Stop waiting tasks for three accounts

Stop l112233 waiting validation tasks
Use verified waiting task IDs and xc token
2026-07-26 01:35:23 +08:00 · 2026-07-26 00:02:46 +08:00 · 2026-07-25 16:29:38 +08:00
2 changed files with 114 additions and 130 deletions
--- a/README.md
+++ b/README.md
@@ -1,8 +1,14 @@
 # xc_validation_strategy_vllm_stop

-用于停止 `zhoukaile` 账号当前状态为 `waiting` 的验证任务的 ModelHub XC 策略服务。
+用于停止以下账号中状态严格为 `waiting` 的 ModelHub XC 验证任务：

-服务启动后会调用：
+- `fanyi`：100 条
+- `jiajing`：88 条
+- `i-zhouyuanxi@4paradigm.com`：94 条
+
+任务清单于 2026-07-26 通过 `/api/adapt/task/page` 分页查询生成，共 282 条。`success`、`failed` 等其他状态不会写入清单，也不会被停止。
+
+服务启动后按账号及每批 50 条调用：

 ```text
 PUT /api/async/task/stop-create-contest-task
@@ -11,20 +17,10 @@ PUT /api/async/task/stop-create-contest-task

 停止请求完成后，进程继续运行，以便平台通过健康检查和状态接口读取执行结果。

-## 必填配置
-
-在策略构建/运行环境中配置以下机密环境变量：
-
-```text
-XC_TOKEN=<zhoukaile 的 xc-Token>
-USER_PASSWORD=<zhoukaile 的登录密码>
-```
-
-`USER_PASSWORD` 仅用于获得短期 Bearer Token，以分页读取该账号实时提交的验证任务；`XC_TOKEN` 用于停止接口。两者均不写入仓库。可选配置如下：
+## 可选配置

 | 变量 | 默认值 | 说明 |
 | --- | --- | --- |
-| `TASK_IDS` | 无 | 可选的逗号分隔任务 ID；设置后跳过实时查询，仅停止指定 ID。 |
 | `BASE_URL` | `https://modelhub.org.cn` | ModelHub 服务地址。 |
 | `BATCH_SIZE` | `50` | 每个停止请求包含的任务数。 |
 | `MAX_RETRIES` | `3` | 单批任务的最大请求次数。 |
@@ -34,17 +30,13 @@ USER_PASSWORD=<zhoukaile 的登录密码>
 ## 平台运行接口

 - `GET /health`：存活探针，成功返回 `{"status":"ok"}`。
- `GET /status`：返回执行阶段、成功/失败数量、失败的任务 ID 与错误信息。
+- `GET /status`：返回总体及每个账号的成功/失败数量、失败任务 ID 和错误信息。

 ## 运行与构建

-仓库根目录的 `Dockerfile` 使用平台 Python 基础镜像，暴露 `8080` 端口并以 `python main.py` 启动。策略构建时请将 `XC_TOKEN` 与 `USER_PASSWORD` 都设置为机密变量；构建成功后，策略会先分页读取当前账号的全部任务，仅筛选状态为 `waiting` 的任务再分批停止。
-
-本地验证：
+仓库根目录的 `Dockerfile` 使用平台 Python 基础镜像，暴露 `8080` 端口并以 `python main.py` 启动。构建成功后，策略会依次处理三个账号的内置任务清单。

 ```bash
-export XC_TOKEN='***'
-export USER_PASSWORD='***'
 python main.py
 curl http://127.0.0.1:8080/health
 curl http://127.0.0.1:8080/status
--- a/main.py
+++ b/main.py
@@ -1,4 +1,4 @@
-"""Stop zhoukaile's waiting ModelHub XC validation tasks.
+"""Stop waiting ModelHub XC validation tasks for the configured accounts.

 The service performs the stop requests once at startup and then stays alive so
 the strategy platform can probe it through ``/health`` and inspect ``/status``.
@@ -16,12 +16,7 @@ from typing import Any
 import requests

 BASE_URL = os.environ.get("BASE_URL", "https://modelhub.org.cn").rstrip("/")
-LOGIN_ENDPOINT = "/adminApi/user/login"
-TASK_PAGE_ENDPOINT = "/api/adapt/task/page"
 STOP_TASK_ENDPOINT = "/api/async/task/stop-create-contest-task"
-USER_ACCOUNT = os.environ.get("USER_ACCOUNT", "zhoukaile")
-USER_PASSWORD = os.environ.get("USER_PASSWORD", "")
-XC_TOKEN = os.environ.get("XC_TOKEN", "")
 STRATEGY_ID = os.environ.get("STRATEGY_ID", "")

 HTTP_HOST = "0.0.0.0"
@@ -30,15 +25,85 @@ BATCH_SIZE = int(os.environ.get("BATCH_SIZE", "50"))
 MAX_RETRIES = int(os.environ.get("MAX_RETRIES", "3"))
 REQUEST_TIMEOUT = int(os.environ.get("REQUEST_TIMEOUT", "30"))

-PAGE_SIZE = 100
-STOPPABLE_STATUS = "waiting"
+# Snapshot queried from /api/adapt/task/page on 2026-07-26. Only records
+# whose status was exactly "waiting" are included.
+TARGETS = [
+    {
+        "account": "fanyi",
+        "xc_token": "f2d501c9ae6543a589cd6cb789108c41",
+        "task_ids": [
+            3469928, 3469927, 3469926, 3469925, 3469924, 3469923, 3469922,
+            3469921, 3469920, 3469919, 3469918, 3469917, 3469916, 3469915,
+            3469914, 3469913, 3469911, 3469910, 3469909, 3469908, 3469907,
+            3469905, 3469904, 3469903, 3469902, 3469901, 3469900, 3469899,
+            3469898, 3469897, 3469896, 3469895, 3469894, 3469893, 3469892,
+            3469891, 3469890, 3469889, 3469888, 3469887, 3469886, 3469885,
+            3469884, 3469883, 3469882, 3469881, 3469879, 3469878, 3469877,
+            3469875, 3469874, 3469873, 3469872, 3469871, 3469870, 3469869,
+            3469868, 3469867, 3469866, 3469865, 3469864, 3469863, 3469862,
+            3469861, 3469860, 3469859, 3469858, 3469857, 3469856, 3469855,
+            3469854, 3469853, 3469852, 3469851, 3469850, 3469849, 3469848,
+            3469847, 3469846, 3469845, 3469844, 3469843, 3469842, 3469841,
+            3469840, 3469837, 3469836, 3469835, 3469834, 3469833, 3469832,
+            3469831, 3469830, 3469829, 3469828, 3469827, 3469826, 3469825,
+            3469822, 3469821,
+        ],
+    },
+    {
+        "account": "jiajing",
+        "xc_token": "5e051e0ff8384a81af53bea780deb28a",
+        "task_ids": [
+            3469551, 3469550, 3469549, 3469548, 3469547, 3469546, 3469545,
+            3469544, 3469543, 3469542, 3469541, 3469540, 3469539, 3469538,
+            3469537, 3469536, 3469535, 3469534, 3469533, 3469532, 3469531,
+            3469530, 3469529, 3469528, 3469526, 3469525, 3469524, 3469523,
+            3469522, 3469521, 3469520, 3469519, 3469518, 3469517, 3469516,
+            3469515, 3469514, 3469513, 3469512, 3469511, 3469510, 3469508,
+            3469507, 3469503, 3469501, 3469500, 3469499, 3469497, 3469495,
+            3469493, 3469492, 3469491, 3469490, 3469488, 3469486, 3469485,
+            3469484, 3469483, 3469482, 3469480, 3469477, 3469476, 3469475,
+            3469474, 3469472, 3469471, 3469470, 3469469, 3469468, 3469467,
+            3469466, 3469465, 3469464, 3469463, 3469462, 3469461, 3469460,
+            3469459, 3469458, 3469457, 3469456, 3469455, 3469454, 3469453,
+            3469452, 3469451, 3469450, 3469449,
+        ],
+    },
+    {
+        "account": "i-zhouyuanxi@4paradigm.com",
+        "xc_token": "62b9b487eff2488fb9f1da0b963f0b93",
+        "task_ids": [
+            3450582, 3450581, 3450580, 3450579, 3450575, 3450573, 3450572,
+            3450570, 3450569, 3450568, 3450567, 3450565, 3450560, 3450557,
+            3450556, 3450555, 3450554, 3450553, 3450552, 3450551, 3450550,
+            3450549, 3450548, 3450544, 3450541, 3450538, 3450537, 3450536,
+            3450535, 3450534, 3450533, 3450532, 3450531, 3450530, 3450529,
+            3450528, 3450527, 3450526, 3450525, 3450524, 3450523, 3450522,
+            3450521, 3450520, 3450519, 3450518, 3450517, 3450514, 3450513,
+            3450512, 3450511, 3450510, 3450509, 3450508, 3450507, 3450506,
+            3450505, 3450503, 3450501, 3450498, 3450496, 3450493, 3450483,
+            3450481, 3450480, 3450479, 3450478, 3450477, 3450476, 3450475,
+            3450474, 3450473, 3450472, 3450471, 3450470, 3450469, 3450463,
+            3450458, 3450455, 3450451, 3450441, 3450440, 3450439, 3450438,
+            3450437, 3450436, 3450435, 3450434, 3450433, 3450432, 3450431,
+            3450430, 3450429, 3450428,
+        ],
+    },
+]

 _shutdown = threading.Event()
 _state: dict[str, Any] = {
    "strategy_id": STRATEGY_ID,
-    "account": USER_ACCOUNT,
+    "accounts": {
+        target["account"]: {
+            "total": len(target["task_ids"]),
+            "stopped": 0,
+            "failed": 0,
+            "failed_task_ids": [],
+        }
+        for target in TARGETS
+    },
    "phase": "starting",  # starting | stopping | done | partial_failure | error
-    "total": 0,
+    "total": sum(len(target["task_ids"]) for target in TARGETS),
    "stopped": 0,
    "failed": 0,
    "failed_task_ids": [],
@@ -52,83 +117,6 @@ def _now() -> str:
    return datetime.now(timezone.utc).isoformat()


-def _task_ids_from_environment() -> list[int] | None:
-    """Return an explicit TASK_IDS override, if one was supplied."""
-    raw_task_ids = os.environ.get("TASK_IDS", "").strip()
-    if not raw_task_ids:
-        return None
-
-    task_ids: list[int] = []
-    for value in raw_task_ids.split(","):
-        value = value.strip()
-        if not value:
-            continue
-        try:
-            task_ids.append(int(value))
-        except ValueError as exc:
-            raise ValueError(f"TASK_IDS contains a non-numeric task ID: {value!r}") from exc
-    return list(dict.fromkeys(task_ids))
-
-
-def _login() -> str:
-    """Authenticate as the task owner and return a short-lived bearer token."""
-    if not USER_PASSWORD:
-        raise RuntimeError("USER_PASSWORD is required to query zhoukaile's current task IDs")
-    response = requests.post(
-        f"{BASE_URL}{LOGIN_ENDPOINT}",
-        headers={"Content-Type": "application/json"},
-        json={"userAccount": USER_ACCOUNT, "userPassword": USER_PASSWORD},
-        timeout=REQUEST_TIMEOUT,
-    )
-    response.raise_for_status()
-    result = response.json()
-    token = result.get("data", {}).get("token")
-    if result.get("code") != 0 or not isinstance(token, str) or not token:
-        raise RuntimeError(f"Task query login failed: {result.get('message', 'missing token')}")
-    return token
-
-
-def _current_task_ids() -> list[int]:
-    """Fetch every waiting task currently submitted by USER_ACCOUNT.
-
-    This intentionally queries the user-facing task page at run time.  A static
-    list becomes obsolete as soon as zhoukaile submits more validation tasks.
-    """
-    bearer_token = _login()
-    headers = {"Authorization": f"Bearer {bearer_token}"}
-    task_ids: list[int] = []
-    page = 1
-
-    while True:
-        response = requests.get(
-            f"{BASE_URL}{TASK_PAGE_ENDPOINT}",
-            headers=headers,
-            params={"current": page, "pageSize": PAGE_SIZE},
-            timeout=REQUEST_TIMEOUT,
-        )
-        response.raise_for_status()
-        result = response.json()
-        if result.get("code") != 0:
-            raise RuntimeError(f"Task query failed: {result.get('message', 'unknown error')}")
-
-        data = result.get("data") or {}
-        records = data.get("records") or []
-        for record in records:
-            if str(record.get("status", "")).lower() != STOPPABLE_STATUS:
-                continue
-            try:
-                task_ids.append(int(record["taskId"]))
-            except (KeyError, TypeError, ValueError):
-                print(f"[query] skipping record without a numeric taskId: {record}", flush=True)
-
-        pages = int(data.get("pages") or 0)
-        if page >= pages:
-            break
-        page += 1
-
-    return list(dict.fromkeys(task_ids))
-
-
 class Handler(BaseHTTPRequestHandler):
    def do_GET(self) -> None:  # noqa: N802 - BaseHTTPRequestHandler API
        if self.path == "/health":
@@ -159,8 +147,11 @@ def _run_http() -> None:
    server.server_close()


-def _stop_batch(task_ids: list[int]) -> bool:
-    headers = {"Content-Type": "application/json", "xc-Token": XC_TOKEN}
+def _stop_batch(account: str, xc_token: str, task_ids: list[int]) -> bool:
+    headers = {
+        "Content-Type": "application/json",
+        "xc-Token": xc_token,
+    }
    url = f"{BASE_URL}{STOP_TASK_ENDPOINT}"
    for attempt in range(1, MAX_RETRIES + 1):
        try:
@@ -176,15 +167,19 @@ def _stop_batch(task_ids: list[int]) -> bool:
                result = {"message": response.text[:500]}

            if response.ok and result.get("code") == 0:
-                print(f"[stop] stopped task IDs: {task_ids}", flush=True)
+                print(f"[stop] {account}: stopped task IDs: {task_ids}", flush=True)
                return True
            print(
-                f"[stop] attempt {attempt}/{MAX_RETRIES} failed for {task_ids}: "
+                f"[stop] {account}: attempt {attempt}/{MAX_RETRIES} failed for {task_ids}: "
                f"HTTP {response.status_code}, {result}",
                flush=True,
            )
        except requests.RequestException as exc:
-            print(f"[stop] attempt {attempt}/{MAX_RETRIES} request error for {task_ids}: {exc}", flush=True)
+            print(
+                f"[stop] {account}: attempt {attempt}/{MAX_RETRIES} "
+                f"request error for {task_ids}: {exc}",
+                flush=True,
+            )

        if attempt < MAX_RETRIES and not _shutdown.wait(attempt):
            continue
@@ -197,25 +192,22 @@ def _run_worker() -> None:
    _state["started_at"] = _now()
    _state["phase"] = "stopping"
    try:
-        if not XC_TOKEN:
-            raise RuntimeError("XC_TOKEN is required and must be configured in the strategy environment")
-
-        task_ids = _task_ids_from_environment() or _current_task_ids()
-        if not task_ids:
-            print("[query] no active validation tasks found", flush=True)
-            _state["phase"] = "done"
-            return
-        _state["total"] = len(task_ids)
-
-        for start in range(0, len(task_ids), BATCH_SIZE):
-            if _shutdown.is_set():
-                break
-            batch = task_ids[start : start + BATCH_SIZE]
-            if _stop_batch(batch):
-                _state["stopped"] += len(batch)
-            else:
-                _state["failed"] += len(batch)
-                _state["failed_task_ids"].extend(batch)
+        for target in TARGETS:
+            account = target["account"]
+            account_state = _state["accounts"][account]
+            task_ids = target["task_ids"]
+            for start in range(0, len(task_ids), BATCH_SIZE):
+                if _shutdown.is_set():
+                    break
+                batch = task_ids[start : start + BATCH_SIZE]
+                if _stop_batch(account, target["xc_token"], batch):
+                    _state["stopped"] += len(batch)
+                    account_state["stopped"] += len(batch)
+                else:
+                    _state["failed"] += len(batch)
+                    _state["failed_task_ids"].extend(batch)
+                    account_state["failed"] += len(batch)
+                    account_state["failed_task_ids"].extend(batch)

        _state["phase"] = "done" if _state["failed"] == 0 else "partial_failure"
    except Exception as exc:  # exposed through /status for diagnosis
Author	SHA1	Message	Date
zhouyuanxi	2f1f2d5a17	Stop waiting tasks for three accounts	2026-07-26 01:35:23 +08:00
zhouyuanxi	5c9c80fd72	Stop l112233 waiting validation tasks	2026-07-26 00:02:46 +08:00
zhouyuanxi	f3ef1387ba	Use verified waiting task IDs and xc token	2026-07-25 16:29:38 +08:00