Files
vps-manager/app/services/sync_service.py
T
gouki f09c0f6e4b fix(review): 处理全量审查剩余项(M4-M8/M11/M12 + 21 条 low)
后端
- asset_service: delete_asset 先清 subdomains/site_certs 子表(SQLite 不强制
  FK 且 assets.id 无 AUTOINCREMENT,删最大 id 后新建会复用 rowid 导致孤儿误挂);
  metrics.db 清理改 best-effort,不再让已成功的删除返回 500
- provider_service: slug 唯一约束竞态补 IntegrityError 兜底(与 account_service 一致)
- account_service/sync_service: raise HTTPException 补 from e,保留原始异常链
- sync_service: 删死代码 _load_config;新增 BalanceUnsupported,平台无余额接口时
  计入 skipped 而不是静默当作失败;到期日写库前归一
- ai: 删掉 OpenAI 早已下线的 billing 余额接口实现,能力探测如实报「不支持」
- adapters/base: _overridden 改走 MRO,中间层基类实现的能力不再被误判为未实现
- core/timeutils: 删死代码 utcnow_iso(注释声称 Z 后缀,实际输出 +00:00)
- main: lifespan 退出时 await 被 cancel 的清理任务,避免 pending task 被销毁
- ssl_service: 写 site_certs/subdomains 前校验 asset_id 存在,杜绝悬空引用;
  到期阈值贯穿 _judge_status/_apply_probe/check_one/check_all_site_certs
- routers/credentials、accounts: 明文密码 GET 补 Cache-Control: no-store
- routers/stats: /stats/expiring 的 Query 描述与实际语义对齐(不设下界,含已过期)

Agent
- collector: 解析 sshd_config 的 Include(含 sshd_config.d/*.conf,按 sshd 语义
  「首次出现的值生效」),未显式配置的关键字按编译默认值判定,消除假 pass;
  防火墙检测改用 iptables -S 统计 -A 规则行并读 INPUT 策略(-L -n 恒有链头输出)
- reporter: 4xx(除 429)不再重试直接退出,429 优先按 Retry-After 退避

安全与部署
- core/security: require_agent_key 改为对外 fail-closed——tailnet/本机放行,
  外部来源未配 AGENT_KEY 时一律 401(此前全放行,而服务默认监听 0.0.0.0,
  公网上任何人可伪造监控指标与安全检查结果)
- main: 启动时把「未配 API_KEY/AGENT_KEY 则外部来源不可用」打进日志
- run.sh: HOST/PORT 可用环境变量覆盖,并说明默认 0.0.0.0 的权衡
- deploy/setup.sh: 新装时自动生成随机 API_KEY/AGENT_KEY 并提示同步到各 VPS;
  注册 vps-cert-check.timer
- deploy/vps-cert-check.{service,timer}: 新增每日 09:30 的站点证书到期探测
  (此前 check_site_certs.py 没有任何 unit,证书提醒不会自动跑)
- deploy/update.sh: sync_units 对 timer 用 enable --now(新增 timer 只 restart
  不会建开机自启软链),主服务 unit 变更后额外 restart;pip 安装失败不再被吞
- scripts/check_site_certs.py: threshold 参数真正生效并打印
- scripts/make_migration_bundle.py: 缺 assets.db 直接 abort;umask 077 + chmod 600
  (包内含 .env 的 MASTER_KEY 与 escrow,不能默认 644 落在 /tmp)
- scripts/restore_migration_bundle.py: 校验之外补必备文件预检,缺文件在触碰
  目标机状态之前就 abort

前端
- store: 拆分列表口径 assets 与统计口径 allAssets(无筛选时同源,不重复请求);
  删除已无消费者的 expiring/loadExpiring(每次 loadAll 白发一个请求)
- dashboard/monitor/providers/subscriptions/modals: 统计与计数改用 allAssets,
  资产页的筛选(含统计卡片点击跳转带的筛选)不再让其他视图的数字缩水
- modals: 账号同步后补 loadOverview;settings: 导入后改 loadAll
- api: countdownText 支持 unit/empty 选项,monitor/domains 各自的实现改为委托,
  assets/domains/subscriptions 的到期列不再直出「(-3天)」
- servers: cpu/mem/disk 百分比统一判空,采样缺项不再渲染「null%」
- assets: 同步 AI 余额的结果提示区分 skipped(平台无余额接口)与 failed
- modals: 修正凭据库头注释——2FA 动态码是选填 + 可事后补验,与实现一致

测试
- 新增 tests/test_agent_collector.py(8 条)、tests/test_security.py(11 条)
- 全量 113 passed;改动 JS 均通过 node --check,deploy 脚本通过 bash -n
- 浏览器实测:筛选态下总览/监控中心/平台页/付费资产的统计仍为全量口径,
  到期文案与 null 百分比渲染正确,console 零报错
2026-09-23 10:15:52 +00:00

448 lines
17 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
"""云厂商同步服务
将适配器返回的标准化资产(NormalizedVPS / NormalizedDomain / AccountInfo)
写入或更新到资产库。以 (provider_id, external_id) 作为去重键,
已存在则更新状态/详情,不存在则新建资产。
凭证层级:API 配置存在账号(Account.api_config_encrypted)上,
同步按账号维度进行(test_account/sync_account);同步产出的资产
自动挂到该账号名下(Asset.account)。平台不再持有凭证。
"""
import json
import logging
from typing import Optional
from fastapi import HTTPException, status
from sqlmodel import Session, select
logger = logging.getLogger("vps-manager.sync")
from app.adapters import registry
from app.adapters.base import BaseAdapter, parse_iso_date
from app.core import crypto
from app.core.timeutils import utcnow
from app.models.asset import (
AIAccount,
Account,
Asset,
AssetStatus,
AssetType,
DomainDetail,
VPSDetail,
)
from app.models.provider import Provider
_VALID_STATUS = {s.value for s in AssetStatus}
def _load_account_config(account: Account) -> dict:
"""解密账号的 API 配置 JSON"""
plain = crypto.decrypt(account.api_config_encrypted)
if not plain:
return {}
try:
return json.loads(plain)
except (json.JSONDecodeError, TypeError):
return {}
def _find_provider_by_platform(session: Session, platform: str) -> Provider:
"""按账号的 platform(slug 或名称)定位平台"""
provider = session.exec(select(Provider).where(Provider.slug == platform)).first()
if not provider:
provider = session.exec(select(Provider).where(Provider.name == platform)).first()
if not provider:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"账号所属平台不存在:{platform}(请先在账号编辑中选择有效平台)",
)
return provider
def _get_provider(session: Session, provider_id: int) -> Provider:
provider = session.get(Provider, provider_id)
if not provider:
raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="平台不存在")
return provider
def _build_adapter(provider: Provider, config: dict) -> BaseAdapter:
if not provider.sdk_type:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST, detail="该平台未配置 SDK 类型(sdk_type)"
)
if not registry.is_supported(provider.sdk_type):
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"暂不支持的 SDK 类型:{provider.sdk_type}",
)
return registry.get_adapter(provider.sdk_type, config)
def _get_account(session: Session, account_id: int) -> Account:
account = session.get(Account, account_id)
if not account:
raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="账号不存在")
return account
def _build_adapter_for_account(session: Session, account: Account):
"""按账号构建适配器:平台定 sdk_type,账号提供凭证"""
if not account.platform:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="该账号未指定所属平台,无法确定 SDK 类型",
)
provider = _find_provider_by_platform(session, account.platform)
config = _load_account_config(account)
adapter = _build_adapter(provider, config)
return provider, adapter
def _norm_status(raw: Optional[str]) -> AssetStatus:
return AssetStatus(raw) if raw in _VALID_STATUS else AssetStatus.UNKNOWN
def _missing_config(adapter: BaseAdapter) -> list:
"""检查适配器所需凭证是否已配置"""
return [f for f in adapter.required_config if not adapter.config.get(f)]
def test_provider(session: Session, provider_id: int) -> dict:
"""(兼容入口)测试平台连接:凭证已下沉到账号,自动找该平台第一个配了凭证的账号"""
provider = _get_provider(session, provider_id)
account = _first_account_with_config(session, provider)
if not account:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="凭证已改为在账号上配置:请先在该平台的账号管理中新建账号并填写 API 配置",
)
return test_account(session, account.id)
def _first_account_with_config(session: Session, provider: Provider):
"""找该平台下第一个配置了 API 凭证的账号(兼容旧入口)"""
return session.exec(
select(Account)
.where(Account.platform == provider.slug)
.where(Account.api_config_encrypted.is_not(None)) # type: ignore[union-attr]
.order_by(Account.id.asc())
).first()
def test_account(session: Session, account_id: int) -> dict:
"""测试账号凭证有效性(按账号 platform 定 SDK,凭证取自账号)"""
account = _get_account(session, account_id)
provider, adapter = _build_adapter_for_account(session, account)
base = {"capabilities": adapter.capabilities(), "sdk_type": provider.sdk_type, "account": account.name}
missing = _missing_config(adapter)
if missing:
return {
"ok": False,
"message": f"缺少凭证配置:{', '.join(missing)}(请在账号编辑里填写 API 配置)",
**base,
}
result = adapter.test_connection()
result.update(base)
return result
def _find_asset(session: Session, provider_id: int, external_id: str, asset_type: AssetType):
return session.exec(
select(Asset).where(
Asset.provider_id == provider_id,
Asset.external_id == external_id,
Asset.asset_type == asset_type,
)
).first()
def _sync_vps(session: Session, provider: Provider, adapter: BaseAdapter, account_id: Optional[int] = None) -> dict:
created = updated = 0
for vps in adapter.list_vps():
existing = _find_asset(session, provider.id, vps.external_id, AssetType.VPS)
if existing:
existing.name = vps.name or existing.name
existing.status = _norm_status(vps.status)
if vps.monthly_cost is not None:
existing.cost = vps.monthly_cost
existing.currency = vps.currency
if account_id:
existing.account_id = account_id
session.add(existing)
detail = session.exec(
select(VPSDetail).where(VPSDetail.asset_id == existing.id)
).first()
if detail:
detail.ip_address = vps.ip_address or detail.ip_address
detail.region = vps.region or detail.region
detail.os = vps.os or detail.os
detail.cpu_cores = vps.cpu_cores or detail.cpu_cores
detail.memory_gb = vps.memory_gb or detail.memory_gb
detail.disk_gb = vps.disk_gb or detail.disk_gb
session.add(detail)
updated += 1
else:
asset = Asset(
name=vps.name or vps.external_id,
asset_type=AssetType.VPS,
provider=provider.slug,
provider_id=provider.id,
external_id=vps.external_id,
account_id=account_id,
status=_norm_status(vps.status),
cost=vps.monthly_cost or 0,
currency=vps.currency,
)
session.add(asset)
session.flush() # 获取 asset.id,统一在循环外提交
session.add(
VPSDetail(
asset_id=asset.id,
ip_address=vps.ip_address or "",
region=vps.region,
os=vps.os,
cpu_cores=vps.cpu_cores or 1,
memory_gb=vps.memory_gb or 1,
disk_gb=vps.disk_gb or 20,
)
)
created += 1
session.commit()
return {"created": created, "updated": updated}
def _sync_domains(session: Session, provider: Provider, adapter: BaseAdapter, account_id: Optional[int] = None) -> dict:
created = updated = 0
for dom in adapter.list_domains():
# 适配器契约上 expiry_date 已是 date;这里再归一一次,
# 防止旧适配器/自定义适配器返回字符串时整个 commit 抛 TypeError
expiry = parse_iso_date(dom.expiry_date)
existing = _find_asset(session, provider.id, dom.external_id, AssetType.DOMAIN)
if existing:
existing.status = _norm_status(dom.status)
if expiry:
existing.expiry_date = expiry
if account_id:
existing.account_id = account_id
session.add(existing)
detail = session.exec(
select(DomainDetail).where(DomainDetail.asset_id == existing.id)
).first()
if detail:
detail.domain_name = dom.domain_name or detail.domain_name
detail.registrar = dom.registrar or detail.registrar
session.add(detail)
updated += 1
else:
asset = Asset(
name=dom.domain_name,
asset_type=AssetType.DOMAIN,
provider=provider.slug,
provider_id=provider.id,
external_id=dom.external_id,
account_id=account_id,
status=_norm_status(dom.status),
expiry_date=expiry,
)
session.add(asset)
session.flush() # 获取 asset.id,统一在循环外提交
session.add(
DomainDetail(
asset_id=asset.id,
domain_name=dom.domain_name,
registrar=dom.registrar or provider.slug,
)
)
created += 1
session.commit()
return {"created": created, "updated": updated}
def sync_provider(session: Session, provider_id: int) -> dict:
"""(兼容入口)同步平台资产:凭证已下沉到账号,自动找该平台第一个配了凭证的账号"""
provider = _get_provider(session, provider_id)
account = _first_account_with_config(session, provider)
if not account:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="凭证已改为在账号上配置:请先在该平台的账号管理中新建账号并填写 API 配置",
)
return sync_account(session, account.id)
def sync_account(session: Session, account_id: int) -> dict:
"""同步账号资产到本地库(凭证取自账号,同步产出自动挂到该账号名下)"""
account = _get_account(session, account_id)
provider, adapter = _build_adapter_for_account(session, account)
missing = _missing_config(adapter)
if missing:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"缺少凭证配置:{', '.join(missing)}(请先在账号编辑里填写 API 配置)",
)
caps = adapter.capabilities()
result = {"provider": provider.slug, "sdk_type": provider.sdk_type, "account": account.name}
if caps["list_vps"]:
try:
result["vps"] = _sync_vps(session, provider, adapter, account.id)
except Exception as e: # noqa: BLE001
# 写库失败会让 Session 进入 pending-rollback 状态,必须先回滚才能继续,
# 否则下面更新 last_synced_at 的 commit 会抛 PendingRollbackError → 接口 500
session.rollback()
result["vps_error"] = str(e)
if caps["list_domains"]:
try:
result["domains"] = _sync_domains(session, provider, adapter, account.id)
except Exception as e: # noqa: BLE001
session.rollback()
result["domains_error"] = str(e)
if caps["get_account"]:
try:
result["account_info"] = adapter.get_account().to_dict()
except Exception as e: # noqa: BLE001
result["account_info_error"] = str(e)
account.last_synced_at = utcnow()
session.add(account)
session.commit()
result["last_synced_at"] = account.last_synced_at.isoformat()
logger.info("同步账号 account=%s provider=%s result=%s", account.name, provider.slug, result)
return result
# AI 服务商 slug -> sdk_type 映射(Provider 未配 sdk_type 时的回退推断)
_AI_SDK_MAP = {
"openai": "openai-api",
"deepseek": "deepseek-api",
"kimi": "moonshot-api",
"moonshot": "moonshot-api",
"minimax": "minimax-api",
}
class BalanceUnsupported(HTTPException):
"""该平台没有可用的余额查询接口
与「查询失败」区分开:sync_all_ai_balances 把它计入 skipped 而不是 failed,
否则每次「同步 AI 余额」都会把这类平台报成错误。
"""
def __init__(self, sdk_type: str):
super().__init__(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"{sdk_type} 暂不支持余额查询",
)
def refresh_ai_balance(session: Session, asset_id: int) -> dict:
"""刷新 AI 账号余额:解密 api_key → 适配器 get_account → 更新余额"""
asset = session.get(Asset, asset_id)
if not asset or asset.asset_type != AssetType.AI_AGENT:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST, detail="非 AI 账号资产"
)
ai = session.exec(select(AIAccount).where(AIAccount.asset_id == asset_id)).first()
if not ai:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND, detail="AI 账号详情不存在"
)
sdk_type = None
if asset.provider_id:
provider = session.get(Provider, asset.provider_id)
if provider:
sdk_type = provider.sdk_type
if not sdk_type:
sdk_type = _AI_SDK_MAP.get((ai.provider or "").lower())
if not sdk_type or not registry.is_supported(sdk_type):
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"无法确定 AI 适配器({ai.provider})",
)
api_key = crypto.decrypt(ai.api_key_encrypted)
if not api_key:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST, detail="未配置 API Key"
)
adapter = registry.get_adapter(sdk_type, {"api_key": api_key})
if not adapter.capabilities().get("get_account"):
raise BalanceUnsupported(sdk_type)
try:
acc = adapter.get_account()
except NotImplementedError as e:
raise BalanceUnsupported(sdk_type) from e
except HTTPException:
raise
except Exception as e: # noqa: BLE001
raise HTTPException(
status_code=status.HTTP_502_BAD_GATEWAY, detail=f"查询余额失败:{e}"
) from e
if acc.balance is not None:
ai.balance = acc.balance
ai.currency = acc.currency
ai.last_synced_at = utcnow()
session.add(ai)
session.commit()
session.refresh(ai)
return {
"balance": ai.balance,
"currency": ai.currency,
"last_synced_at": ai.last_synced_at.isoformat(),
}
def sync_all_ai_balances(session: Session, max_workers: int = 5) -> dict:
"""并发刷新所有 AI 账号余额(单个失败不中断整体)
每个账号需调用外部 API(单次最长 30s),串行时总耗时随账号数线性增长;
改为线程池并发后显著提速。注意:SQLite Session 不能跨线程共享,
每个 worker 使用独立 Session(WAL 模式下多连接读写安全)。
"""
from concurrent.futures import ThreadPoolExecutor
from app.database import assets_engine
ai_assets = session.exec(
select(Asset).where(Asset.asset_type == AssetType.AI_AGENT)
).all()
asset_ids = [(a.id, a.name) for a in ai_assets]
def _refresh_one(asset_id: int):
with Session(assets_engine) as worker_session:
refresh_ai_balance(worker_session, asset_id)
success = 0
failed = 0
skipped = 0
errors = []
with ThreadPoolExecutor(max_workers=max_workers) as pool:
futures = {pool.submit(_refresh_one, aid): name for aid, name in asset_ids}
for future in futures:
name = futures[future]
try:
future.result()
success += 1
except BalanceUnsupported:
# 平台本身没余额接口,不算失败(否则每次都报一堆红)
skipped += 1
except HTTPException as e:
failed += 1
errors.append(f"{name}: {e.detail}")
except Exception as e: # noqa: BLE001
failed += 1
errors.append(f"{name}: {e}")
return {
"total": len(asset_ids),
"success": success,
"failed": failed,
"skipped": skipped,
"errors": errors,
"synced_at": utcnow().isoformat(),
}