diff --git a/chv-guest/build.sh b/chv-guest/build.sh index 2896c87..d75db9d 100755 --- a/chv-guest/build.sh +++ b/chv-guest/build.sh @@ -26,6 +26,12 @@ # 失败关闭(不 --force、不自动 rebase)。token 绝不进镜像、 # 代码、日志。 # +# 对齐 Muse P2(2026-11-07):镜像里加 systemd-nspawn 运行时单元 +#(chv-cell:codex 跑在单元内)+ 单元外 broker(chv-broker: +#Unix socket + SO_PEERCRED 白名单代办)+ 浏览器子代理(chv-browser: +#单元外独立进程,只返回无障碍树快照)。boot clone 仍在单元外做; +#单元隔离口径见 docs/chv-muse-p2-gap.md。 +# # 用法(必须在带 /dev/kvm 的宿主上跑全量模式): # CODEX_BIN_DIR=bin UBUNTU_VERSION=24.04 CODEX_VERSION=0.164.0-alpha.9 \ # REPO_COMMIT=$(git rev-parse --short HEAD) \ @@ -166,6 +172,13 @@ if [ ! -f "$SCRIPT_DIR/chv-sync.sh" ]; then echo "缺少 chv-sync.sh(Gitea 双向同步脚本):$SCRIPT_DIR" >&2 exit 1 fi +# 对齐 Muse P2:单元启动器 / broker / 浏览器子代理三脚本缺一即失败。 +for p2_script in chv-cell.py chv-broker.py chv-browser.py; do + if [ ! -f "$SCRIPT_DIR/$p2_script" ]; then + echo "缺少 $p2_script(Muse P2 客机脚本):$SCRIPT_DIR" >&2 + exit 1 + fi +done if [ "$DRY_RUN" = "0" ]; then if [ ! -e /dev/kvm ]; then echo "CHV_GUEST_BUILD needs a KVM host: /dev/kvm 不存在" >&2 @@ -249,7 +262,7 @@ run qemu-img convert -f qcow2 -O raw "$IMG" "$OUT/disk.raw" # 写进 /app/guest-versions-chromium.txt)。 GT_INSTALL_CMD="curl -fsSL -o /tmp/gt-install.sh https://gitea.com/gitea/cli/raw/branch/main/scripts/install.sh && echo \"${GT_INSTALL_SH_SHA256} /tmp/gt-install.sh\" | sha256sum -c - && GITEA_CLI_VERSION=${GT_VERSION} GITEA_CLI_INSTALL_DIR=/usr/local/bin sh /tmp/gt-install.sh && rm -f /tmp/gt-install.sh && /usr/local/bin/gt version" VC_ARGS=(-a "$OUT/disk.raw" - --run-command 'apt-get update -qq && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends git openssh-server openssh-client cloud-init python3 python3-pip curl ca-certificates xz-utils fonts-wqy-microhei fontconfig && apt-get clean' + --run-command 'apt-get update -qq && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends git openssh-server openssh-client cloud-init python3 python3-pip curl ca-certificates xz-utils fonts-wqy-microhei fontconfig systemd-container iproute2 iptables && apt-get clean' --copy-in "$CODEX_BIN_DIR/codex:/usr/local/bin/" --copy-in "$CODEX_BIN_DIR/codex-code-mode-host:/usr/local/bin/" --copy-in "$CODEX_BIN_DIR/codex-resources:/usr/local/bin/" @@ -258,13 +271,32 @@ VC_ARGS=(-a "$OUT/disk.raw" --copy-in "$REPO_DIR/host.py:/app/" --copy-in "$SCRIPT_DIR/chv-sync.sh:/usr/local/bin/" --run-command 'mv /usr/local/bin/chv-sync.sh /usr/local/bin/chv-sync && chmod +x /usr/local/bin/chv-sync' + --copy-in "$SCRIPT_DIR/chv-cell.py:/usr/local/bin/" + --copy-in "$SCRIPT_DIR/chv-broker.py:/usr/local/bin/" + --copy-in "$SCRIPT_DIR/chv-browser.py:/usr/local/bin/" + --run-command 'mv /usr/local/bin/chv-cell.py /usr/local/bin/chv-cell && mv /usr/local/bin/chv-broker.py /usr/local/bin/chv-broker && mv /usr/local/bin/chv-browser.py /usr/local/bin/chv-browser && chmod +x /usr/local/bin/chv-cell /usr/local/bin/chv-broker /usr/local/bin/chv-browser' --run-command 'printf "runcmd:\n - systemctl restart ssh || service ssh restart || true\n - /usr/local/bin/chv-sync boot >>/var/log/chv-sync.log 2>&1 || true\n" > /etc/cloud/cloud.cfg.d/99-chv-sync.cfg' --run-command 'git config --system user.name chv && git config --system user.email chv@localhost && git config --system credential.helper store && git config --system init.defaultBranch main' --run-command "$GT_INSTALL_CMD" --run-command '/usr/local/bin/gt version > /app/guest-versions-gt.txt' --run-command 'printf "ssh_deletekeys: false\nbootcmd:\n - \"[ -f /etc/ssh/ssh_host_ed25519_key ] || ssh-keygen -A\"\n" > /etc/cloud/cloud.cfg.d/99-chv-hostkey.cfg; systemctl enable ssh || true' --run-command 'printf "PasswordAuthentication no\nPermitRootLogin prohibit-password\nChallengeResponseAuthentication no\nKbdInteractiveAuthentication no\nPubkeyAuthentication yes\n" > /etc/ssh/sshd_config.d/99-chv.conf && mkdir -p /run/sshd && ssh-keygen -A && sshd -t && grep -q "^PasswordAuthentication no" /etc/ssh/sshd_config.d/99-chv.conf && grep -q "^PermitRootLogin prohibit-password" /etc/ssh/sshd_config.d/99-chv.conf && rm -f /etc/ssh/ssh_host_* && test -z "$(ls /etc/ssh/ssh_host_* 2>/dev/null)"' - --run-command 'dpkg-query -W -f "\${Package} \${Version}\n" > /app/guest-versions-dpkg.txt; pip3 freeze > /app/guest-versions-pip.txt || true') + --run-command 'dpkg-query -W -f "\${Package} \${Version}\n" > /app/guest-versions-dpkg.txt; pip3 freeze > /app/guest-versions-pip.txt || true' + --run-command 'useradd --system --home-dir /var/lib/chv-browser --shell /usr/sbin/nologin chv-browser && mkdir -p /var/lib/chv-browser /run/chv/shots && chown chv-browser:chv-browser /var/lib/chv-browser /run/chv/shots && chmod 0755 /run/chv /run/chv/shots' + --run-command 'printf "[Unit]\nDescription=CHV cell broker (Muse P2)\nAfter=network.target\n\n[Service]\nType=simple\nExecStart=/usr/bin/python3 /usr/local/bin/chv-broker\nRestart=on-failure\nRestartSec=2\n\n[Install]\nWantedBy=multi-user.target\n" > /etc/systemd/system/chv-broker.service' + --run-command 'printf "[Unit]\nDescription=CHV browser sub-agent (Muse P2)\nAfter=network.target\n\n[Service]\nType=simple\nUser=chv-browser\nGroup=chv-browser\nExecStart=/usr/bin/python3 /usr/local/bin/chv-browser\nRestart=on-failure\nRestartSec=2\n\n[Install]\nWantedBy=multi-user.target\n" > /etc/systemd/system/chv-browser.service' + --run-command 'printf "d /run/chv 0755 root root -\nd /run/chv/shots 0755 chv-browser chv-browser -\n" > /etc/tmpfiles.d/chv.conf && systemctl enable chv-broker chv-browser || true') +# P1 交接(Muse P1 的 vsock_notify.py):文件在仓库里存在时拷入 +# 镜像(P1 分支才有,本分支尚无);不存在时只警告、不失败—— +# 合并后自动生效,P2 不改该脚本内容。 +VSOCK_NOTIFY_PRESENT=0 +if [ -f "$REPO_DIR/cloud/chv/guest/vsock_notify.py" ]; then + VC_ARGS+=(--copy-in "$REPO_DIR/cloud/chv/guest/vsock_notify.py:/usr/local/bin/" + --run-command 'mv /usr/local/bin/vsock_notify.py /usr/local/bin/vsock-notify && chmod +x /usr/local/bin/vsock-notify') + VSOCK_NOTIFY_PRESENT=1 +else + echo "警告:cloud/chv/guest/vsock_notify.py 不在仓库里(P1 未合并),镜像暂不含 vsock notify" >&2 +fi if [ "${PLAYWRIGHT:-1}" = "1" ]; then VC_ARGS+=( --run-command '(pip3 install --no-cache-dir --break-system-packages playwright || pip3 install --no-cache-dir playwright) && PLAYWRIGHT_BROWSERS_PATH=/usr/local/share/ms-playwright python3 -m playwright install --with-deps chromium-headless-shell' @@ -316,14 +348,16 @@ fi # 产物校验(干跑只打印,不执行校验命令的副作用部分)。 if [ "$DRY_RUN" = "1" ]; then echo "DRY-RUN: test -s $OUT/vmlinux && test -s $OUT/disk.raw" - echo "DRY-RUN: python3 $SCRIPT_DIR/buildinfo.py $OUT $TAG $UBUNTU_VERSION $CODEX_VERSION $REPO_COMMIT ${PLAYWRIGHT:-1} ${CHROMIUM_HEADLESS_SHELL_REV:-} # 生成 $OUT/BUILDINFO.json 并校验 tag=$TAG" + echo "DRY-RUN: python3 $SCRIPT_DIR/buildinfo.py $OUT $TAG $UBUNTU_VERSION $CODEX_VERSION $REPO_COMMIT ${PLAYWRIGHT:-1} ${CHROMIUM_HEADLESS_SHELL_REV:-} [vsock_notify?] # 生成 $OUT/BUILDINFO.json 并校验 tag=$TAG" else test -s "$OUT/vmlinux" || { echo "内核产物为空:$OUT/vmlinux" >&2; exit 1; } test -s "$OUT/disk.raw" || { echo "磁盘产物为空:$OUT/disk.raw" >&2; exit 1; } if [ "${PLAYWRIGHT:-1}" = "1" ] && [ -z "$CHROMIUM_HEADLESS_SHELL_REV" ]; then echo "警告:CHROMIUM_HEADLESS_SHELL_REV 未设,Chromium 版本只记录、不钉死(V-9 验收须设该变量重建)" >&2 fi - python3 "$SCRIPT_DIR/buildinfo.py" "$OUT" "$TAG" "$UBUNTU_VERSION" "$CODEX_VERSION" "$REPO_COMMIT" "$UBUNTU_IMAGE_SHA256" "${PLAYWRIGHT:-1}" "${CHROMIUM_HEADLESS_SHELL_REV:-}" + EXTRA_CONTENTS="" + [ "$VSOCK_NOTIFY_PRESENT" = "1" ] && EXTRA_CONTENTS="vsock_notify" + python3 "$SCRIPT_DIR/buildinfo.py" "$OUT" "$TAG" "$UBUNTU_VERSION" "$CODEX_VERSION" "$REPO_COMMIT" "$UBUNTU_IMAGE_SHA256" "${PLAYWRIGHT:-1}" "${CHROMIUM_HEADLESS_SHELL_REV:-}" "$EXTRA_CONTENTS" fi echo "tag=$TAG" echo "out=$OUT" diff --git a/chv-guest/buildinfo.py b/chv-guest/buildinfo.py index 112b16c..f3af51d 100644 --- a/chv-guest/buildinfo.py +++ b/chv-guest/buildinfo.py @@ -37,6 +37,8 @@ BASE_CONTENTS = [ "xz-utils", "fonts-wqy-microhei", "fontconfig", "codex", "codex-code-mode-host", "codex-resources/bwrap", "/app/host.py", "gt", "chv-sync", + # 对齐 Muse P2:nspawn 单元启动器 + broker + 浏览器子代理。 + "systemd-container", "chv-cell", "chv-broker", "chv-browser", ] # 执行器默认内核 cmdline(config.py),构建产物须与它一起记录(B-12), @@ -67,12 +69,18 @@ def _sha256_size(path): def build(outdir, tag, ubuntu, codex_version, repo_commit, - ubuntu_image_sha256, playwright, chromium_pinned_rev=None): + ubuntu_image_sha256, playwright, chromium_pinned_rev=None, + extra_contents=None): disk_path = os.path.join(outdir, "disk.raw") vmlinux_path = os.path.join(outdir, "vmlinux") disk_sha, disk_size = _sha256_size(disk_path) vmlinux_sha, vmlinux_size = _sha256_size(vmlinux_path) contents = list(BASE_CONTENTS) + # extra_contents:构建时才确定的附加内容(如 P1 的 vsock_notify, + # 文件存在才拷入)。调用方传什么记什么,不编默认值。 + for item in extra_contents or []: + if item and item not in contents: + contents.append(item) versions = { "dpkg": _read_lines( os.path.join(outdir, "guest-versions-dpkg.txt")), @@ -123,14 +131,18 @@ def build(outdir, tag, ubuntu, codex_version, repo_commit, def main(argv): - if len(argv) not in (8, 9): + if len(argv) not in (8, 9, 10): print("用法:buildinfo.py OUT TAG UBUNTU CODEX_VERSION " "REPO_COMMIT UBUNTU_IMAGE_SHA256 PLAYWRIGHT " - "[CHROMIUM_HEADLESS_SHELL_REV]", file=sys.stderr) + "[CHROMIUM_HEADLESS_SHELL_REV] [EXTRA_CONTENTS]", file=sys.stderr) return 2 _, outdir, tag, ubuntu, codex_version, repo_commit, \ ubuntu_image_sha256, playwright = argv[:8] - chromium_pinned_rev = argv[8] if len(argv) == 9 else None + chromium_pinned_rev = argv[8] if len(argv) >= 9 else None + # 第 10 参数:逗号分隔的附加内容(如 vsock_notify),可空。 + extra_contents = None + if len(argv) == 10 and argv[9]: + extra_contents = [s for s in argv[9].split(",") if s] if chromium_pinned_rev and not re.fullmatch(r"[0-9]+", chromium_pinned_rev): print("CHROMIUM_HEADLESS_SHELL_REV 须为纯数字:%s" @@ -138,7 +150,7 @@ def main(argv): return 2 info = build(outdir, tag, ubuntu, codex_version, repo_commit, ubuntu_image_sha256, playwright == "1", - chromium_pinned_rev) + chromium_pinned_rev, extra_contents) if not info["versions"]["dpkg"]: print("BUILDINFO versions.dpkg 为空", file=sys.stderr) return 1 diff --git a/chv-guest/chv-broker.py b/chv-guest/chv-broker.py new file mode 100644 index 0000000..9a82df2 --- /dev/null +++ b/chv-guest/chv-broker.py @@ -0,0 +1,267 @@ +#!/usr/bin/env python3 +"""chv-broker:单元外 IPC broker(客机侧、单元外服务)。 + +对齐 Muse P2(见 docs/chv-muse-p2-gap.md):单元内外走 Unix socket + +SO_PEERCRED 的最小权限 IPC。broker 只认白名单操作,凭证与浏览器 +都在单元外——单元看不见 token,只能请 broker 代办。 + +校验(双重,缺一不可才放行;cell.pid 不存在时只做 uid 检查并打日志): + 1. SO_PEERCRED 的 uid 须在单元映射范围 [1000, 1000+65536) 内; + 2. 对端 pid 的 pid ns 须与 /run/chv/cell.pid 记的单元 init 同 ns + (读不到 cell.pid 时跳过第 2 项并警告;cell.pid 存在但对端 ns + 读不到时拒绝,不降级)。 + +操作白名单: + ping -> {"ok": true} + sync {action: pull|push|status, repo?} -> 单元外以 chv 身份跑 + chv-sync(list argv,不进 shell;仓目录只允许 + ~/workspace 下,realpath + commonpath 校验)。 + browser {…} -> 原样转发给 chv-browser.sock(浏览器子代理), + 回包原样返回;浏览器服务不在时报错,不替它跑 playwright。 + +只用标准库。单连接一请求一响应(JSON 行,≤64KB);异常只断本 +连接,绝不杀死 broker 循环。 +""" +import grp +import json +import os +import socket +import struct +import subprocess +import sys + +BROKER_SOCK = "/run/chv/broker.sock" +BROWSER_SOCK = "/run/chv/browser.sock" +CELL_PID_FILE = "/run/chv/cell.pid" +UID_BASE = 1000 +UID_COUNT = 65536 +WORKSPACE = "/home/chv/workspace" +SYNC_USER = "chv" +SYNC_BIN = "/usr/local/bin/chv-sync" +MAX_REQUEST_BYTES = 65536 +MAX_OUTPUT_BYTES = 8192 + +SYNC_ACTIONS = ("pull", "push", "status") + + +def _read_ns_link(pid, proc_root="/proc"): + """读 /proc//ns/pid 的链接目标;读不到返回 None。""" + try: + return os.readlink(os.path.join(proc_root, str(pid), "ns", "pid")) + except OSError: + return None + + +def peer_allowed(peer_uid, peer_pid, cell_pid=None, proc_root="/proc"): + """对端是否允许用 broker。返回 (allowed, reason)。 + cell_pid 为 None 时只做 uid 检查(broker 打日志警告,调用方负责)。""" + if not (UID_BASE <= peer_uid < UID_BASE + UID_COUNT): + return False, "uid %d 不在单元映射范围 [%d, %d)" % ( + peer_uid, UID_BASE, UID_BASE + UID_COUNT) + if cell_pid is None: + return True, "uid 在范围内(cell.pid 缺失,未做 pid ns 对照)" + cell_ns = _read_ns_link(cell_pid, proc_root) + if cell_ns is None: + return False, "cell.pid 的 pid ns 读不到,拒绝(不降级)" + peer_ns = _read_ns_link(peer_pid, proc_root) + if peer_ns is None: + return False, "对端 pid ns 读不到,拒绝" + if peer_ns != cell_ns: + return False, "对端 pid ns 与单元不一致" + return True, "ok" + + +def _read_cell_pid(): + try: + with open(CELL_PID_FILE) as fh: + return int(fh.read().strip()) + except (OSError, ValueError): + return None + + +def _resolve_repo(repo): + """仓目录只允许 ~/workspace 下。repo 为空时返回 None(由 + chv-sync 自己取唯一仓)。返回解析后的绝对路径或抛 ValueError。""" + if not repo: + return None + base = os.path.realpath(WORKSPACE) + target = os.path.realpath(repo) + if os.path.commonpath([base, target]) != base or target == base: + raise ValueError("仓目录只允许 %s 下:%s" % (WORKSPACE, repo)) + return target + + +def _truncate(text, limit=MAX_OUTPUT_BYTES): + if text is None: + return "" + if len(text) > limit: + return text[:limit] + "\n…(已截断)" + return text + + +def _run_sync(action, repo): + if action not in SYNC_ACTIONS: + return {"ok": False, + "error": "sync action 只允许 %s:%s" + % ("/".join(SYNC_ACTIONS), action)} + try: + repo_path = _resolve_repo(repo) + except ValueError as e: + return {"ok": False, "error": str(e)} + cmd = ["runuser", "-u", SYNC_USER, "--", "env", "HOME=/home/chv", + SYNC_BIN, action] + if repo_path: + cmd.append(repo_path) + try: + proc = subprocess.run(cmd, capture_output=True, text=True, + timeout=300) + except subprocess.TimeoutExpired: + return {"ok": False, "error": "chv-sync 超时(300 秒)"} + except OSError as e: + return {"ok": False, "error": "chv-sync 起不来:%s" % e} + return {"ok": proc.returncode == 0, + "stdout": _truncate(proc.stdout), + "stderr": _truncate(proc.stderr), + "returncode": proc.returncode} + + +def _forward_browser(req, sock_path=BROWSER_SOCK, timeout=120): + """转发给浏览器子代理;回包须是带 ok 的 dict,否则报错。""" + try: + conn = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM) + conn.settimeout(timeout) + conn.connect(sock_path) + except OSError as e: + return {"ok": False, + "error": "浏览器子代理未就绪:%s" % e} + try: + conn.sendall((json.dumps(req, ensure_ascii=False) + "\n").encode()) + data = b"" + while len(data) < MAX_REQUEST_BYTES: + chunk = conn.recv(4096) + if not chunk: + break + data += chunk + if data.endswith(b"\n"): + break + except OSError as e: + return {"ok": False, "error": "浏览器子代理通信失败:%s" % e} + finally: + conn.close() + try: + resp = json.loads(data.decode()) + except (ValueError, UnicodeDecodeError): + return {"ok": False, "error": "浏览器子代理回包不是 JSON"} + if not isinstance(resp, dict) or "ok" not in resp: + return {"ok": False, "error": "浏览器子代理回包格式不对"} + return resp + + +def handle_request(req, browser_sock=BROWSER_SOCK): + """处理一个请求 dict,返回响应 dict(纯逻辑,可单测)。""" + if not isinstance(req, dict): + return {"ok": False, "error": "请求须是 JSON 对象"} + op = req.get("op") + if op == "ping": + return {"ok": True, "cell": True} + if op == "sync": + return _run_sync(req.get("action"), req.get("repo")) + if op == "browser": + return _forward_browser(req, sock_path=browser_sock) + return {"ok": False, "error": "未知操作:%r(白名单:ping/sync/browser)" + % (op,)} + + +def _peer_creds(conn): + """SO_PEERCRED:返回 (pid, uid, gid)。Linux struct ucred = 3 个 int。""" + raw = conn.getsockopt(socket.SOL_SOCKET, socket.SO_PEERCRED, + struct.calcsize("3i")) + pid, uid, gid = struct.unpack("3i", raw) + return pid, uid, gid + + +def serve(sock_path=BROKER_SOCK, browser_sock=BROWSER_SOCK): + """监听 Unix socket,逐连接处理。返回码 0 只能是被信号杀死; + socket 建不起来即抛错(systemd 会重启服务,不要静默带病跑)。""" + if os.path.exists(sock_path): + os.remove(sock_path) + os.makedirs(os.path.dirname(sock_path), exist_ok=True) + srv = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM) + srv.bind(sock_path) + srv.listen(16) + # 0660 + 组 chv:单元内 root 在客机侧是 uid 1000(chv), + # 组 chv 可连;其它用户连不上(再叠加 SO_PEERCRED 校验)。 + os.chmod(sock_path, 0o660) + try: + gid = grp.getgrnam("chv").gr_gid + except KeyError: + print("chv-broker: 组 chv 不存在,socket 权限置不全,拒绝启动", + file=sys.stderr) + raise SystemExit(1) + os.chown(sock_path, -1, gid) + print("chv-broker: 监听 %s" % sock_path, file=sys.stderr, flush=True) + while True: + try: + conn, _ = srv.accept() + except OSError: + continue + try: + pid, uid, _gid = _peer_creds(conn) + cell_pid = _read_cell_pid() + allowed, reason = peer_allowed(uid, pid, cell_pid) + if not allowed: + print("chv-broker: 拒绝连接 pid=%d uid=%d:%s" + % (pid, uid, reason), file=sys.stderr, flush=True) + conn.close() + continue + if cell_pid is None: + print("chv-broker: 警告:cell.pid 缺失,只做了 uid 检查 " + "(pid=%d uid=%d)" % (pid, uid), file=sys.stderr, + flush=True) + conn.settimeout(310) + data = b"" + while len(data) < MAX_REQUEST_BYTES: + chunk = conn.recv(4096) + if not chunk: + break + data += chunk + if data.endswith(b"\n"): + break + try: + req = json.loads(data.decode()) + except (ValueError, UnicodeDecodeError): + resp = {"ok": False, "error": "请求不是合法 JSON"} + else: + resp = handle_request(req, browser_sock=browser_sock) + conn.sendall((json.dumps(resp, ensure_ascii=False) + + "\n").encode()) + except OSError as e: + print("chv-broker: 连接处理异常:%s" % e, file=sys.stderr, + flush=True) + finally: + try: + conn.close() + except OSError: + pass + + +def main(argv): + sock_path = BROKER_SOCK + browser_sock = BROWSER_SOCK + i = 0 + while i < len(argv): + if argv[i] == "--socket" and i + 1 < len(argv): + sock_path = argv[i + 1] + i += 2 + elif argv[i] == "--browser-socket" and i + 1 < len(argv): + browser_sock = argv[i + 1] + i += 2 + else: + print("chv-broker: 未知参数:%s" % argv[i], file=sys.stderr) + return 2 + serve(sock_path, browser_sock) + return 0 + + +if __name__ == "__main__": + raise SystemExit(main(sys.argv[1:])) diff --git a/chv-guest/chv-browser.py b/chv-guest/chv-browser.py new file mode 100644 index 0000000..743feb9 --- /dev/null +++ b/chv-guest/chv-browser.py @@ -0,0 +1,301 @@ +#!/usr/bin/env python3 +"""chv-browser:浏览器子代理(客机侧、单元外独立进程)。 + +对齐 Muse P2(见 docs/chv-muse-p2-gap.md):浏览器任务由专门子代理 +完成,与 codex 单元分离。子代理只向单元返回页面无障碍树快照 +(Playwright accessibility.snapshot),不见原始 DOM,不能在页面 +上下文跑 JS,不暴露 evaluate/content,不开 DevTools。 + +操作(Unix socket /run/chv/browser.sock,JSON 行): + open {url} -> 导航并返回 {title, url, a11y, truncated} + snapshot {} -> 当前页的 {title, url, a11y, truncated} + screenshot {} -> 截图落盘 /run/chv/shots/.png,返回 {path} + close {} -> 关当前页(浏览器进程继续跑,等下一次 open) + ping {} -> {"ok": true} + +URL 只允许 http/https;localhost 与私有地址字面量直接拒绝(主机名 +判定,不做 DNS 解析——DNS 重绑定防护是 P3 Sentinel 的事,这里不 +冒充已做)。无障碍树截断到 200 节点 / 64KB,超了标 truncated=true, +绝不把整棵树无上限灌进 socket。 + +playwright 是懒加载:import 本模块不需要 playwright 已装(单测 +在无 playwright 环境跑);真正 open 时才加载,缺失即报错。 + +只用标准库 + playwright(运行时)。运行身份:系统用户 chv-browser, +socket 0600(broker 以 root 连,不受模式限制)。 +""" +import json +import os +import re +import socket +import sys +import uuid + +BROWSER_SOCK = "/run/chv/browser.sock" +SHOTS_DIR = "/run/chv/shots" +PLAYWRIGHT_BROWSERS_PATH = "/usr/local/share/ms-playwright" +MAX_REQUEST_BYTES = 65536 +MAX_A11Y_NODES = 200 +MAX_A11Y_CHARS = 65536 +NAV_TIMEOUT_MS = 30000 + +# 拒绝的 URL 主机:localhost 与私有地址字面量(主机名判定)。 +_PRIVATE_HOST_RE = re.compile( + r"^(localhost|127\.\d+\.\d+\.\d+|10\.\d+\.\d+\.\d+|" + r"172\.(1[6-9]|2\d|3[01])\.\d+\.\d+|192\.168\.\d+\.\d+|" + r"\[::1\]|\[fe8[0-9a-f]*:.*\])$", + re.IGNORECASE) + + +def validate_url(url): + """URL 校验:只允许 http/https,且主机不是 localhost/私有地址 + 字面量。返回 (ok, error)。DNS 重绑定不在本层防(见模块 docstring)。""" + if not isinstance(url, str) or not url: + return False, "url 不能为空" + m = re.match(r"^(https?)://([^/]+)", url, re.IGNORECASE) + if not m: + return False, "只允许 http/https URL:%s" % url[:80] + host = m.group(2).split("@")[-1].split(":")[0].lower() + if _PRIVATE_HOST_RE.match(host): + return False, "拒绝 localhost/私有地址:%s" % host + return True, "" + + +def truncate_a11y(node, max_nodes=MAX_A11Y_NODES, + max_chars=MAX_A11Y_CHARS): + """截断无障碍树:节点数超上限时剪掉多余子树。返回 (tree, truncated)。 + 纯函数,可单测。节点计数按 dict 节点数(含 children 展开)。""" + state = {"count": 0, "truncated": False} + + def walk(n): + if not isinstance(n, dict): + return n + state["count"] += 1 + if state["count"] > max_nodes: + state["truncated"] = True + return None + out = {} + for key, value in n.items(): + if key == "children" and isinstance(value, list): + children = [] + for child in value: + walked = walk(child) + if walked is None: + state["truncated"] = True + break + children.append(walked) + out[key] = children + else: + out[key] = value + return out + + tree = walk(node) + text = json.dumps(tree, ensure_ascii=False) + if len(text) > max_chars: + state["truncated"] = True + # 字符超限时只留顶层角色/名字/URL 信息,不灌整棵树。 + tree = {"role": (node or {}).get("role"), + "name": (node or {}).get("name"), + "truncated_chars": True} + return tree, state["truncated"] + + +def _load_playwright(): + """懒加载 playwright;缺失时报明错(不静默假装有浏览器)。""" + os.environ.setdefault("PLAYWRIGHT_BROWSERS_PATH", + PLAYWRIGHT_BROWSERS_PATH) + try: + from playwright.sync_api import sync_playwright + except ImportError as e: + raise RuntimeError( + "chv-browser: playwright 未装(%s),浏览器子代理不可用" + % e) + return sync_playwright + + +class BrowserAgent: + """单页浏览器子代理。playwright_factory 供单测注入假实现; + 正常运行时用 _load_playwright()。""" + + def __init__(self, playwright_factory=None, shots_dir=SHOTS_DIR): + self._factory = playwright_factory or _load_playwright + self._shots_dir = shots_dir + self._pw = None + self._browser = None + self._page = None + + def _ensure_browser(self): + if self._browser is not None: + return + self._pw = self._factory() + # sync_playwright() 返回上下文管理器;__enter__ 后拿 chromium。 + entered = self._pw.__enter__() + self._browser = entered.chromium.launch(headless=True) + + def open(self, url): + ok, err = validate_url(url) + if not ok: + return {"ok": False, "error": err} + try: + self._ensure_browser() + if self._page is not None: + try: + self._page.close() + except Exception: + pass + self._page = self._browser.new_page() + self._page.goto(url, timeout=NAV_TIMEOUT_MS) + self._page.wait_for_load_state("domcontentloaded", + timeout=NAV_TIMEOUT_MS) + except Exception as e: + return {"ok": False, + "error": "打开 %s 失败:%s" % (url[:80], e)} + return self.snapshot() + + def snapshot(self): + if self._page is None: + return {"ok": False, "error": "还没有打开的页面(先 open)"} + try: + title = self._page.title() + url = self._page.url + tree = self._page.accessibility.snapshot() + except Exception as e: + return {"ok": False, "error": "取快照失败:%s" % e} + tree, truncated = truncate_a11y(tree or {}) + return {"ok": True, "title": title, "url": url, + "a11y": tree, "truncated": truncated} + + def screenshot(self): + if self._page is None: + return {"ok": False, "error": "还没有打开的页面(先 open)"} + try: + os.makedirs(self._shots_dir, exist_ok=True) + path = os.path.join(self._shots_dir, "%s.png" + % uuid.uuid4().hex) + self._page.screenshot(path=path) + except Exception as e: + return {"ok": False, "error": "截图失败:%s" % e} + return {"ok": True, "path": path} + + def close(self): + if self._page is not None: + try: + self._page.close() + except Exception: + pass + self._page = None + return {"ok": True, "closed": True} + + def shutdown(self): + try: + if self._browser is not None: + self._browser.close() + except Exception: + pass + try: + if self._pw is not None: + self._pw.__exit__(None, None, None) + except Exception: + pass + + +# 本层不暴露的操作:evaluate / content(原始 DOM)/ DevTools。 +# 收到即拒绝,不转发、不执行。 +FORBIDDEN_OPS = ("evaluate", "content", "devtools", "js") + + +def handle_request(agent, req): + """处理一个请求 dict,返回响应 dict(纯逻辑,可单测)。""" + if not isinstance(req, dict): + return {"ok": False, "error": "请求须是 JSON 对象"} + op = req.get("op") + if op in FORBIDDEN_OPS: + return {"ok": False, + "error": "操作 %r 不暴露(无 JS/无原始 DOM)" % op} + if op == "ping": + return {"ok": True, "browser": True} + if op == "open": + return agent.open(req.get("url")) + if op == "snapshot": + return agent.snapshot() + if op == "screenshot": + return agent.screenshot() + if op == "close": + return agent.close() + return {"ok": False, + "error": "未知操作:%r(白名单:open/snapshot/screenshot/" + "close/ping)" % (op,)} + + +def serve(sock_path=BROWSER_SOCK, shots_dir=SHOTS_DIR): + """监听 Unix socket,逐连接处理。socket 0600(只给属主 + + broker 的 root 连)。建不起来即抛错,不要静默带病跑。""" + if os.path.exists(sock_path): + os.remove(sock_path) + os.makedirs(os.path.dirname(sock_path), exist_ok=True) + srv = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM) + srv.bind(sock_path) + srv.listen(16) + os.chmod(sock_path, 0o600) + print("chv-browser: 监听 %s" % sock_path, file=sys.stderr, flush=True) + agent = BrowserAgent(shots_dir=shots_dir) + try: + while True: + try: + conn, _ = srv.accept() + except OSError: + continue + try: + conn.settimeout(120) + data = b"" + while len(data) < MAX_REQUEST_BYTES: + chunk = conn.recv(4096) + if not chunk: + break + data += chunk + if data.endswith(b"\n"): + break + try: + req = json.loads(data.decode()) + except (ValueError, UnicodeDecodeError): + resp = {"ok": False, "error": "请求不是合法 JSON"} + else: + try: + resp = handle_request(agent, req) + except Exception as e: + resp = {"ok": False, + "error": "浏览器子代理内部错误:%s" % e} + conn.sendall((json.dumps(resp, ensure_ascii=False) + + "\n").encode()) + except OSError as e: + print("chv-browser: 连接处理异常:%s" % e, file=sys.stderr, + flush=True) + finally: + try: + conn.close() + except OSError: + pass + finally: + agent.shutdown() + + +def main(argv): + sock_path = BROWSER_SOCK + shots_dir = SHOTS_DIR + i = 0 + while i < len(argv): + if argv[i] == "--socket" and i + 1 < len(argv): + sock_path = argv[i + 1] + i += 2 + elif argv[i] == "--shots-dir" and i + 1 < len(argv): + shots_dir = argv[i + 1] + i += 2 + else: + print("chv-browser: 未知参数:%s" % argv[i], file=sys.stderr) + return 2 + serve(sock_path, shots_dir) + return 0 + + +if __name__ == "__main__": + raise SystemExit(main(sys.argv[1:])) diff --git a/chv-guest/chv-cell.py b/chv-guest/chv-cell.py new file mode 100644 index 0000000..dfdd654 --- /dev/null +++ b/chv-guest/chv-cell.py @@ -0,0 +1,409 @@ +#!/usr/bin/env python3 +"""chv-cell:客机内 systemd-nspawn 运行时单元(Hatch 式)启动器。 + +对齐 Muse P2(见 docs/chv-muse-p2-gap.md):codex 跑在单元内,不再 +直接跑在客机用户态。单元是单命令容器(--as-pid2,不 boot 第二个 +systemd),单元即沙箱——单元内 codex 用 danger-full-access,不再 +套 bwrap(bwrap 的 --proc 挂载在嵌套命名空间里走不通,见 AGENTS.md +内层 proc 定位)。 + +关键映射(与 Muse“单元 root 映射为 host 侧非特权用户”同款): + --private-users=1000:65536 —— 单元内 uid 0 → 客机侧 chv(uid 1000)。 + 效果:workspace / ~/.codex(chv 所有)单元内仍可写;host-root 0600 + 文件(如 /etc/chv/guest-env)在单元内显示为 overflow uid、不可读; + 再加 --inaccessible 遮蔽作纵深防御。 + +网络:--network-veth。宿主侧 ve-chv-cell(10.200.0.1/24),单元侧 +host0(10.200.0.2/24,默认路由经 10.200.0.1),客机侧 NAT 出站 +(MASQUERADE 走客机默认路由的上行口,动态发现)。单元内配网经 +nsenter 进单元 init 的 netns 做。veth 建不起来即大声失败,不静默 +回落共享网络;CHV_CELL_NETWORK=shared 是显式调试逃生口,须显式 +设置并打日志,绝不默认。 + +环境透传:nspawn 不透传调用方环境,本脚本只转发白名单变量(模型 +网关与代理变量)。GITEA_TOKEN 等凭证变量不在白名单里,绝不进单元。 + +用法(客机侧 root 运行,执行器经 SSH 调): + chv-cell run [--network veth|shared] -- + # 例:chv-cell run -- python3 /app/host.py --state + # /home/chv/.codex/host-state.json --codex-home /home/chv/.codex + # --sandbox danger-full-access dispatch --name t --task ... + chv-cell argv [--network veth|shared] -- + # 只打印 nspawn 参数表(JSON),不启动 + chv-cell network-commands [--uplink IFACE] [--cell-pid PID] + # 只打印配网命令表(JSON),不执行 +只用标准库。单元启动失败、配网失败一律非零退出,不静默降级。 +""" +import json +import os +import pwd +import re +import subprocess +import sys +import time + +MACHINE = "chv-cell" +# 单元映射:单元内 uid 0 → 客机侧 chv(uid 1000,非特权用户)。 +UID_BASE = 1000 +UID_COUNT = 65536 +CELL_USER = "chv" +CELL_UID = 1000 +CELL_GID = 1000 +WORKSPACE = "/home/chv/workspace" +CODEX_HOME = "/home/chv/.codex" +BROKER_DIR = "/run/chv" +CELL_PID_FILE = "/run/chv/cell.pid" + +# 单元网络(与客机 tap 网 192.168.10.0/24 不重叠的独立 /24)。 +CELL_NET = "10.200.0.0/24" +HOST_IP = "10.200.0.1" +CELL_IP = "10.200.0.2" +HOST_IFACE = "ve-chv-cell" +CELL_IFACE = "host0" + +# Muse 点名要丢的能力 + 同批丢的危险能力(nspawn 默认集已有限, +# 这里是显式再丢,--drop-capability 逐项写,测试逐项断言)。 +DROP_CAPABILITIES = [ + "CAP_SYS_ADMIN", + "CAP_SYS_PTRACE", + "CAP_NET_ADMIN", + "CAP_NET_RAW", + "CAP_SYS_MODULE", + "CAP_SYS_RAWIO", + "CAP_SYS_BOOT", + "CAP_MKNOD", +] + +# syscall 过滤:@system-service 集合再扣掉 io_uring 三件套 +#(Muse 点名禁用 io_uring)。 +SYSCALL_FILTER = ("@system-service ~io_uring_setup " + "~io_uring_enter ~io_uring_register") + +# 单元内遮掉的路径:凭证与浏览器(单元内跑不起自己的浏览器, +# 只能走 broker 的 chv-browser)。 +INACCESSIBLE_BASE = [ + "/etc/chv", + "/home/chv/.git-credentials", + "/usr/local/share/ms-playwright", +] + +# 只转发白名单环境变量进单元。凭证类变量(GITEA_TOKEN 等) +# 不在白名单里,绝不进单元。 +FORWARD_ENV_ALLOWLIST = ( + "SUB2API_BASE_URL", + "SUB2API_API_KEY", + "XCM_API_KEY", + "HTTP_PROXY", + "HTTPS_PROXY", + "NO_PROXY", + "http_proxy", + "https_proxy", + "no_proxy", +) + + +def _playwright_package_dir(): + """playwright pip 包目录(遮蔽用)。算不出时回落到 noble 默认路径。""" + try: + import sysconfig + purelib = sysconfig.get_path("purelib") + if purelib: + return os.path.join(purelib, "playwright") + except Exception: + pass + return "/usr/local/lib/python3.12/dist-packages/playwright" + + +def build_nspawn_argv(command, network="veth", + workspace=WORKSPACE, codex_home=CODEX_HOME, + broker_dir=BROKER_DIR, + playwright_package_dir=None): + """产 nspawn 参数表(纯函数,可单测)。command 是单元内要跑的命令 + (list,如 ["python3", "/app/host.py", ...])。""" + if not command: + raise ValueError("chv-cell: 单元内命令不能为空") + if network not in ("veth", "shared"): + raise ValueError("chv-cell: network 只能是 veth|shared:%r" + % network) + if playwright_package_dir is None: + playwright_package_dir = _playwright_package_dir() + argv = [ + "systemd-nspawn", "--quiet", + "--machine=%s" % MACHINE, + "--directory=/", + "--as-pid2", + "--private-users=%d:%d" % (UID_BASE, UID_COUNT), + "--no-new-privileges", + "--system-call-filter=%s" % SYSCALL_FILTER, + ] + for cap in DROP_CAPABILITIES: + argv.append("--drop-capability=%s" % cap) + for path in INACCESSIBLE_BASE + [playwright_package_dir]: + argv.append("--inaccessible=%s" % path) + argv += [ + "--bind=%s" % workspace, + "--bind=%s" % codex_home, + "--bind=%s" % broker_dir, + "--tmpfs=/tmp", + ] + if network == "veth": + argv.append("--network-veth") + argv += [ + "--setenv=CHV_CELL=1", + "--setenv=HOME=/home/chv", + ] + for name in FORWARD_ENV_ALLOWLIST: + value = os.environ.get(name) + if value: + argv.append("--setenv=%s=%s" % (name, value)) + argv += ["--chdir=%s" % workspace] + argv += list(command) + return argv + + +def parse_default_route_iface(ip_route_output): + """从 `ip route show default` 的输出里解析上行口名。 + 无默认路由时返回 None(调用方大声失败)。""" + for line in (ip_route_output or "").splitlines(): + line = line.strip() + if not line.startswith("default "): + continue + m = re.search(r"\bdev\s+(\S+)", line) + if m: + return m.group(1) + return None + + +def find_uplink_iface(run=None): + """发现客机上行口(默认路由的 dev)。run 供单测注入。""" + run = run or subprocess.run + proc = run(["ip", "route", "show", "default"], + capture_output=True, text=True) + if proc.returncode != 0: + return None + return parse_default_route_iface(proc.stdout) + + +def host_network_commands(uplink_iface): + """客机侧配网命令表(纯函数):ve 接口起 IP + 转发 + NAT。 + 调用方按顺序执行,任一失败即停。""" + if not uplink_iface: + raise ValueError("chv-cell: 上行口名为空,配不出 NAT") + return [ + ["ip", "addr", "add", "%s/24" % HOST_IP, "dev", HOST_IFACE], + ["ip", "link", "set", HOST_IFACE, "up"], + ["sysctl", "-w", "net.ipv4.ip_forward=1"], + ["iptables", "-t", "nat", "-A", "POSTROUTING", + "-s", CELL_NET, "-o", uplink_iface, "-j", "MASQUERADE"], + ["iptables", "-A", "FORWARD", "-i", HOST_IFACE, + "-o", uplink_iface, "-j", "ACCEPT"], + ["iptables", "-A", "FORWARD", "-i", uplink_iface, + "-o", HOST_IFACE, "-m", "state", "--state", + "RELATED,ESTABLISHED", "-j", "ACCEPT"], + ] + + +def host_network_cleanup_commands(uplink_iface): + """配网清理命令表(与 host_network_commands 配对,-A 改 -D)。 + 清理是 best-effort:删不掉(规则已不在)不报错。""" + if not uplink_iface: + raise ValueError("chv-cell: 上行口名为空,清不了 NAT") + cmds = [] + for cmd in host_network_commands(uplink_iface): + if "iptables" not in cmd: + continue + cleaned = [c if c != "-A" else "-D" for c in cmd] + cmds.append(cleaned) + return cmds + + +def find_cell_init_pid(nspawn_pid, proc_root="/proc"): + """找单元 init 的 pid:nspawn 进程的第一个子进程 + (--as-pid2 时它是 stub init,payload 是 PID 2;两者同 netns)。 + 找不到返回 None。""" + try: + task_dir = os.path.join(proc_root, str(nspawn_pid), "task") + for tid in os.listdir(task_dir): + children_path = os.path.join(task_dir, tid, "children") + try: + with open(children_path) as fh: + children = fh.read().split() + except OSError: + continue + if children: + return int(children[0]) + except (OSError, ValueError): + return None + return None + + +def cell_network_commands(cell_pid): + """单元内配网命令表(纯函数):nsenter 进单元 init 的 netns, + 给 host0 配 IP、起接口、加默认路由。""" + if not cell_pid: + raise ValueError("chv-cell: 单元 init pid 为空,配不了单元内网络") + prefix = ["nsenter", "-t", str(cell_pid), "-n"] + return [ + prefix + ["ip", "addr", "add", "%s/24" % CELL_IP, + "dev", CELL_IFACE], + prefix + ["ip", "link", "set", CELL_IFACE, "up"], + prefix + ["ip", "route", "add", "default", "via", HOST_IP], + ] + + +def _ensure_bind_source(path, uid=CELL_UID, gid=CELL_GID): + """bind 源目录须存在且属 chv(单元内 root 即客机 chv,属主不对 + 就写不进 workspace)。root 建的目录要 chown 回 chv。""" + os.makedirs(path, exist_ok=True) + try: + st = os.stat(path) + if st.st_uid != uid or st.st_gid != gid: + os.chown(path, uid, gid) + except OSError as e: + print("chv-cell: chown %s 失败:%s" % (path, e), file=sys.stderr) + raise + + +def _wait_for_iface(iface, timeout=10.0): + deadline = time.time() + timeout + while time.time() < deadline: + proc = subprocess.run(["ip", "link", "show", iface], + capture_output=True) + if proc.returncode == 0: + return True + time.sleep(0.2) + return False + + +def _run_cmds(cmds, what): + for cmd in cmds: + print("chv-cell: + %s" % " ".join(cmd), file=sys.stderr) + proc = subprocess.run(cmd) + if proc.returncode != 0: + raise RuntimeError( + "chv-cell: %s 失败(rc=%d):%s" + % (what, proc.returncode, " ".join(cmd))) + + +def run_cell(command, network="veth"): + """起单元、配网、等 payload 退出、清 NAT,返回 payload 退出码。""" + if os.geteuid() != 0: + print("chv-cell: 须以 root 运行(nspawn + 配网都要 root)", + file=sys.stderr) + return 2 + _ensure_bind_source(WORKSPACE) + _ensure_bind_source(CODEX_HOME) + _ensure_bind_source(BROKER_DIR, uid=0, gid=0) + argv = build_nspawn_argv(command, network=network) + print("chv-cell: 启动单元:%s" % " ".join(argv), file=sys.stderr) + proc = subprocess.Popen(argv) + uplink = None + cell_pid = None + try: + if network == "veth": + if not _wait_for_iface(HOST_IFACE): + raise RuntimeError( + "chv-cell: veth 接口 %s 未出现,拒绝回落共享网络" + % HOST_IFACE) + uplink = find_uplink_iface() + if not uplink: + raise RuntimeError( + "chv-cell: 客机无默认路由,配不出 NAT") + _run_cmds(host_network_commands(uplink), "客机侧配网") + cell_pid = find_cell_init_pid(proc.pid) + if not cell_pid: + raise RuntimeError("chv-cell: 找不到单元 init pid") + with open(CELL_PID_FILE, "w") as fh: + fh.write(str(cell_pid)) + _run_cmds(cell_network_commands(cell_pid), "单元内配网") + return proc.wait() + finally: + if uplink: + for cmd in host_network_cleanup_commands(uplink): + subprocess.run(cmd, capture_output=True) + try: + if os.path.exists(CELL_PID_FILE): + os.remove(CELL_PID_FILE) + except OSError: + pass + if proc.poll() is None: + proc.terminate() + try: + proc.wait(timeout=10) + except subprocess.TimeoutExpired: + proc.kill() + + +def main(argv): + if len(argv) < 1: + print("用法:chv-cell run|argv|network-commands …", + file=sys.stderr) + return 2 + verb = argv[0] + rest = argv[1:] + network = os.environ.get("CHV_CELL_NETWORK", "veth") + # --network 只在 run/argv 前接受,其它位置一律不认。 + if rest and rest[0] == "--network": + if len(rest) < 2: + print("chv-cell: --network 缺值", file=sys.stderr) + return 2 + network = rest[1] + rest = rest[2:] + if network not in ("veth", "shared"): + print("chv-cell: network 只能是 veth|shared:%r" % network, + file=sys.stderr) + return 2 + if network == "shared": + print("chv-cell: 警告:共享网络模式(调试逃生口),单元与" + "客机共用网络命名空间", file=sys.stderr) + if verb == "run": + if not rest or rest[0] != "--": + print("chv-cell: run 用法:chv-cell run -- ", + file=sys.stderr) + return 2 + command = rest[1:] + if not command: + print("chv-cell: 单元内命令不能为空", file=sys.stderr) + return 2 + try: + return run_cell(command, network=network) + except RuntimeError as e: + print(str(e), file=sys.stderr) + return 1 + if verb == "argv": + if not rest or rest[0] != "--": + print("chv-cell: argv 用法:chv-cell argv -- ", + file=sys.stderr) + return 2 + print(json.dumps(build_nspawn_argv(rest[1:], network=network), + ensure_ascii=False)) + return 0 + if verb == "network-commands": + uplink = None + cell_pid = None + i = 0 + while i < len(rest): + if rest[i] == "--uplink" and i + 1 < len(rest): + uplink = rest[i + 1] + i += 2 + elif rest[i] == "--cell-pid" and i + 1 < len(rest): + cell_pid = int(rest[i + 1]) + i += 2 + else: + print("chv-cell: network-commands 未知参数:%s" + % rest[i], file=sys.stderr) + return 2 + out = {} + if uplink: + out["host"] = host_network_commands(uplink) + out["host_cleanup"] = host_network_cleanup_commands(uplink) + if cell_pid: + out["cell"] = cell_network_commands(cell_pid) + print(json.dumps(out, ensure_ascii=False)) + return 0 + print("chv-cell: 未知子命令:%s" % verb, file=sys.stderr) + return 2 + + +if __name__ == "__main__": + raise SystemExit(main(sys.argv[1:])) diff --git a/cloud/tests/test_chv_muse_p2.py b/cloud/tests/test_chv_muse_p2.py new file mode 100644 index 0000000..2ca4069 --- /dev/null +++ b/cloud/tests/test_chv_muse_p2.py @@ -0,0 +1,616 @@ +"""对齐 Muse P2 测试:客机 nspawn 单元 + 浏览器子代理独立。 + +只测本轨交付物(chv-guest/chv-cell.py、chv-broker.py、chv-browser.py、 +build.sh 干跑、buildinfo.py 内容项)。全部用假 socket/假进程/假 +playwright,绝不连真实网络、不起真实 VM、不改生产行为。 +""" +import importlib.util +import json +import os +import shutil +import socket +import struct +import subprocess +import sys +import tempfile +import unittest +from pathlib import Path +from unittest import mock + +ROOT = Path(__file__).resolve().parents[2] + + +def _load_script(name): + """把 chv-guest 下的独立脚本当模块加载(无包结构)。""" + path = ROOT / "chv-guest" / name + spec = importlib.util.spec_from_file_location( + "chv_guest_p2_" + name[:-3], str(path)) + module = importlib.util.module_from_spec(spec) + spec.loader.exec_module(module) + return module + + +CELL = _load_script("chv-cell.py") +BROKER = _load_script("chv-broker.py") +BROWSER = _load_script("chv-browser.py") +BUILDINFO = _load_script("buildinfo.py") + + +class CellArgvTest(unittest.TestCase): + def test_private_users_maps_cell_root_to_chv(self): + argv = CELL.build_nspawn_argv(["true"]) + self.assertIn("--private-users=1000:65536", argv) + self.assertIn("--machine=chv-cell", argv) + self.assertIn("--directory=/", argv) + self.assertIn("--as-pid2", argv) + self.assertIn("--no-new-privileges", argv) + + def test_muse_named_capabilities_dropped(self): + argv = CELL.build_nspawn_argv(["true"]) + for cap in ("CAP_SYS_PTRACE", "CAP_NET_ADMIN", "CAP_SYS_ADMIN", + "CAP_NET_RAW", "CAP_SYS_MODULE"): + self.assertIn("--drop-capability=%s" % cap, argv) + # 绝不能出现“保留这些能力”的写法。 + for token in argv: + self.assertFalse(token.startswith("--capability=CAP_SYS_PTRACE")) + self.assertFalse(token.startswith("--capability=CAP_NET_ADMIN")) + + def test_io_uring_blocked_in_syscall_filter(self): + argv = CELL.build_nspawn_argv(["true"]) + filt = [a for a in argv if a.startswith("--system-call-filter=")] + self.assertEqual(len(filt), 1) + self.assertIn("@system-service", filt[0]) + for call in ("io_uring_setup", "io_uring_enter", + "io_uring_register"): + self.assertIn("~%s" % call, filt[0]) + + def test_secrets_and_browser_hidden_from_cell(self): + argv = CELL.build_nspawn_argv( + ["true"], + playwright_package_dir="/usr/local/lib/python3.12/dist-packages/playwright") + hidden = [a.split("=", 1)[1] for a in argv + if a.startswith("--inaccessible=")] + for path in ("/etc/chv", "/home/chv/.git-credentials", + "/usr/local/share/ms-playwright", + "/usr/local/lib/python3.12/dist-packages/playwright"): + self.assertIn(path, hidden) + + def test_writable_binds_and_tmpfs(self): + argv = CELL.build_nspawn_argv(["true"]) + self.assertIn("--bind=/home/chv/workspace", argv) + self.assertIn("--bind=/home/chv/.codex", argv) + self.assertIn("--bind=/run/chv", argv) + self.assertIn("--tmpfs=/tmp", argv) + self.assertIn("--chdir=/home/chv/workspace", argv) + self.assertIn("--setenv=CHV_CELL=1", argv) + self.assertIn("--setenv=HOME=/home/chv", argv) + + def test_veth_default_and_shared_opt_in(self): + self.assertIn("--network-veth", + CELL.build_nspawn_argv(["true"])) + self.assertNotIn("--network-veth", + CELL.build_nspawn_argv(["true"], + network="shared")) + with self.assertRaises(ValueError): + CELL.build_nspawn_argv(["true"], network="macvlan") + with self.assertRaises(ValueError): + CELL.build_nspawn_argv([]) + + def test_env_forward_allowlist(self): + env = {"SUB2API_API_KEY": "sk-test", + "SUB2API_BASE_URL": "https://example.invalid", + "GITEA_TOKEN": "must-never-enter-cell", + "XCM_API_KEY": "xk-test"} + with mock.patch.dict(os.environ, env, clear=False): + argv = CELL.build_nspawn_argv(["true"]) + self.assertIn("--setenv=SUB2API_API_KEY=sk-test", argv) + self.assertIn("--setenv=XCM_API_KEY=xk-test", argv) + self.assertFalse(any("GITEA_TOKEN" in a for a in argv)) + + +class CellNetworkTest(unittest.TestCase): + def test_host_network_commands_nat_via_uplink(self): + cmds = CELL.host_network_commands("enp0s5") + flat = [" ".join(c) for c in cmds] + self.assertIn("ip addr add 10.200.0.1/24 dev ve-chv-cell", flat) + self.assertIn("ip link set ve-chv-cell up", flat) + self.assertIn("sysctl -w net.ipv4.ip_forward=1", flat) + self.assertTrue(any("iptables -t nat -A POSTROUTING -s 10.200.0.0/24 " + "-o enp0s5 -j MASQUERADE" in line + for line in flat)) + self.assertTrue(any("-A FORWARD -i ve-chv-cell -o enp0s5 -j ACCEPT" + in line for line in flat)) + with self.assertRaises(ValueError): + CELL.host_network_commands("") + + def test_host_network_cleanup_mirrors_with_delete(self): + cmds = CELL.host_network_cleanup_commands("enp0s5") + self.assertTrue(cmds) + for cmd in cmds: + self.assertIn("iptables", cmd) + self.assertIn("-D", cmd) + self.assertNotIn("-A", cmd) + + def test_cell_network_commands_via_nsenter(self): + cmds = CELL.cell_network_commands(4321) + flat = [" ".join(c) for c in cmds] + self.assertIn("nsenter -t 4321 -n ip addr add 10.200.0.2/24 " + "dev host0", flat) + self.assertIn("nsenter -t 4321 -n ip link set host0 up", flat) + self.assertIn("nsenter -t 4321 -n ip route add default via " + "10.200.0.1", flat) + with self.assertRaises(ValueError): + CELL.cell_network_commands(0) + + def test_parse_default_route_iface(self): + out = ("default via 192.168.10.1 dev enp0s5 proto dhcp metric 100\n" + "192.168.10.0/24 dev enp0s5 proto kernel scope link src " + "192.168.10.10\n") + self.assertEqual(CELL.parse_default_route_iface(out), "enp0s5") + self.assertIsNone(CELL.parse_default_route_iface("")) + self.assertIsNone(CELL.parse_default_route_iface( + "192.168.10.0/24 dev enp0s5 proto kernel scope link\n")) + + def test_find_uplink_iface_uses_injected_runner(self): + class FakeProc: + returncode = 0 + stdout = "default via 192.168.10.1 dev tap0 proto static\n" + + def fake_run(cmd, **kwargs): + self.assertEqual(cmd, ["ip", "route", "show", "default"]) + return FakeProc() + + self.assertEqual(CELL.find_uplink_iface(run=fake_run), "tap0") + + def test_find_cell_init_pid_from_fake_proc(self): + with tempfile.TemporaryDirectory() as tmp: + task_dir = os.path.join(tmp, "777", "task", "777") + os.makedirs(task_dir) + with open(os.path.join(task_dir, "children"), "w") as fh: + fh.write("778 779") + self.assertEqual( + CELL.find_cell_init_pid(777, proc_root=tmp), 778) + self.assertIsNone( + CELL.find_cell_init_pid(999, proc_root=tmp)) + + +class BrokerTest(unittest.TestCase): + def _fake_proc(self, root, mapping): + """mapping: pid -> ns 标识。建出 /proc//ns/pid 软链。""" + for pid, ns in mapping.items(): + d = os.path.join(root, str(pid), "ns") + os.makedirs(d) + os.symlink("pid:[%s]" % ns, os.path.join(d, "pid")) + + def test_peer_allowed_same_pid_ns(self): + with tempfile.TemporaryDirectory() as tmp: + self._fake_proc(tmp, {100: "111", 200: "111"}) + ok, _ = BROKER.peer_allowed(1000, 200, cell_pid=100, + proc_root=tmp) + self.assertTrue(ok) + + def test_peer_denied_uid_outside_cell_range(self): + with tempfile.TemporaryDirectory() as tmp: + self._fake_proc(tmp, {100: "111", 200: "111"}) + for uid in (0, 999, 66536, 100000): + ok, reason = BROKER.peer_allowed(uid, 200, cell_pid=100, + proc_root=tmp) + self.assertFalse(ok, uid) + self.assertIn("单元映射范围", reason) + + def test_peer_denied_different_pid_ns(self): + with tempfile.TemporaryDirectory() as tmp: + self._fake_proc(tmp, {100: "111", 200: "222"}) + ok, reason = BROKER.peer_allowed(1000, 200, cell_pid=100, + proc_root=tmp) + self.assertFalse(ok) + self.assertIn("pid ns", reason) + + def test_peer_denied_when_cell_ns_unreadable(self): + with tempfile.TemporaryDirectory() as tmp: + self._fake_proc(tmp, {200: "111"}) + ok, _ = BROKER.peer_allowed(1000, 200, cell_pid=100, + proc_root=tmp) + self.assertFalse(ok) + + def test_peer_uid_only_when_cell_pid_missing(self): + ok, reason = BROKER.peer_allowed(1000, 200, cell_pid=None) + self.assertTrue(ok) + self.assertIn("cell.pid 缺失", reason) + + def test_ping_and_unknown_op(self): + self.assertEqual(BROKER.handle_request({"op": "ping"}), + {"ok": True, "cell": True}) + resp = BROKER.handle_request({"op": "rm"}) + self.assertFalse(resp["ok"]) + resp = BROKER.handle_request("not-a-dict") + self.assertFalse(resp["ok"]) + + def test_sync_rejects_bad_action_and_outside_repo(self): + resp = BROKER.handle_request({"op": "sync", "action": "reset"}) + self.assertFalse(resp["ok"]) + for bad in ("/etc/passwd", "/home/chv/workspace/../..", + "/home/chv", "/tmp/x"): + resp = BROKER.handle_request( + {"op": "sync", "action": "pull", "repo": bad}) + self.assertFalse(resp["ok"], bad) + + def test_sync_runs_chv_sync_as_chv_without_shell(self): + seen = {} + + class FakeProc: + returncode = 0 + stdout = "已同步\n" + stderr = "" + + def fake_run(cmd, **kwargs): + seen["cmd"] = cmd + self.assertFalse(kwargs.get("shell", False)) + return FakeProc() + + with mock.patch.object(BROKER.subprocess, "run", fake_run): + resp = BROKER.handle_request( + {"op": "sync", "action": "status", + "repo": "/home/chv/workspace/demo"}) + self.assertTrue(resp["ok"]) + self.assertEqual(seen["cmd"][:6], + ["runuser", "-u", "chv", "--", "env", + "HOME=/home/chv"]) + self.assertIn("/usr/local/bin/chv-sync", seen["cmd"]) + self.assertIn("status", seen["cmd"]) + self.assertIn("/home/chv/workspace/demo", seen["cmd"]) + + def test_sync_failure_reported_not_swallowed(self): + class FakeProc: + returncode = 1 + stdout = "" + stderr = "分叉\n" + + with mock.patch.object(BROKER.subprocess, "run", + lambda *a, **k: FakeProc()): + resp = BROKER.handle_request({"op": "sync", "action": "pull"}) + self.assertFalse(resp["ok"]) + self.assertEqual(resp["returncode"], 1) + + def test_browser_forward_and_down(self): + payload = {"ok": True, "title": "Example Domain"} + + class FakeConn: + def __init__(self, *args, **kwargs): + self.sent = b"" + + def settimeout(self, timeout): + pass + + def connect(self, path): + self.path = path + + def sendall(self, data): + self.sent = data + + def recv(self, n): + data, self._buf = self._buf, b"" + return data + + def close(self): + pass + + fake = FakeConn() + fake._buf = (json.dumps(payload) + "\n").encode() + with mock.patch.object(BROKER.socket, "socket", + lambda *a, **k: fake): + resp = BROKER.handle_request( + {"op": "browser", "action": "open"}, + browser_sock="/run/chv/browser.sock") + self.assertEqual(resp, payload) + self.assertEqual(json.loads(fake.sent.decode())["op"], "browser") + + def boom(*args, **kwargs): + raise OSError("no socket") + + with mock.patch.object(BROKER.socket, "socket", boom): + resp = BROKER.handle_request({"op": "browser"}) + self.assertFalse(resp["ok"]) + self.assertIn("浏览器子代理未就绪", resp["error"]) + + def test_peer_creds_unpack(self): + # struct ucred = pid/uid/gid 三个 int;broker 用它验对端。 + raw = struct.pack("3i", 4321, 1000, 1000) + + class FakeConn: + def getsockopt(self, level, opt, size): + self.seen = (level, opt, size) + return raw + + fake = FakeConn() + self.assertEqual(BROKER._peer_creds(fake), (4321, 1000, 1000)) + self.assertEqual(fake.seen[1], socket.SO_PEERCRED) + + +class BrowserTest(unittest.TestCase): + def test_validate_url(self): + self.assertTrue(BROWSER.validate_url("https://example.com")[0]) + self.assertTrue(BROWSER.validate_url("http://example.com/x")[0]) + self.assertTrue( + BROWSER.validate_url("https://user@example.com/")[0]) + for bad in ("ftp://example.com", "file:///etc/passwd", + "https://localhost:8080/", "http://127.0.0.1/", + "https://192.168.1.1/", "https://10.0.0.5/", + "https://172.17.0.2/", "", None): + ok, _ = BROWSER.validate_url(bad) + self.assertFalse(ok, bad) + + def test_truncate_a11y_small_tree_untouched(self): + tree = {"role": "WebArea", "name": "Example", + "children": [{"role": "heading", "name": "Hi"}]} + out, truncated = BROWSER.truncate_a11y(tree) + self.assertFalse(truncated) + self.assertEqual(out, tree) + + def test_truncate_a11y_node_cap(self): + tree = {"role": "WebArea", + "children": [{"role": "link", "name": "l%d" % i} + for i in range(300)]} + out, truncated = BROWSER.truncate_a11y(tree) + self.assertTrue(truncated) + self.assertLessEqual(len(out["children"]), 200) + + def test_truncate_a11y_char_cap(self): + tree = {"role": "WebArea", "name": "x" * 70000, "children": []} + out, truncated = BROWSER.truncate_a11y(tree) + self.assertTrue(truncated) + self.assertTrue(out.get("truncated_chars")) + + def test_forbidden_ops_never_execute(self): + agent = mock.Mock() + for op in ("evaluate", "content", "devtools", "js"): + resp = BROWSER.handle_request(agent, {"op": op}) + self.assertFalse(resp["ok"]) + self.assertIn("不暴露", resp["error"]) + agent.open.assert_not_called() + resp = BROWSER.handle_request(agent, {"op": "bogus"}) + self.assertFalse(resp["ok"]) + + def test_open_rejects_bad_url_without_loading_playwright(self): + def exploding_factory(): + raise AssertionError("坏 URL 不应加载 playwright") + + agent = BROWSER.BrowserAgent(playwright_factory=exploding_factory) + resp = agent.open("https://127.0.0.1/secret") + self.assertFalse(resp["ok"]) + + def test_snapshot_and_screenshot_need_open_first(self): + agent = BROWSER.BrowserAgent( + playwright_factory=lambda: (_ for _ in ()).throw( + AssertionError("不应加载 playwright"))) + self.assertFalse(agent.snapshot()["ok"]) + self.assertFalse(agent.screenshot()["ok"]) + self.assertTrue(agent.close()["ok"]) + + def test_open_returns_a11y_snapshot_with_fake_playwright(self): + page = FakePage() + with tempfile.TemporaryDirectory() as tmp: + agent = BROWSER.BrowserAgent( + playwright_factory=lambda: FakePlaywright(page), + shots_dir=tmp) + resp = agent.open("https://example.com") + self.assertTrue(resp["ok"]) + self.assertEqual(resp["title"], "Example Domain") + self.assertEqual(resp["url"], "https://example.com/") + self.assertEqual(resp["a11y"]["role"], "WebArea") + snap = agent.snapshot() + self.assertTrue(snap["ok"]) + shot = agent.screenshot() + self.assertTrue(shot["ok"]) + self.assertTrue(shot["path"].endswith(".png")) + self.assertTrue(os.path.exists(shot["path"])) + self.assertTrue(agent.close()["ok"]) + + +class FakePage: + def __init__(self): + self.url = "https://example.com/" + self._closed = False + + def goto(self, url, timeout=None): + self.url = url.rstrip("/") + "/" + + def wait_for_load_state(self, state, timeout=None): + pass + + def title(self): + return "Example Domain" + + @property + def accessibility(self): + return self + + def snapshot(self): + return {"role": "WebArea", "name": "Example Domain", + "children": [{"role": "heading", "name": "Example"}]} + + def screenshot(self, path=None): + with open(path, "wb") as fh: + fh.write(b"\x89PNG") + + def close(self): + self._closed = True + + +class FakeBrowser: + def __init__(self, page): + self._page = page + + def new_page(self): + return self._page + + def close(self): + pass + + +class FakeEntered: + def __init__(self, page): + self.chromium = mock.Mock() + self.chromium.launch.return_value = FakeBrowser(page) + + +class FakePlaywright: + def __init__(self, page): + self._page = page + + def __enter__(self): + return FakeEntered(self._page) + + def __exit__(self, *exc): + return False + + +class CellCliTest(unittest.TestCase): + def test_cli_argv_prints_json(self): + import contextlib + import io + buf = io.StringIO() + with contextlib.redirect_stdout(buf): + rc = CELL.main(["argv", "--", "echo", "hi"]) + self.assertEqual(rc, 0) + argv = json.loads(buf.getvalue()) + self.assertIn("--private-users=1000:65536", argv) + self.assertEqual(argv[-2:], ["echo", "hi"]) + + def test_cli_run_requires_double_dash(self): + self.assertEqual(CELL.main(["run", "echo", "hi"]), 2) + self.assertEqual(CELL.main(["bogus"]), 2) + self.assertEqual(CELL.main([]), 2) + + def test_cli_network_commands_prints_tables(self): + import contextlib + import io + buf = io.StringIO() + with contextlib.redirect_stdout(buf): + rc = CELL.main(["network-commands", "--uplink", "enp0s5", + "--cell-pid", "99"]) + self.assertEqual(rc, 0) + out = json.loads(buf.getvalue()) + self.assertIn("host", out) + self.assertIn("cell", out) + self.assertIn("host_cleanup", out) + + def test_cli_unknown_args_rejected(self): + self.assertEqual(BROKER.main(["--bogus"]), 2) + self.assertEqual(BROWSER.main(["--bogus"]), 2) + + +class BuildInfoTest(unittest.TestCase): + def _synth_out(self, tmp): + out = os.path.join(tmp, "out") + os.makedirs(out) + for name, content in (("disk.raw", b"D"), ("vmlinux", b"K")): + with open(os.path.join(out, name), "wb") as fh: + fh.write(content) + for name, lines in ( + ("guest-versions-dpkg.txt", ["git 9.9.9-9ubuntu9"]), + ("guest-versions-pip.txt", ["playwright==9.9.9"]), + ("guest-versions-gt.txt", ["gt version 1.0.0"]), + ("guest-versions-chromium.txt", + ["chromium_headless_shell-1248"])): + with open(os.path.join(out, name), "w") as fh: + fh.write("\n".join(lines) + "\n") + return out + + def test_contents_include_p2_items(self): + with tempfile.TemporaryDirectory() as tmp: + out = self._synth_out(tmp) + info = BUILDINFO.build( + out, "chv-guest-24.04-0.164.0-alpha.9-abc1234", "24.04", + "0.164.0-alpha.9", "abc1234", "0" * 64, True, "1248") + for item in ("systemd-container", "chv-cell", "chv-broker", + "chv-browser"): + self.assertIn(item, info["contents"]) + + def test_extra_contents_appended_once(self): + with tempfile.TemporaryDirectory() as tmp: + out = self._synth_out(tmp) + info = BUILDINFO.build( + out, "chv-guest-24.04-0.164.0-alpha.9-abc1234", "24.04", + "0.164.0-alpha.9", "abc1234", "0" * 64, True, "1248", + ["vsock_notify"]) + self.assertIn("vsock_notify", info["contents"]) + info2 = BUILDINFO.build( + out, "chv-guest-24.04-0.164.0-alpha.9-abc1234", "24.04", + "0.164.0-alpha.9", "abc1234", "0" * 64, True, "1248", + ["vsock_notify", "chv-cell"]) + self.assertEqual(info2["contents"].count("chv-cell"), 1) + self.assertIn("vsock_notify", info2["contents"]) + + def test_main_accepts_tenth_arg(self): + with tempfile.TemporaryDirectory() as tmp: + out = self._synth_out(tmp) + args = ["buildinfo.py", out, + "chv-guest-24.04-0.164.0-alpha.9-abc1234", "24.04", + "0.164.0-alpha.9", "abc1234", "0" * 64, "1", "1248", + "vsock_notify"] + self.assertEqual(BUILDINFO.main(args), 0) + with open(os.path.join(out, "BUILDINFO.json")) as fh: + info = json.load(fh) + self.assertIn("vsock_notify", info["contents"]) + self.assertEqual(BUILDINFO.main(args[:9]), 0) + + +class BuildDryRunTest(unittest.TestCase): + def test_dry_run_contains_p2_steps(self): + # 在隔离的临时 git 仓里跑 build.sh --dry-run:工作树干净、 + # HEAD 与 REPO_COMMIT 一致、假 codex 二进制报对版本。 + with tempfile.TemporaryDirectory() as tmp: + repo = os.path.join(tmp, "repo") + guest = os.path.join(repo, "chv-guest") + bindir = os.path.join(tmp, "bin") + os.makedirs(guest) + os.makedirs(os.path.join(bindir, "codex-resources")) + for name in ("build.sh", "buildinfo.py", "check_lockfile.py", + "generate_lockfile.py", "chv-sync.sh", + "chv-cell.py", "chv-broker.py", "chv-browser.py"): + shutil.copy2(ROOT / "chv-guest" / name, + os.path.join(guest, name)) + shutil.copy2(ROOT / "host.py", os.path.join(repo, "host.py")) + codex = os.path.join(bindir, "codex") + with open(codex, "w") as fh: + fh.write("#!/bin/sh\necho 'codex-cli 0.164.0-alpha.9'\n") + os.chmod(codex, 0o755) + host_bin = os.path.join(bindir, "codex-code-mode-host") + with open(host_bin, "w") as fh: + fh.write("#!/bin/sh\nexit 0\n") + os.chmod(host_bin, 0o755) + subprocess.run(["git", "init", "-q"], cwd=repo, check=True) + subprocess.run(["git", "add", "-A"], cwd=repo, check=True) + subprocess.run( + ["git", "-c", "user.email=t@t", "-c", "user.name=t", + "commit", "-qm", "p2 dry-run fixture"], + cwd=repo, check=True) + head = subprocess.run( + ["git", "rev-parse", "--short", "HEAD"], cwd=repo, + capture_output=True, text=True, check=True).stdout.strip() + env = dict(os.environ) + env.update({ + "CODEX_BIN_DIR": bindir, + "UBUNTU_VERSION": "24.04", + "CODEX_VERSION": "0.164.0-alpha.9", + "REPO_COMMIT": head, + "UBUNTU_IMAGE_SHA256": "a" * 64, + }) + proc = subprocess.run( + ["bash", os.path.join(guest, "build.sh"), + "--dry-run", "out"], + cwd=repo, capture_output=True, text=True, env=env) + self.assertEqual(proc.returncode, 0, proc.stderr) + out = proc.stdout + proc.stderr + for needle in ("systemd-container", "chv-cell", "chv-broker", + "chv-browser", "chv-browser.service", + "chv-broker.service", "useradd --system", + "systemctl enable chv-broker chv-browser"): + self.assertIn(needle, out, needle) + # P1 文件不在本分支:干跑只警告、不失败。 + self.assertIn("vsock_notify.py 不在仓库里", out) + + +if __name__ == "__main__": + unittest.main() diff --git a/docs/chv-muse-p2-gap.md b/docs/chv-muse-p2-gap.md new file mode 100644 index 0000000..1d06ad6 --- /dev/null +++ b/docs/chv-muse-p2-gap.md @@ -0,0 +1,127 @@ +# 对齐 Muse P2 差距表:客机 nspawn 单元 + 浏览器子代理独立 + +依据:`~/workspace/research_notes/meta-muse-ai-architecture-20261101-0755/report.md` +(官方以 Meta security.muse.ai 2026-09-08 为准;第三方拆解只作参考, +逆向细节不得写成官方架构)。Muse 模型一句话:VM 内 systemd-nspawn +运行时单元(Hatch)跑代理,单元 root 映射为 host 侧非特权用户,自有 +虚拟网络接口,过滤 syscall(如禁用 io_uring),受限能力(如无 +CAP_SYS_PTRACE、无 CAP_NET_ADMIN);浏览器任务由专门浏览器子代理 +完成,只见页面无障碍树快照、不见原始 DOM,不能在页面上下文跑 JS, +CDP 由单元外 broker 管理。 + +本轨范围只限 P2:客机镜像层——systemd-nspawn 运行时单元 + 浏览器 +子代理独立。不碰 P1(执行器 WarmPool/气球/vsock,常驻 VM)、不碰 P3 +(authd 替身凭证、Sentinel 式出口审批)。分支 `feat-chv-muse-p2` +(从 `feat-chv-cover100` 分出,基线 HEAD `06c3e17`)。 + +## 1. 现状 vs Muse + +| 项 | 现状(实测/代码为准) | Muse 目标 | 本轨做法 | +|---|---|---|---| +| 代理运行位置 | codex 直接跑在客机用户态(执行器经 SSH 以 `chv` 身份调 `python3 /app/host.py`,它再起 `codex app-server`);无内层隔离 | 代理跑在 systemd-nspawn 运行时单元内 | `chv-cell run -- python3 /app/host.py --state … --codex-home … `:单命令单元(`--as-pid2`,不 boot 第二个 systemd);单元即沙箱 | +| 单元 root 映射 | 无此概念(`chv` 就是普通用户) | 单元 root 映射为 host 侧非特权用户 | `--private-users=1000:65536`:单元内 uid 0 → 客机侧 `chv`(uid 1000)。效果:workspace/`~/.codex`(`chv` 所有)单元内仍可写;host-root 0600 文件(如 `/etc/chv/guest-env`)在单元内显示为 overflow uid、不可读;再加 `--inaccessible=/etc/chv` 纵深防御 | +| 单元根文件系统 | 无(直接用客机根文件系统) | 单元自有根文件系统(完整 Debian 镜像),与 host 系统分离 | **本轮不做独立根文件系统**:单元复用客机根文件系统(`--directory=/`)。独立 minbase 根文件系统约 200–300MB,另要把 codex/playwright 依赖再装一份,会顶破 qcow2 ≤1GB 验收。文件隔离本轮是“读多写少 + 遮蔽”:`/usr`、`/app` 等 host-root 所有路径单元内只读不可写,`/etc/chv`、playwright 包与浏览器二进制 `--inaccessible` 遮掉,可写路径只有显式 bind(workspace、`~/.codex`、`/run/chv`)与 tmpfs(`/tmp`)。完整独立根文件系统列为下一阶段,需先重定镜像体积预算 | +| syscall 过滤 | 无 | 过滤 syscall,如禁用 io_uring | `--system-call-filter=@system-service ~io_uring_setup ~io_uring_enter ~io_uring_register` | +| 能力 | 客机 `chv` 是普通用户,但 codex 的 bwrap 沙箱在单元内另有问题(见下) | 无 CAP_SYS_PTRACE、无 CAP_NET_ADMIN | `--drop-capability=CAP_SYS_ADMIN,CAP_SYS_PTRACE,CAP_NET_ADMIN,CAP_NET_RAW,CAP_SYS_MODULE,CAP_SYS_RAWIO,CAP_SYS_BOOT,CAP_MKNOD` + `--no-new-privileges` | +| 单元内 codex 沙箱 | `host.py` 默认 `sandbox="workspace-write"`(bwrap) | 单元本身就是边界 | 单元内 codex 用 `danger-full-access`:bwrap 的 `--proc /proc` 挂载在嵌套命名空间里是已知走不通的形状(见 AGENTS.md 内层 proc 定位),单元内再套 bwrap 只会假隔离。执行器侧改 `--sandbox` 透传是 P1 的文件,本轨只定义接口(见 §4),不改 `executor.py` | +| 单元网络 | 无(与客机共享网络命名空间) | 单元自有虚拟网络接口 | `--network-veth`:宿主侧 `ve-chv-cell`(10.200.0.1/24)、单元侧 `host0`(10.200.0.2/24,默认路由经 10.200.0.1),客机侧 NAT 出站(MASQUERADE 走客机默认路由的上行口,动态发现,不写死口名)。单元内配网经 nsenter 进单元 init 的 netns 做(`chv-cell.py` 内实现)。veth 建不起来即大声失败,不静默回落共享网络;调试逃生口 `CHV_CELL_NETWORK=shared` 须显式设置、打日志,绝不默认 | +| 单元内外 IPC | 无(SSH 直达,host key 钉死) | Unix socket + SO_PEERCRED + 对端 ACL 的最小权限 IPC | `chv-broker`(客机侧、单元外):Unix socket `/run/chv/broker.sock`(bind 进单元)。SO_PEERCRED 校验:对端 uid 须在单元映射范围 1000–66535,且对端 pid 的 pid ns 须与 `/run/chv/cell.pid` 记的单元 init 同 ns(cell.pid 不存在时只做 uid 检查,并打日志)。操作白名单:`ping`、`sync pull|push|status`(仓目录只允许 `~/workspace` 下,`chv-sync` 以 `chv` 身份跑,不进 shell)、`browser …`(转发给浏览器子代理)。其余一律拒绝 | +| 浏览器 | Playwright headless-shell 由 agent 在客机内直接跑,能读原始 DOM、能跑 JS | 浏览器任务由专门浏览器子代理完成,只见无障碍树快照,不见原始 DOM,不能在页面上下文跑 JS,CDP 由单元外 broker 管理 | `chv-browser`(客机侧、单元外独立进程,用户 `chv-browser`):Unix socket `/run/chv/browser.sock`,操作只有 `open|snapshot|screenshot|close`。`open`/`snapshot` 返回标题、URL 与 Playwright `accessibility.snapshot()` 的无障碍树(截断到 200 节点/64KB);`screenshot` 落盘到 `/run/chv/shots/`(bind 进单元,单元只读路径取图)。**不暴露** `evaluate`、`content`(原始 DOM)、DevTools。单元内 playwright 包与浏览器二进制用 `--inaccessible` 遮掉,单元内跑不起自己的浏览器,只能走 broker | +| 凭证 | `/etc/chv/guest-env`(0600 root)+ `~/.git-credentials`(0600 chv),codex 在客机内直接可读后者 | 凭证由 authd 保管,代理只见替身令牌 | 本轨只做“单元看不见”:`/etc/chv` 与 `~/.git-credentials` 都 `--inaccessible`,同步走 broker 的 `sync` 操作(broker 在单元外跑 `chv-sync`,token 不进单元)。替身化本身是 P3,不在本轨冒充已做 | +| 客机启动同步 | `chv-sync boot` 由 cloud-init runcmd 以 root 跑,clone 到 `~/workspace` | — | 不动:boot clone 仍在单元外做;单元内只经 broker 做 pull/push/status | + +## 2. 改了什么(本轨交付物) + +- `chv-guest/chv-cell.py`(新,标准库):单元启动器。`build_nspawn_argv()` 纯函数产 + nspawn 参数表(单元映射、能力丢弃、syscall 过滤、遮蔽、bind、tmpfs、veth、 + `--as-pid2`、`CHV_CELL=1`);`configure_cell_network()` 产配网命令表(宿主侧 + ve 接口 + NAT + nsenter 进单元配 host0);`chv-cell run -- ` 起单元、 + 写 `/run/chv/cell.pid`、等退出码、清理 NAT 规则;`chv-cell argv` 只打印 + 参数表、不启动(干检查用)。 +- `chv-guest/chv-broker.py`(新,标准库):单元外 IPC broker。Unix socket + `/run/chv/broker.sock`,SO_PEERCRED + pid ns 双重校验,白名单操作见 §1, + `sync` 调 `chv-sync`(`runuser -u chv`,list argv、不进 shell),`browser` + 转发 `chv-browser.sock`。 +- `chv-guest/chv-browser.py`(新,标准库 + playwright 懒加载):浏览器子代理。 + Unix socket `/run/chv/browser.sock`,操作 `open|snapshot|screenshot|close`, + 只返回无障碍树快照(截断)+ 标题/URL/截图路径;无 `evaluate`/`content`。 +- `chv-guest/build.sh`:apt 加 `systemd-container`、`iproute2`、`iptables`; + 拷入 `chv-cell.py`/`chv-broker.py`/`chv-browser.py`(`--copy-in` + chmod); + 建系统用户 `chv-browser`;装 `chv-broker.service`、`chv-browser.service` + 并 `systemctl enable`;`/etc/tmpfiles.d/chv.conf` 建 `/run/chv` 与 + `/run/chv/shots`。P1 交接:`cloud/chv/guest/vsock_notify.py` 若在仓库里 + 存在则一并拷入(条件式,不存在时只警告、不失败——P1 文件在本分支尚无, + 合并后才有)。 +- `chv-guest/buildinfo.py`:`BASE_CONTENTS` 加 `systemd-container`、 + `chv-cell`、`chv-broker`、`chv-browser`;`build()` 支持可选 `extra_contents` + (vsock 接入时用,不改现有调用签名)。 +- `cloud/tests/test_chv_muse_p2.py`(新):单元参数表、broker 校验与白名单、 + 浏览器 API schema 与快照截断、build.sh 干跑含新步骤、buildinfo 新内容项。 + 全部用假 socket/假进程,绝不连真实网络、不起真实 VM。 +- `docs/chv-production-readiness.md`:补 P2 一节(单元隔离验收与待验证项), + 旧定案不删、只追加。 + +## 3. 诚实边界(本轮不宣称的东西) + +1. **没有独立根文件系统**:单元与客机共享根文件系统(只读挂载 + 遮蔽 + + bind),不是 Muse 的独立 Debian 镜像。一个有决心的单元内进程仍能读到 + 客机侧 world-readable 的文件内容(如 `/app/host.py`、系统库)。真正 + 的文件系统分离要等下一阶段的独立根文件系统 + 体积重预算。 +2. **SO_PEERCRED 的 uid 检查不区分单元 root 与客机 `chv`**:单元映射 + `--private-users=1000:65536` 让单元内 uid 0 在客机侧就是 uid 1000, + 与客机 `chv` 同 uid。broker 的主校验是 pid ns 对照(`cell.pid`), + uid 范围检查只是第二道;`cell.pid` 不存在时只做 uid 检查,并打日志。 + 这不是“单元身份”的密码学证明,是命名空间对照。 +3. **a11y-tree-only 是方向,不是已封死的强制**:浏览器 socket 不暴露 + DOM/JS 操作,且单元内跑不起 playwright(已遮蔽),但单元内进程仍 + 有 curl/Python 标准库,可以直接发 HTTP 请求取页面原文——那不是 + “浏览器 DOM 访问”,但也不是无障碍树。完整封堵(出口只走浏览器 + 子代理)是 P3 Sentinel 的事。 +4. **性能数字全是待验证目标**:单元启动耗时、veth+NAT 对模型调用延迟 + 的影响、快照截断后的 token 占用——本轮无真机验证(第三棒 V-1~V-9 + 占着 WSL 测试机),一律记入“真机待验证清单”,不在代码/文档里写成 + 已测得。 +5. **nspawn 参数以 systemd 255(`systemd-nspawn --help` 实测)为准**: + `--private-users=UIDBASE[:NUIDS]`、`--drop-capability`、 + `--system-call-filter`、`--inaccessible`、`--network-veth`、 + `--as-pid2` 均在本机 help 里确认存在;但在 CHV 客机内核(云镜像 + generic 内核)里 userns/veth/idmap 的实际行为,须真机验证。 +6. **WarmPool/常驻路径与本轨无关**:P1 的常驻 VM 若接进来,单元在常驻 + VM 内同样起(`chv-cell run` 与 VM 生命周期无关);本轨不改执行器, + 不碰 P1 的文件。 + +## 4. 与 P1/P3 的接口(不碰对方文件,只定契约) + +- **P1 → P2(执行器接单元)**:执行器调客内 host.py 的命令从 + `python3 /app/host.py …` 改为 + `chv-cell run -- python3 /app/host.py --state /home/chv/.codex/host-state.json --codex-home /home/chv/.codex --sandbox danger-full-access …`。 + `--sandbox danger-full-access` 只在单元内生效(单元即沙箱);非单元 + 路径(本地/Docker/冷启动回落)保持原沙箱不动。改动在 `executor.py`, + 由 P1 做,本轨只提供客机侧 `chv-cell` 命令与本节契约。 +- **P1 → P2(vsock 进镜像)**:P1 的 `cloud/chv/guest/vsock_notify.py` + 由本轨 `build.sh` 条件拷入(§2),合并后自动生效;P2 不改该脚本内容。 +- **P2 → P3(broker 预留)**:broker 的 `sync` 操作现在单元外直接跑 + `chv-sync`(token 仍在客机侧落盘);P3 做 authd 替身化时,只改 broker + 内部的凭证获取方式(调 authd 拿短期替身),不改 broker 的 socket 协议 + 与白名单。浏览器出口审批(只允许经 `chv-browser` 出站)同样是 P3 的 + Sentinel 接管点,本轨只做到“单元内无浏览器可跑”。 + +## 5. 真机待验证清单(round 11,测试机空闲后) + +- lite 验收照旧:qcow2 ≤1GB、冷启动到 SSH、codex 版本、headless-shell + 例、Gitea 双向同步(boot clone + push + pull + 分叉失败关闭)。 +- 单元隔离验收(新增,否定测试为主): + 1. 单元内 `cat /etc/chv/guest-env` 失败(遮蔽 + 0600 overflow uid)。 + 2. 单元内 `cat ~/.git-credentials` 失败;经 broker `sync status` 成功。 + 3. 单元内写 `/usr/local/bin/x` 失败(只读根文件系统)。 + 4. 单元内 `nsenter`/ptrace 客机侧进程失败(pid ns + dropped caps)。 + 5. 含 `io_uring_setup` 的程序在单元内报 EPERM/ENOSYS(syscall 过滤)。 + 6. 单元内 `python3 -c "import playwright"` 失败(已遮蔽);经 broker + `browser open https://example.com` 返回标题 Example Domain 与 + 无障碍树快照。 + 7. 单元内 codex 一轮 turn 完成(`danger-full-access` 下,不再套 bwrap)。 + 8. 单元内出站到模型网关通(veth + NAT);`chv-cell` 退出后 NAT 规则已清理。 + 9. broker 拒绝非单元进程的连接(pid ns 不符时拒绝;`cell.pid` 缺失时 + 只做 uid 检查并打日志——该降级须在日志里可见)。 +- 性能记录(只记录、不设门):单元启动到 codex 就绪耗时、veth+NAT 下 + 模型调用往返、qcow2 实际占用(systemd-container 与新脚本的增量)。 diff --git a/docs/chv-production-readiness.md b/docs/chv-production-readiness.md index bb4e6f2..2427b6b 100644 --- a/docs/chv-production-readiness.md +++ b/docs/chv-production-readiness.md @@ -833,3 +833,54 @@ UNSTABLE、未合并、未部署、真机 0 项。差:外部为主。 - 动态安全 2.0 → 2.0,未达 10:本轨无任何动态验证, V-7 未跑。差:外部(KVM 宿主 + 审计结论)。 + +## 15. 对齐 Muse P2 轨 feat-chv-muse-p2(客机 nspawn 单元 + 浏览器子代理) + +本节是准备工作,不是启用结论。本轨只到本地分支,不合并、不部署; +生产就绪与动态安全的外部关卡原样未动。完整差距表见 +docs/chv-muse-p2-gap.md(现状 vs Muse、改了什么、诚实边界、 +真机待验证清单、与 P1/P3 的接口)。 + +### 15.1 本轨量化结果(实测) + +- 新增 `cloud/tests/test_chv_muse_p2.py` 40 项,全过。命令(仓库根目录): + `python3 -m unittest cloud.tests.test_chv_muse_p2` +- 冒烟(真实 Unix socket 回路,/tmp 临时路径,nobody 身份): + broker ping 通、白名单拒绝通、sync 校验通、browser ping 通、 + evaluate 拒绝、localhost URL 拒绝、broker→browser 转发通; + root(uid 0)连 broker 被拒并有日志。见 §15.2。 +- 本机无 /dev/kvm:镜像 round 11 未构建,真机验证 0 项(待第三棒 + V-1~V-9 让出 WSL 测试机)。 + +### 15.2 本轨交付物 + +- `chv-guest/chv-cell.py`:单元启动器。`--private-users=1000:65536` + (单元内 root → 客机 chv)、丢 8 项能力(含 CAP_SYS_PTRACE / + CAP_NET_ADMIN)、syscall 过滤扣 io_uring 三件套、遮蔽 /etc/chv / + ~/.git-credentials / playwright、`--network-veth`(10.200.0.0/24 + + NAT,nsenter 配单元内网络)、白名单环境透传(GITEA_TOKEN 不进单元)。 +- `chv-guest/chv-broker.py`:单元外 IPC broker。Unix socket + + SO_PEERCRED(uid 范围 + pid ns 对照 cell.pid),白名单 ping / + sync / browser;sync 只调 chv-sync(runuser -u chv,list argv)。 +- `chv-guest/chv-browser.py`:浏览器子代理(用户 chv-browser)。 + 只暴露 open/snapshot/screenshot/close/ping;返回无障碍树快照 + (截断 200 节点/64KB);evaluate/content/DevTools 一律拒绝; + localhost 与私有地址字面量拒绝。 +- `chv-guest/build.sh`:apt 加 systemd-container/iproute2/iptables; + 拷入三脚本;建 chv-browser 系统用户;装 chv-broker/chv-browser + 两个 systemd 单元并 enable;tmpfiles.d 建 /run/chv。P1 的 + vsock_notify.py 条件拷入(文件不存在时只警告、不失败)。 +- `chv-guest/buildinfo.py`:内容项加 systemd-container/chv-cell/ + chv-broker/chv-browser;支持可选第 10 参数 extra_contents。 + +### 15.3 诚实边界(本轨不宣称) + +- 无独立根文件系统:单元与客机共享根文件系统(只读 + 遮蔽 + + bind),不是 Muse 的独立 Debian 镜像;world-readable 的客机文件 + 单元内仍可读。完整分离要等下一阶段 + 体积重预算。 +- SO_PEERCRED 的 uid 检查不区分单元 root 与客机 chv(同 uid 1000); + 主校验是 pid ns 对照;cell.pid 缺失时只做 uid 检查并打日志。 +- a11y-tree-only 是方向:单元内仍有 curl/Python 标准库可直接发 + HTTP 请求;完整出口封堵是 P3 Sentinel 的事。 +- 单元内 codex 用 danger-full-access(单元即沙箱,不再套 bwrap); + 执行器侧改 `--sandbox` 透传是 P1 的文件,本轨只定义了接口契约。