_schema = schema.get("item", {})
+        return [
+            render_response(
+                {
+                    "type": "object",
+                    "fields": item_schema.get("fields", {})
+                },
+                item
+            )
+            for item in data
+        ]
+
+    # ---------- OBJECT ----------
+    if schema_type == "object":
+        fields = schema.get("fields", {})
+        result = {}
+
+        for key, field_cfg in fields.items():
+            if key in data:
+                result[key] = data[key]
+
+        return result
+
+    # ---------- FALLBACK ----------
+    return data
+

app/utils/security.py

@@ -0,0 +1,23 @@
+SENSITIVE_KEYS = {
+    "password",
+    "token",
+    "access_token",
+    "refresh_token",
+    "authorization",
+}
+
+def mask_sensitive(data):
+    if isinstance(data, dict):
+        return {
+            k: (
+                "***"
+                if k.lower() in SENSITIVE_KEYS
+                else mask_sensitive(v)
+            )
+            for k, v in data.items()
+        }
+
+    if isinstance(data, list):
+        return [mask_sensitive(v) for v in data]
+
+    return data

app/utils/template.py

@@ -0,0 +1,20 @@
+import re
+
+_PATTERN = re.compile(r"\{\{(\w+)\}\}")
+
+def render(value, context: dict):
+    if not isinstance(value, str):
+        return value
+
+    def replace(match):
+        key = match.group(1)
+        return str(context.get(key, ""))
+
+    return _PATTERN.sub(replace, value)
+
+
+def render_headers(headers: dict, context: dict):
+    return {
+        k: render(v, context)
+        for k, v in headers.items()
+    }

docker-compose.yml

@@ -0,0 +1,18 @@
+version: "3.9"
+
+services:
+  health-agent:
+    build: .
+    container_name: health-agent
+    ports:
+      - "8000:8000"
+
+    volumes:
+      # 🔥 mount folder services
+      - ./services:/app/services
+
+    environment:
+      - SERVICES_DIR=/app/services
+      - TZ=Asia/Ho_Chi_Minh
+
+    restart: unless-stopped

requirements.txt

@@ -0,0 +1,5 @@
+fastapi
+uvicorn
+pyyaml
+httpx
+jsonpath-ng

services/bids-service.yaml

@@ -0,0 +1,129 @@
+service: bids-service
+enabled: true
+interval: 30
+base_url: http://localhost:4000/api/v1/admin
+
+checks:
+  - type: http
+    name: login
+    url: /auth/login
+    method: POST
+    body:
+      username: admin
+      password: Admin@123
+    expect:
+      status_code: 201
+    save:
+      cookies: __cookies__
+
+  - type: http
+    name: me
+    depends_on: [login]
+    url: /auth/me
+    expect:
+      status_code: 200
+
+  - type: http
+    name: webs
+    depends_on: [login]
+    url: /web-bids
+    expect:
+      status_code: 200
+
+  - type: http
+    name: create-webs
+    depends_on: [login]
+    body:
+      origin_url: http://abc.com
+    method: POST
+    url: /web-bids
+    expect:
+      status_code: 400
+
+  - type: http
+    name: update-webs
+    depends_on: [login]
+    body:
+      origin_url: http://abc.com
+    method: PUT
+    url: /web-bids/0
+    expect:
+      status_code: 404
+
+  - type: http
+    name: delete-webs
+    depends_on: [login]
+    method: DELETE
+    url: /web-bids/0
+    expect:
+      status_code: 404
+
+  - type: http
+    name: admins
+    depends_on: [login]
+    url: /admins
+    expect:
+      status_code: 200
+
+  - type: http
+    name: admins
+    depends_on: [login]
+    url: /admins
+    expect:
+      status_code: 200
+
+  - type: http
+    name: admins
+    depends_on: [login]
+    url: /admins
+    expect:
+      status_code: 200
+
+  - type: http
+    name: bids
+    depends_on: [login]
+    url: /bids
+    expect:
+      status_code: 200
+
+  - type: http
+    name: create-bids
+    depends_on: [login]
+    body:
+      url: http://abc.com
+    method: POST
+    url: /bids
+    expect:
+      status_code: 400
+
+  - type: http
+    name: update-bids
+    depends_on: [login]
+    body:
+      url: http://abc.com
+    method: PUT
+    url: /bids/0
+    expect:
+      status_code: 400
+
+  - type: http
+    name: delete-bids
+    depends_on: [login]
+    method: DELETE
+    url: /bids/0
+    expect:
+      status_code: 404
+
+  - type: http
+    name: out-bid-logs
+    depends_on: [login]
+    url: /out-bid-logs
+    expect:
+      status_code: 200
+
+  - type: http
+    name: send-message-histories
+    depends_on: [login]
+    url: /send-message-histories
+    expect:
+      status_code: 200

services/default.yaml

@@ -0,0 +1,90 @@
+# ================================
+# SERVICE CONFIG
+# ================================
+
+service: default-service # Tên service (duy nhất trong toàn hệ thống)
+enabled: false # true = bật check, false = bỏ qua service này
+interval: 30 # Chu kỳ chạy health check (giây)
+
+# ================================
+# CHECK LIST
+# ================================
+checks:
+  # ------------------------------------------------
+  # 1️⃣ LOGIN CHECK (TOKEN TRONG RESPONSE BODY)
+  # - API trả access_token trong JSON
+  # - Token sẽ được save vào context
+  # ------------------------------------------------
+  - type: http # Loại check: http | shell
+    name: login-token # Tên check
+
+    url: http://localhost:4000/api/v1/admin/auth/login
+    method: POST
+    timeout: 5
+
+    body:
+      email: admin@example.com
+      password: Admin@1234
+
+    # Điều kiện pass
+    expect:
+      status_code: 200
+      json:
+        access_token: $.data.access_token
+        # ↑ JSONPath phải tồn tại
+
+    # Lưu token vào context
+    save:
+      token: $.data.access_token
+      # context["token"] = response.data.access_token
+      # cookies: __cookies__ => Trường hợp dùng cho cần save cookie
+
+    # 👉 Check này BLOCKING
+    # 👉 Fail là các check phụ thuộc không chạy
+
+  # ------------------------------------------------
+  # 2️⃣ ME CHECK (DÙNG TOKEN TỪ CONTEXT)
+  # - Authorization header dùng {{token}}
+  # ------------------------------------------------
+  - type: http
+    name: me
+
+    url: http://localhost:4000/api/v1/admin/me
+    method: GET
+    timeout: 3
+
+    headers:
+      Authorization: "Bearer {{token}}"
+      # ↑ token lấy từ context của service này
+
+    expect:
+      status_code: 200
+
+  # ------------------------------------------------
+  # 3️⃣ HEALTH CHECK (ĐỘC LẬP)
+  # - Không cần login
+  # - Login fail vẫn chạy
+  # ------------------------------------------------
+  - type: http
+    name: health
+
+    url: http://localhost:4000/health
+    method: GET
+    timeout: 2
+
+    expect:
+      status_code: 200
+
+    independent: true # 🔥 Không phụ thuộc login
+
+  # ------------------------------------------------
+  # 4️⃣ PROCESS CHECK (SHELL)
+  # - Kiểm tra process OS
+  # -> Chưa test nào dùng thì viết tiếp
+  # ------------------------------------------------
+  - type: shell
+    name: process
+
+    command: "pgrep -f auth-service"
+
+    independent: true