feat(chv): 安全审计修复 High/Medium/Low(堆叠于 #239)

审计报告 ~/workspace/chv-security-audit/REPORT.md(No-Go,
Critical 0 / High 7 / Medium 8 / Low 5)代码层修复,mock
测试覆盖,真机动态验证一律未做、只列清单(见
docs/chv-production-readiness.md §11)。

High:
- H1 egress 代码门:代理地址校验、直连显式允许+审计、
  每 VM 独立 tap/mac/ip 校验、net 字段白名单;宿主防火墙
  列为部署清单,代码门+防火墙缺一不可。
- H2 网关 key 移出 argv:provisioned seed 盘客内文件
  /etc/chv/guest-env(0600)投递,传输层密钥进 argv 直接
  ValueError;keyscan 模式密钥投递 fail-closed。
- H3 任务盘删除:stop/recycle/cleanup 删 disk.raw(保留期
  0)+ 残留盘 GC,删失败审计+脏标记。
- H4 默认值改 provisioned:keyscan 只作显式 opt-in。
- H5 seed.iso 生成即 0600;宿主私钥 seed 盘建好后即删;
  崩溃残留扫描 cleanup_stale_secrets。
- H6 state_dir 默认出 /tmp(~/.local/share/.../chv-vms,
  preflight 保 0700);api-socket 组/其它可写即 fail-fast。
- H7 tap 隔离校验:net.json 登记+stop 释放,共享即拒绝审计。
Medium:M1 shlex 远程转义;M2 清理失败即拒起(脏标记);
M3 配额默认 50GB + 快照数/总量/保留期上限;M4 SHA256 必填
+ BUILDINFO versions 实测记录(事前 lockfile 列清单);
M5 host key/ preflight/失败审计补齐;M6 reuse 须声明单租户
项目;M7 容量计数只认连通 socket;M8 sshd 加固进 build.sh
断言。Low:L1 链接不跟随;L2 路径白名单;L3 失败审计;
L4 双变量警告+字段白名单;L5 预热盘/任务盘对应关系定义。
旧测试:实验室默认值与生产默认值区分标注;行为已变的旧
测试按新语义更新(容量死 socket、快照白名单、磁盘路径
白名单、网关 argv、reuse 租户声明)。
测试:新 test_chv_secfix 39 项;旧 117 项 + 2 项新增断言 =
158 项;worker 相关 129 项;前端 vitest 9 项;cloud 全量
1588 项(skipped 4)。
This commit is contained in:
host committed 2026-10-31 03:02:10 +00:00
1 parent 003ea357f9
commit baf26beb78
16 files changed
+2106 -132

No files matched your search

+65 -13
View File
@@ -150,14 +150,23 @@ class ChvFixCase(unittest.TestCase):
class SeedDiskTest(ChvFixCase):
def test_vm_config_carries_seed_disk_readonly(self):
ex = self.make_exec()
cfg = ex.build_vm_config("/tmp/x/disk.raw",
seed_disk="/tmp/x/seed.iso")
# feat-chv-secfix(L2):磁盘路径须在 state_dir 内,
# 旧 /tmp/x 路径不再合法,改用 state_dir 下路径。
disk = os.path.join(self.state_dir, "x", "disk.raw")
seed = os.path.join(self.state_dir, "x", "seed.iso")
cfg = ex.build_vm_config(disk, seed_disk=seed)
self.assertEqual(len(cfg["disks"]), 2)
self.assertEqual(cfg["disks"][1]["path"], "/tmp/x/seed.iso")
self.assertEqual(cfg["disks"][1]["path"], seed)
self.assertTrue(cfg["disks"][1]["readonly"])
cfg2 = ex.build_vm_config("/tmp/x/disk.raw")
cfg2 = ex.build_vm_config(disk)
self.assertEqual(len(cfg2["disks"]), 1)
def test_vm_config_rejects_disk_outside_state_dir(self):
ex = self.make_exec()
with self.assertRaises(ChvFailure) as ctx:
ex.build_vm_config("/tmp/x/disk.raw")
self.assertEqual(ctx.exception.kind, "disk_path_not_allowed")
def test_provisioned_wake_mounts_seed_disk(self):
ex = self.make_exec(
env_extra={"CHV_HOST_KEY_MODE": "provisioned"})
@@ -208,8 +217,11 @@ class RestoreNoFallbackTest(ChvFixCase):
env_extra={"CHV_HOST_KEY_MODE": "provisioned"})
launches = []
ex.launcher = lambda cmd: launches.append(cmd) or FakeProc(True)
# feat-chv-secfix(L2):恢复源须在白名单目录内,
# 本测试的意图是“起 VMM 前失败”,用 state_dir 下路径。
snap_url = "file://" + os.path.join(self.state_dir, "snap", "x")
with self.assertRaises(ChvFailure) as ctx:
ex.restore_vm({"id": "r1"}, "file:///snap/x")
ex.restore_vm({"id": "r1"}, snap_url)
self.assertEqual(ctx.exception.kind, "host_key_not_pinned")
self.assertEqual(launches, [])
@@ -226,7 +238,9 @@ class ReuseBlockTest(ChvFixCase):
def test_reuse_passes_with_experiment_flag_and_is_audited(self):
ex = self.make_exec(env_extra={
"CHV_DISK_MODE": "reuse",
"CHV_REUSE_SINGLE_TENANT_EXPERIMENT": "1"})
"CHV_REUSE_SINGLE_TENANT_EXPERIMENT": "1",
# feat-chv-secfix(M6):实验放行还须声明单租户项目。
"CHV_SINGLE_TENANT_PROJECT": "p1"})
self.assertTrue(ex.preflight())
out = ex.wake_vm({"id": "reuse1"})
self.assertEqual(out["state"], "running")
@@ -264,12 +278,19 @@ class SshTargetTest(ChvFixCase):
# -- P1-1 网关 extra_env ------------------------------------------------
class GatewayExtraEnvTest(ChvFixCase):
def test_chv_ssh_command_carries_extra_env(self):
# feat-chv-secfix(H2):密钥值绝不再进 argv。
# 非密钥 extra_env 仍走 env 前缀;密钥子集须 seed
# 投递后才 source 客内文件(见 worker 网关测试)。
ex = self.make_exec()
cmd = ex.build_ssh_command({"id": "t"}, ["true"],
extra_env={"SUB2API_API_KEY": "k1"})
extra_env={"SUB2API_BASE_URL":
"http://gw/"})
joined = " ".join(cmd)
self.assertIn("SUB2API_API_KEY=k1", joined)
self.assertIn("SUB2API_BASE_URL=http://gw/", joined)
self.assertIn("HTTPS_PROXY=http://10.0.0.1:8888", joined)
with self.assertRaises(ChvFailure):
ex.build_ssh_command({"id": "t"}, ["true"],
extra_env={"SUB2API_API_KEY": "k1"})
def test_worker_gateway_guard_allows_chv(self):
from cloud.server import init_db
@@ -287,8 +308,20 @@ class GatewayExtraEnvTest(ChvFixCase):
self.assertIsNone(error)
self.assertEqual(result["status"], "completed")
ssh_calls = [c for c in self.calls if c and c[0] == "ssh"]
self.assertTrue(any("SUB2API_API_KEY=gwkey" in " ".join(c)
# feat-chv-secfix(H2):密钥值不得出现在任何 ssh
# argv 里;远程命令改为 source 客内 seed 密钥文件。
self.assertFalse(any("SUB2API_API_KEY=gwkey" in " ".join(c)
for c in ssh_calls))
self.assertTrue(any("/etc/chv/guest-env" in " ".join(c)
for c in ssh_calls))
# 任务结束 stop 后 seed 目录与宿主私钥均已清(H3/H5;
# seed 内容本身的断言见 test_chv_secfix 的直接
# provision_host_key 测试,stop 前 seed 仍在)。
self.assertFalse(os.path.exists(
os.path.join(self.state_dir, "chv-gw1", "seed")))
self.assertFalse(os.path.exists(
os.path.join(self.state_dir, "chv-gw1",
"ssh_host_ed25519_key")))
# -- P1-2 取消/清理/看门狗 ----------------------------------------------
@@ -420,14 +453,30 @@ class PoolAcquireResumeTest(unittest.TestCase):
# -- P1-4 容量持久化计数 -------------------------------------------------
class CapacityPersistentTest(ChvFixCase):
def test_socket_counts_after_restart(self):
# feat-chv-secfix(M7):只有真能连通的 socket 才
# 计数——用真实监听 socket 代表活 VMM。
ex = self.make_exec(env_extra={"CHV_MAX_VMS": "1"})
os.makedirs(self.state_dir, exist_ok=True)
sock_path = os.path.join(self.state_dir, "chv-other.sock")
srv = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
srv.bind(sock_path)
srv.listen(1)
try:
with self.assertRaises(ChvFailure) as ctx:
ex.wake_vm({"id": "new1"})
self.assertEqual(ctx.exception.kind, "host_capacity_exceeded")
finally:
srv.close()
def test_dead_socket_file_does_not_block(self):
# M7 的另一面:空文件/死 socket 不再占满容量。
ex = self.make_exec(env_extra={"CHV_MAX_VMS": "1"})
os.makedirs(self.state_dir, exist_ok=True)
with open(os.path.join(self.state_dir, "chv-other.sock"),
"w") as fh:
fh.write("")
with self.assertRaises(ChvFailure) as ctx:
ex.wake_vm({"id": "new1"})
self.assertEqual(ctx.exception.kind, "host_capacity_exceeded")
out = ex.wake_vm({"id": "new1"})
self.assertEqual(out["state"], "running")
def test_own_socket_does_not_block_rewake(self):
ex = self.make_exec(env_extra={"CHV_MAX_VMS": "1"})
@@ -523,8 +572,11 @@ class ApiTruncationTest(unittest.TestCase):
class SnapshotPersistTest(ChvFixCase):
def test_snapshot_missing_file_does_not_shutdown(self):
ex = self.make_exec()
# feat-chv-secfix(L2):快照须在白名单目录内,
# 用 state_dir 下的不存在路径,仍验 snapshot_not_persisted。
missing = os.path.join(self.state_dir, "missing.snap")
with self.assertRaises(ChvFailure) as ctx:
ex.hibernate_vm({"id": "h1"}, "file:///nonexistent/snap")
ex.hibernate_vm({"id": "h1"}, "file://" + missing)
self.assertEqual(ctx.exception.kind, "snapshot_not_persisted")
kinds = [c[-1] for c in self.calls if c[0] == self.ch_remote]
self.assertIn("pause", kinds)