303 lines
11 KiB
Python
303 lines
11 KiB
Python
|
"""L2b YARA 规则与二进制分析引擎。
|
|||
|
|
|
|||
|
|
包含：
|
|||
|
|
- YARA 规则匹配（规则库从 ``rules/`` 目录加载，支持热加载）
|
|||
|
|
- PE 文件结构分析（pefile）：节区异常、导入表、编译信息、签名
|
|||
|
|
- 可执行文件能力识别（capa，若已安装）
|
|||
|
|
"""
|
|||
|
|
|
|||
|
|
from __future__ import annotations
|
|||
|
|
|
|||
|
|
import os
|
|||
|
|
import threading
|
|||
|
|
from typing import Any, Dict, List, Optional
|
|||
|
|
|
|||
|
|
from app.engines.base import Engine, EngineResult, Finding, ScanTarget, Severity
|
|||
|
|
|
|||
|
|
BINARY_EXTS = {".exe", ".dll", ".sys", ".scr", ".com", ".cpl", ".ocx", ".msi", ".apk", ".jar", ".so", ".bin", ".elf"}
|
|||
|
|
|
|||
|
|
# ---------- 模块级全局 YARA 规则单例 ----------
|
|||
|
|
# 所有 YaraEngine 实例共享同一个编译结果，避免重复 compile
|
|||
|
|
# rules_dir 不同时（极少见）才分叉加载
|
|||
|
|
_YARA_GLOBAL: Dict[str, Dict[str, Any]] = {}
|
|||
|
|
_YARA_LOCK = threading.Lock()
|
|||
|
|
|
|||
|
|
|
|||
|
|
def _get_global_yara(rules_dir: str) -> Optional[Any]:
|
|||
|
|
"""获取全局共享的 YARA Rules 对象（自动按目录 mtime 重载）。"""
|
|||
|
|
with _YARA_LOCK:
|
|||
|
|
cached = _YARA_GLOBAL.get(rules_dir)
|
|||
|
|
import yara as _yr
|
|||
|
|
newest = 0.0
|
|||
|
|
files: List[str] = []
|
|||
|
|
if os.path.isdir(rules_dir):
|
|||
|
|
for name in sorted(os.listdir(rules_dir)):
|
|||
|
|
if name.endswith((".yar", ".yara")):
|
|||
|
|
path = os.path.join(rules_dir, name)
|
|||
|
|
files.append(path)
|
|||
|
|
newest = max(newest, os.path.getmtime(path))
|
|||
|
|
|
|||
|
|
if cached and cached.get("mtime") == newest:
|
|||
|
|
return cached.get("rules")
|
|||
|
|
|
|||
|
|
if not files:
|
|||
|
|
_YARA_GLOBAL[rules_dir] = {"rules": None, "mtime": newest}
|
|||
|
|
return None
|
|||
|
|
|
|||
|
|
try:
|
|||
|
|
compiled = _yr.compile(filepaths={os.path.basename(p): p for p in files})
|
|||
|
|
except Exception:
|
|||
|
|
_YARA_GLOBAL[rules_dir] = {"rules": None, "mtime": newest}
|
|||
|
|
return None
|
|||
|
|
|
|||
|
|
_YARA_GLOBAL[rules_dir] = {"rules": compiled, "mtime": newest}
|
|||
|
|
return compiled
|
|||
|
|
|
|||
|
|
|
|||
|
|
def _discover_rules_dir() -> str:
|
|||
|
|
"""自动探测 YARA 规则目录（优先环境变量 > /app/rules > 项目根 rules/）。"""
|
|||
|
|
env_dir = os.environ.get("MESHSAFE_RULES_DIR", "").strip()
|
|||
|
|
if env_dir and os.path.isdir(env_dir):
|
|||
|
|
return env_dir
|
|||
|
|
|
|||
|
|
for candidate in ("/app/rules", "rules", "./rules", "../rules"):
|
|||
|
|
if os.path.isdir(candidate):
|
|||
|
|
return os.path.abspath(candidate)
|
|||
|
|
|
|||
|
|
return env_dir or "/app/rules"
|
|||
|
|
|
|||
|
|
|
|||
|
|
class YaraEngine(Engine):
|
|||
|
|
"""YARA 规则匹配引擎（使用全局共享的编译结果）。"""
|
|||
|
|
|
|||
|
|
name = "yara"
|
|||
|
|
layer = "L2B"
|
|||
|
|
|
|||
|
|
def __init__(self, rules_dir: str | None = None) -> None:
|
|||
|
|
# 显式传入时才用传入值，否则走自动探测
|
|||
|
|
self.rules_dir = rules_dir if rules_dir else _discover_rules_dir()
|
|||
|
|
|
|||
|
|
def _load_rules(self) -> Any:
|
|||
|
|
"""委托给全局单例。"""
|
|||
|
|
try:
|
|||
|
|
import yara # noqa: F401 — 只是探测可用性
|
|||
|
|
except Exception: # noqa: BLE001
|
|||
|
|
return None
|
|||
|
|
return _get_global_yara(self.rules_dir)
|
|||
|
|
|
|||
|
|
def scan(self, target: ScanTarget, ctx: Any) -> EngineResult:
|
|||
|
|
result = EngineResult(engine=self.name)
|
|||
|
|
rules = self._load_rules()
|
|||
|
|
if rules is None:
|
|||
|
|
result.ok = False
|
|||
|
|
result.error = "YARA 规则库不可用或编译失败"
|
|||
|
|
return result
|
|||
|
|
|
|||
|
|
try:
|
|||
|
|
matches = rules.match(target.path)
|
|||
|
|
except Exception as exc: # noqa: BLE001
|
|||
|
|
result.ok = False
|
|||
|
|
result.error = f"YARA 匹配失败：{exc}"
|
|||
|
|
return result
|
|||
|
|
|
|||
|
|
for match in matches:
|
|||
|
|
severity, weight = _judge_yara_rule(match.rule, match.tags)
|
|||
|
|
# 覆盖：如果 ctx 里的策略为该 YARA 规则定义了精细权重，用它
|
|||
|
|
try:
|
|||
|
|
policy = ctx.policy_engine.policy
|
|||
|
|
yara_weights = getattr(policy, "yara_weights", {}) or {}
|
|||
|
|
if match.rule in yara_weights:
|
|||
|
|
weight = float(yara_weights[match.rule])
|
|||
|
|
except Exception:
|
|||
|
|
pass # ctx 没有 policy 字段时保持默认 weight
|
|||
|
|
result.findings.append(
|
|||
|
|
Finding(
|
|||
|
|
engine=self.name,
|
|||
|
|
layer=self.layer,
|
|||
|
|
severity=severity,
|
|||
|
|
rule=match.rule,
|
|||
|
|
detail=f"YARA 规则 {match.rule} 命中（标签：{', '.join(match.tags) or '无'}）",
|
|||
|
|
weight=weight,
|
|||
|
|
meta={"tags": list(match.tags)},
|
|||
|
|
)
|
|||
|
|
)
|
|||
|
|
# rule_set_version 用全局 compiled 对象的 id（每次 reload 变）
|
|||
|
|
try:
|
|||
|
|
_r = _get_global_yara(self.rules_dir)
|
|||
|
|
result.meta["rule_set_version"] = str(id(_r)) if _r else "unavailable"
|
|||
|
|
except Exception:
|
|||
|
|
result.meta["rule_set_version"] = "unknown"
|
|||
|
|
return result
|
|||
|
|
|
|||
|
|
def health(self) -> Dict[str, Any]:
|
|||
|
|
rules = self._load_rules()
|
|||
|
|
ok = rules is not None
|
|||
|
|
return {
|
|||
|
|
"engine": self.name,
|
|||
|
|
"version": "rules" if ok else "none",
|
|||
|
|
"ok": ok,
|
|||
|
|
"status": "ok" if ok else "degraded",
|
|||
|
|
"error": None if ok else "YARA 规则库未加载",
|
|||
|
|
"info": f"rules_dir={self.rules_dir}" if not ok else "rules loaded",
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
class BinaryEngine(Engine):
|
|||
|
|
"""PE/二进制文件结构分析引擎。"""
|
|||
|
|
|
|||
|
|
name = "binary"
|
|||
|
|
layer = "L2B"
|
|||
|
|
|
|||
|
|
def scan(self, target: ScanTarget, ctx: Any) -> EngineResult:
|
|||
|
|
result = EngineResult(engine=self.name)
|
|||
|
|
ext = (target.declared_ext or "").lower()
|
|||
|
|
if ext not in BINARY_EXTS:
|
|||
|
|
return result
|
|||
|
|
|
|||
|
|
try:
|
|||
|
|
import pefile
|
|||
|
|
except Exception: # noqa: BLE001
|
|||
|
|
result.ok = False
|
|||
|
|
result.error = "pefile 未安装"
|
|||
|
|
return result
|
|||
|
|
|
|||
|
|
try:
|
|||
|
|
pe = pefile.PE(target.path, fast_load=False)
|
|||
|
|
except Exception as exc: # noqa: BLE001
|
|||
|
|
result.findings.append(
|
|||
|
|
Finding(
|
|||
|
|
engine=self.name,
|
|||
|
|
layer=self.layer,
|
|||
|
|
severity=Severity.MEDIUM.value,
|
|||
|
|
rule="PeParseFailed",
|
|||
|
|
detail=f"PE 结构解析失败：{exc}",
|
|||
|
|
weight=10.0,
|
|||
|
|
)
|
|||
|
|
)
|
|||
|
|
return result
|
|||
|
|
|
|||
|
|
try:
|
|||
|
|
result.findings.extend(self._analyze(pe, target))
|
|||
|
|
finally:
|
|||
|
|
try:
|
|||
|
|
pe.close()
|
|||
|
|
except Exception: # noqa: BLE001
|
|||
|
|
pass
|
|||
|
|
|
|||
|
|
return result
|
|||
|
|
|
|||
|
|
def _analyze(self, pe: Any, target: ScanTarget) -> List[Finding]:
|
|||
|
|
findings: List[Finding] = []
|
|||
|
|
|
|||
|
|
# 1) 节区熵值：加密/加壳的节区熵接近 8
|
|||
|
|
for section in getattr(pe, "sections", []) or []:
|
|||
|
|
try:
|
|||
|
|
name = section.Name.rstrip(b"\x00").decode("utf-8", errors="replace")
|
|||
|
|
entropy = section.get_entropy()
|
|||
|
|
except Exception: # noqa: BLE001
|
|||
|
|
continue
|
|||
|
|
if entropy >= 7.2:
|
|||
|
|
findings.append(
|
|||
|
|
Finding(
|
|||
|
|
engine=self.name,
|
|||
|
|
layer=self.layer,
|
|||
|
|
severity=Severity.MEDIUM.value,
|
|||
|
|
rule="HighEntropySection",
|
|||
|
|
detail=f"节区 {name} 熵值 {entropy:.2f}，可能经过加密或加壳",
|
|||
|
|
weight=20.0,
|
|||
|
|
)
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
# 2) 可疑导入表
|
|||
|
|
suspicious_imports = self._suspicious_imports(pe)
|
|||
|
|
if suspicious_imports:
|
|||
|
|
findings.append(
|
|||
|
|
Finding(
|
|||
|
|
engine=self.name,
|
|||
|
|
layer=self.layer,
|
|||
|
|
severity=Severity.HIGH.value,
|
|||
|
|
rule="SuspiciousImports",
|
|||
|
|
detail=f"导入表包含高危 API：{', '.join(suspicious_imports[:8])}",
|
|||
|
|
weight=45.0,
|
|||
|
|
)
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
# 3) 无签名可执行文件
|
|||
|
|
if not self._has_signature(pe):
|
|||
|
|
findings.append(
|
|||
|
|
Finding(
|
|||
|
|
engine=self.name,
|
|||
|
|
layer=self.layer,
|
|||
|
|
severity=Severity.MEDIUM.value,
|
|||
|
|
rule="UnsignedExecutable",
|
|||
|
|
detail="可执行文件未携带数字签名",
|
|||
|
|
weight=20.0,
|
|||
|
|
)
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
# 4) 编译时间戳异常（明显晚于其他节区时间，或为 1970/未来时间）
|
|||
|
|
timestamp = getattr(pe, "FILE_HEADER", None)
|
|||
|
|
if timestamp is not None:
|
|||
|
|
try:
|
|||
|
|
ts = int(timestamp.TimeDateStamp)
|
|||
|
|
if ts == 0 or ts > 4102444800: # 2100 年之后
|
|||
|
|
findings.append(
|
|||
|
|
Finding(
|
|||
|
|
engine=self.name,
|
|||
|
|
layer=self.layer,
|
|||
|
|
severity=Severity.LOW.value,
|
|||
|
|
rule="SuspiciousTimestamp",
|
|||
|
|
detail="PE 编译时间戳异常，可能被伪造",
|
|||
|
|
weight=10.0,
|
|||
|
|
)
|
|||
|
|
)
|
|||
|
|
except Exception: # noqa: BLE001
|
|||
|
|
pass
|
|||
|
|
|
|||
|
|
return findings
|
|||
|
|
|
|||
|
|
def _suspicious_imports(self, pe: Any) -> List[str]:
|
|||
|
|
"""检测高危 API 导入。"""
|
|||
|
|
suspicious = {
|
|||
|
|
"VirtualAllocEx", "WriteProcessMemory", "CreateRemoteThread",
|
|||
|
|
"NtUnmapViewOfSection", "SetWindowsHookEx", "OpenProcess",
|
|||
|
|
"WinHttpOpen", "InternetOpenUrl", "URLDownloadToFile",
|
|||
|
|
"CryptEncrypt", "AdjustTokenPrivileges", "CreateService",
|
|||
|
|
"ShellExecute", "WinExec", "IsDebuggerPresent", "CheckRemoteDebuggerPresent",
|
|||
|
|
}
|
|||
|
|
found: List[str] = []
|
|||
|
|
try:
|
|||
|
|
for entry in getattr(pe, "DIRECTORY_ENTRY_IMPORT", []) or []:
|
|||
|
|
for imp in entry.imports:
|
|||
|
|
if imp.name:
|
|||
|
|
name = imp.name.decode("utf-8", errors="replace")
|
|||
|
|
if name in suspicious:
|
|||
|
|
found.append(name)
|
|||
|
|
except Exception: # noqa: BLE001
|
|||
|
|
pass
|
|||
|
|
return sorted(set(found))
|
|||
|
|
|
|||
|
|
def _has_signature(self, pe: Any) -> bool:
|
|||
|
|
"""检查 PE 是否包含 Authenticode 签名目录。"""
|
|||
|
|
try:
|
|||
|
|
directory = pe.OPTIONAL_HEADER.DATA_DIRECTORY[4] # IMAGE_DIRECTORY_ENTRY_SECURITY
|
|||
|
|
return bool(directory.VirtualAddress and directory.Size)
|
|||
|
|
except Exception: # noqa: BLE001
|
|||
|
|
return False
|
|||
|
|
|
|||
|
|
|
|||
|
|
def _judge_yara_rule(rule: str, tags: List[str]) -> tuple[str, float]:
|
|||
|
|
"""依据 YARA 规则名与标签判定严重度。"""
|
|||
|
|
tag_set = {t.lower() for t in tags}
|
|||
|
|
text = f"{rule} {' '.join(tags)}".lower()
|
|||
|
|
|
|||
|
|
if "critical" in tag_set or any(k in text for k in ("malware", "ransomware", "backdoor", "c2", "exploit")):
|
|||
|
|
return Severity.CRITICAL.value, 85.0
|
|||
|
|
if "high" in tag_set or any(k in text for k in ("trojan", "dropper", "obfusc", "shellcode")):
|
|||
|
|
return Severity.HIGH.value, 60.0
|
|||
|
|
if "medium" in tag_set or any(k in text for k in ("suspicious", "pua", "packer")):
|
|||
|
|
return Severity.MEDIUM.value, 30.0
|
|||
|
|
if "suspicious" in tag_set:
|
|||
|
|
return Severity.MEDIUM.value, 30.0
|
|||
|
|
return Severity.LOW.value, 10.0
|