fix release targets fallback and git-based ssh deploy

This commit is contained in:
hy 2026-04-23 12:34:10 +08:00
parent 579e8a4aa5
commit 9cbb777674
4 changed files with 161 additions and 108 deletions

View File

@ -205,7 +205,7 @@ python3 ops/deploy_via_ssh.py
默认会:
1. 把当前工作区同步到远端 `/opt/hy-app-monitor`
1. 让远端主机从 Git 拉取指定分支到 `/opt/hy-app-monitor`
2. 把本地项目 `.env` 同步到远端项目 `.env`
3. 创建远端 `.venv`
4. 安装 `requirements-ops.txt`
@ -215,6 +215,13 @@ python3 ops/deploy_via_ssh.py
`ops/deploy_via_tat.py` 仍可保留成带外兜底,不建议继续作为日常主链路。
日常链路应先把本地改动提交并推送到远端仓库,再执行部署脚本,让 deploy 机按同一分支 `git pull`
```bash
git push origin main
python3 ops/deploy_via_ssh.py --repo-ref main
```
## 维护完整 Compose
默认只刷新本地完整模板:

View File

@ -719,7 +719,11 @@ def current_template_image(service_name: str) -> str:
# 优先从本地完整模板里拿当前运行时镜像仓路径。
records = service_records_by_name().get(service_name) or []
for record in records:
model = load_local_host_compose_model(record["host"])
try:
model = load_local_host_compose_model(record["host"])
except Exception: # noqa: BLE001
# 发布页只需要尽量推导仓路径;模板暂时不可读时继续尝试其他主机并允许后续兜底。
continue
image_ref = str(((model.get("services") or {}).get(service_name) or {}).get("image") or "").strip()
if image_ref.startswith(f"{HARBOR_REGISTRY}/{HARBOR_PROJECT}/"):
return image_ref

View File

@ -3,13 +3,14 @@
from __future__ import annotations
import argparse
import base64
import os
from pathlib import Path
import shlex
import subprocess
# 项目根目录,后面会从这里读取 .env 并打包代码
# 项目根目录,后面会从这里读取 .env 和当前 Git 仓信息
ROOT = Path(__file__).resolve().parents[1]
# 默认 .env 路径。
ENV_FILE = Path(os.environ.get("APP_ENV_FILE", ROOT / ".env")).expanduser()
@ -35,12 +36,53 @@ def load_env_file(path: Path) -> None:
os.environ.setdefault(key, value)
def local_git_output(args: list[str], *, label: str) -> str:
# 所有本地 Git 查询统一走这个函数,失败时直接抛明确错误。
try:
result = subprocess.run(
["git", "-C", str(ROOT), *args],
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT,
text=True,
check=False,
)
except FileNotFoundError as exc:
raise SystemExit("missing command for local git query: git") from exc
if result.returncode != 0:
raise SystemExit((result.stdout or "").strip() or f"{label} failed")
return (result.stdout or "").strip()
def default_repo_url() -> str:
# 默认直接沿用当前工作区 origin避免脚本和实际仓库地址漂移。
configured = os.environ.get("MONITOR_REPO_URL", "").strip()
if configured:
return configured
return local_git_output(["remote", "get-url", "origin"], label="detect git remote")
def default_repo_ref() -> str:
# 默认部署当前分支,满足“提交后推送、线上按同一分支拉取”的链路。
configured = os.environ.get("MONITOR_REPO_REF", "").strip()
if configured:
return configured
branch = local_git_output(["rev-parse", "--abbrev-ref", "HEAD"], label="detect current branch")
if branch == "HEAD":
raise SystemExit("current git HEAD is detached; pass --repo-ref explicitly")
return branch
# 先加载项目根目录 .env保证默认 SSH 目标也能从这里读取。
load_env_file(ENV_FILE)
# 默认远端部署目录。
DEFAULT_REMOTE_DIR = os.environ.get("MONITOR_REMOTE_DIR", "/opt/hy-app-monitor")
# 默认远端监听端口。
DEFAULT_APP_PORT = os.environ.get("APP_PORT", "2026")
# 默认远端仓库地址和分支。
DEFAULT_REPO_URL = default_repo_url()
DEFAULT_REPO_REF = default_repo_ref()
# 默认 SSH 主机、用户和端口。
DEFAULT_DEPLOY_SSH_HOST = os.environ.get("DEPLOY_SSH_HOST", "").strip()
DEFAULT_DEPLOY_SSH_USER = os.environ.get("DEPLOY_SSH_USER", "root").strip() or "root"
@ -49,12 +91,14 @@ DEFAULT_DEPLOY_SSH_IDENTITY_FILE = os.environ.get("DEPLOY_SSH_IDENTITY_FILE", ""
def parse_args() -> argparse.Namespace:
# 这里默认按当前项目目录直接同步到远端,不再依赖远端 git clone
parser = argparse.ArgumentParser(description="deploy hy-app-monitor to deploy host via ssh/scp")
# SSH 链路现在和 TAT 一样走远端 git 拉取,不再同步本地工作区文件树
parser = argparse.ArgumentParser(description="deploy hy-app-monitor to deploy host via remote git pull over ssh")
parser.add_argument("--host", default=DEFAULT_DEPLOY_SSH_HOST, help="deploy ssh host")
parser.add_argument("--user", default=DEFAULT_DEPLOY_SSH_USER, help="deploy ssh user")
parser.add_argument("--port", type=int, default=DEFAULT_DEPLOY_SSH_PORT, help="deploy ssh port")
parser.add_argument("--identity-file", default=DEFAULT_DEPLOY_SSH_IDENTITY_FILE, help="optional ssh identity file")
parser.add_argument("--repo-url", default=DEFAULT_REPO_URL, help="git repository url used by remote host")
parser.add_argument("--repo-ref", default=DEFAULT_REPO_REF, help="git ref pulled on remote host")
parser.add_argument("--remote-dir", default=DEFAULT_REMOTE_DIR, help="remote project directory")
parser.add_argument("--app-port", default=DEFAULT_APP_PORT, help="remote app port")
return parser.parse_args()
@ -68,7 +112,7 @@ def require_env_file() -> Path:
def ssh_base_args(host: str, *, user: str, port: int, identity_file: str) -> list[str]:
# 所有 SSH / SCP 命令统一复用同一组参数,避免行为不一致。
# 所有 SSH 命令统一复用同一组参数,避免行为不一致。
args = [
"-p",
str(port),
@ -89,25 +133,8 @@ def ssh_base_args(host: str, *, user: str, port: int, identity_file: str) -> lis
return args
def scp_base_args(*, port: int, identity_file: str) -> list[str]:
# SCP 的端口参数和 SSH 不同,这里单独组装一份。
args = [
"-P",
str(port),
"-o",
"BatchMode=yes",
"-o",
"ConnectTimeout=10",
"-o",
"StrictHostKeyChecking=accept-new",
]
if identity_file:
args.extend(["-i", identity_file])
return args
def run_command(command: list[str], *, input_bytes: bytes | None = None, label: str) -> bytes:
# 小工具统一把 stdout/stderr 合并,失败时直接抛完整错误文本。
# SSH 命令统一把 stdout/stderr 合并,失败时直接抛完整错误文本。
try:
result = subprocess.run(
command,
@ -124,88 +151,55 @@ def run_command(command: list[str], *, input_bytes: bytes | None = None, label:
return result.stdout or b""
def sync_project_tree(*, remote_dir: str, ssh_args: list[str]) -> None:
# 远端保留 .env / .venv / run其他项目内容全部按当前工作区覆盖。
remote_prepare_script = f"""set -euo pipefail
REMOTE_DIR={shlex.quote(remote_dir)}
mkdir -p "$REMOTE_DIR"
find "$REMOTE_DIR" -mindepth 1 -maxdepth 1 \\
! -name '.env' \\
! -name '.venv' \\
! -name 'run' \\
-exec rm -rf {{}} +
"""
run_command(["ssh", *ssh_args, "bash", "-se"], input_bytes=remote_prepare_script.encode("utf-8"), label="remote prepare")
def remote_script(*, repo_url: str, repo_ref: str, remote_dir: str, app_port: str, runtime_env_text: str) -> str:
# .env 文本改走 base64 下发,避免 ssh heredoc 里的转义和编码问题。
env_blob = base64.b64encode(runtime_env_text.encode("utf-8")).decode("ascii")
# 本地 tar 流直接送到远端解压,避免要求远端额外安装 rsync。
tar_command = [
"tar",
"-czf",
"-",
"--exclude=.git",
"--exclude=.env",
"--exclude=.venv",
"--exclude=run",
"--exclude=__pycache__",
"--exclude=.pytest_cache",
"--exclude=.mypy_cache",
"--exclude=.DS_Store",
".",
]
ssh_command = ["ssh", *ssh_args, "tar", "-xzf", "-", "-C", remote_dir]
# 远端固定执行git 拉取代码 -> 写入 .env -> 安装依赖 -> 重启 systemd -> 校验状态。
return f"""set -euo pipefail
try:
tar_process = subprocess.Popen(tar_command, cwd=str(ROOT), stdout=subprocess.PIPE, stderr=subprocess.PIPE)
except FileNotFoundError as exc:
raise SystemExit("missing command for tar sync: tar") from exc
try:
ssh_process = subprocess.run(
ssh_command,
stdin=tar_process.stdout,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT,
check=False,
)
if tar_process.stdout is not None:
tar_process.stdout.close()
tar_stderr = (tar_process.stderr.read() if tar_process.stderr is not None else b"").decode("utf-8", errors="replace")
tar_return = tar_process.wait()
finally:
if tar_process.poll() is None:
tar_process.kill()
tar_process.wait()
if tar_return != 0:
raise SystemExit(tar_stderr.strip() or "local tar packaging failed")
if ssh_process.returncode != 0:
raise SystemExit((ssh_process.stdout or b"").decode("utf-8", errors="replace").strip() or "remote extract failed")
def copy_env_file(
target: str,
*,
remote_dir: str,
env_path: Path,
port: int,
identity_file: str,
) -> None:
# 运行配置单独走 scp和代码同步解耦避免误把远端 .env 覆盖成空文件。
run_command(
["scp", *scp_base_args(port=port, identity_file=identity_file), str(env_path), f"{target}:{remote_dir}/.env"],
label="copy env",
)
def run_remote_install(*, remote_dir: str, app_port: str, ssh_args: list[str]) -> str:
# 远端只负责安装依赖、重启 systemd 并校验监听。
script = f"""set -euo pipefail
REPO_URL={shlex.quote(repo_url)}
REPO_REF={shlex.quote(repo_ref)}
REMOTE_DIR={shlex.quote(remote_dir)}
APP_PORT={shlex.quote(app_port)}
export GIT_SSH_COMMAND='ssh -o StrictHostKeyChecking=accept-new'
mkdir -p "$(dirname "$REMOTE_DIR")"
if [ -d "$REMOTE_DIR/.git" ]; then
git -C "$REMOTE_DIR" fetch --all --tags --prune
else
rm -rf "$REMOTE_DIR"
git clone "$REPO_URL" "$REMOTE_DIR"
git -C "$REMOTE_DIR" fetch --all --tags --prune
fi
if git -C "$REMOTE_DIR" show-ref --verify --quiet "refs/remotes/origin/$REPO_REF"; then
git -C "$REMOTE_DIR" checkout -B "$REPO_REF" "origin/$REPO_REF"
git -C "$REMOTE_DIR" branch --set-upstream-to="origin/$REPO_REF" "$REPO_REF" >/dev/null 2>&1 || true
elif git -C "$REMOTE_DIR" show-ref --verify --quiet "refs/tags/$REPO_REF"; then
git -C "$REMOTE_DIR" checkout --detach "refs/tags/$REPO_REF"
else
git -C "$REMOTE_DIR" checkout --detach "$REPO_REF"
fi
cd "$REMOTE_DIR"
python3 - <<'PY'
import base64
from pathlib import Path
# 本地项目根目录的 .env 必须同步到远端根目录,运行时只认这一份。
content = base64.b64decode({env_blob!r}).decode("utf-8")
Path(".env").write_text(content, encoding="utf-8")
PY
chmod +x scripts/*.sh
python3 -m venv .venv
.venv/bin/pip install -r requirements-ops.txt
.venv/bin/pip install --quiet --disable-pip-version-check -r requirements-ops.txt
bash scripts/install_systemd.sh
systemctl restart hy-app-monitor-yumi.service hy-app-monitor-haiyi.service hy-app-monitor.service
systemctl is-active hy-app-monitor-yumi.service
@ -213,10 +207,29 @@ systemctl is-active hy-app-monitor-haiyi.service
systemctl is-active hy-app-monitor.service
ss -lntp | grep ":$APP_PORT" || true
"""
def run_remote_deploy(
*,
repo_url: str,
repo_ref: str,
remote_dir: str,
app_port: str,
env_path: Path,
ssh_args: list[str],
) -> str:
# SSH 部署主逻辑和 TAT 保持一致,只是传输层从 TAT 换成 SSH。
script = remote_script(
repo_url=repo_url,
repo_ref=repo_ref,
remote_dir=remote_dir,
app_port=app_port,
runtime_env_text=env_path.read_text(encoding="utf-8"),
)
output = run_command(
["ssh", *ssh_args, "bash", "-se"],
input_bytes=script.encode("utf-8"),
label="remote install",
label="remote deploy",
)
return output.decode("utf-8", errors="replace").strip()
@ -227,18 +240,15 @@ def main() -> None:
raise SystemExit("deploy ssh host is required; set DEPLOY_SSH_HOST or pass --host")
env_path = require_env_file()
target = f"{args.user}@{args.host}"
ssh_args = ssh_base_args(args.host, user=args.user, port=args.port, identity_file=args.identity_file)
sync_project_tree(remote_dir=args.remote_dir, ssh_args=ssh_args)
copy_env_file(
target,
output = run_remote_deploy(
repo_url=args.repo_url,
repo_ref=args.repo_ref,
remote_dir=args.remote_dir,
app_port=args.app_port,
env_path=env_path,
port=args.port,
identity_file=args.identity_file,
ssh_args=ssh_args,
)
output = run_remote_install(remote_dir=args.remote_dir, app_port=args.app_port, ssh_args=ssh_args)
print(output)

View File

@ -3,6 +3,7 @@ from __future__ import annotations
import unittest
from unittest.mock import patch
from core.config import HARBOR_PROJECT, HARBOR_REGISTRY
from monitors.yumi.service_release import build_update_targets_payload
@ -90,6 +91,37 @@ class ServiceReleaseTargetsTests(unittest.TestCase):
self.assertEqual(payload["items"][0]["currentImages"], [])
self.assertEqual(payload["items"][0]["warnings"], ["app-1: broken compose"])
@patch("monitors.yumi.service_release.frontend_release_enabled", return_value=False)
@patch("monitors.yumi.service_release.buildable_service_names", return_value=["auth"])
@patch("monitors.yumi.service_release.service_records_by_name")
@patch("monitors.yumi.service_release.load_local_host_compose_model", side_effect=RuntimeError("broken compose"))
def test_build_update_targets_payload_falls_back_repository_when_compose_unavailable(
self,
load_local_host_compose_model_mock,
service_records_by_name_mock,
buildable_service_names_mock,
frontend_release_enabled_mock,
) -> None:
del load_local_host_compose_model_mock
del buildable_service_names_mock, frontend_release_enabled_mock
service_records_by_name_mock.return_value = {
"auth": [
{
"host": "app-1",
"ip": "10.0.0.1",
"instanceId": "ins-app-1",
"port": 8080,
"path": "/auth",
}
]
}
payload = build_update_targets_payload()
self.assertTrue(payload["ok"])
self.assertEqual(len(payload["items"]), 1)
self.assertEqual(payload["items"][0]["repository"], f"{HARBOR_REGISTRY}/{HARBOR_PROJECT}/java/chatapp3-auth")
if __name__ == "__main__":
unittest.main()