modelTesting/webhook.py

#!/usr/bin/env python3
"""
Gitea push webhook receiver for the benchmark dashboard.

- Listens on 0.0.0.0:PORT (Gitea posts to /hook via the dashboard nginx proxy,
  or directly to http://10.0.0.22:PORT/hook)
- Auth: requires Authorization header (Bearer token) matching WEBHOOK_AUTH_TOKEN
- Signature: validates X-Gitea-Signature (HMAC-SHA256 of body) against WEBHOOK_SECRET
- On a valid push to `main`, runs: git fetch + reset to origin/main + ./deploy.sh up
- One concurrent deploy at a time (lock prevents overlapping rebuilds)

Security: no request data reaches the shell. The only shell string is a hardcoded
script (cd here, git fetch/reset, deploy.sh up). Ref validated == refs/heads/main.
Run via systemd unit llm-bench-webhook.service (see deploy-webhook.sh).
"""
import asyncio, hmac, hashlib, os, json, logging

HERE = os.path.dirname(os.path.abspath(__file__))
PORT = int(os.environ.get("WEBHOOK_PORT", "41798"))
SECRET = os.environ.get("WEBHOOK_SECRET", "").encode()
# Bearer token Gitea sends in the "Authorization Header" webhook field.
AUTH_TOKEN = os.environ.get("WEBHOOK_AUTH_TOKEN", "").strip()
REF_FILTER = os.environ.get("WEBHOOK_REF", "refs/heads/main")
MAX_BODY = 2 * 1024 * 1024

logging.basicConfig(level=logging.INFO, format="%(asctime)s %(levelname)s %(message)s")
log = logging.getLogger("webhook")
_deploy_lock = asyncio.Lock()

_DEPLOY_CMD = "cd " + HERE + " && git fetch origin && git reset --hard origin/main && ./deploy.sh up"


def check_auth(auth_header: str) -> bool:
    if not AUTH_TOKEN:
        log.warning("WEBHOOK_AUTH_TOKEN not set — accepting WITHOUT auth header (dev only)")
        return True
    if not auth_header:
        return False
    token = auth_header[7:].strip() if auth_header.lower().startswith("bearer ") else auth_header.strip()
    return hmac.compare_digest(token, AUTH_TOKEN)


def verify(signature_hex: str, body: bytes) -> bool:
    if not SECRET:
        log.warning("WEBHOOK_SECRET not set — accepting WITHOUT signature check (dev only)")
        return True
    if not signature_hex:
        return False
    return hmac.compare_digest(hmac.new(SECRET, body, hashlib.sha256).hexdigest(), signature_hex)


async def redeploy():
    if _deploy_lock.locked():
        log.info("deploy already running, skipping"); return
    async with _deploy_lock:
        log.info("starting redeploy")
        proc = await asyncio.create_subprocess_exec(
            "bash", "-lc", _DEPLOY_CMD,
            stdout=asyncio.subprocess.PIPE, stderr=asyncio.subprocess.STDOUT)
        out, _ = await proc.communicate()
        log.info("redeploy exit=%s\n%s", proc.returncode, (out or b"").decode(errors="replace"))


async def handle(reader, writer):
    try:
        req = await reader.readuntil(b"\r\n\r\n")
    except Exception:
        writer.close(); return
    try:
        head, _, body_start = req.partition(b"\r\n\r\n")
        lines = head.decode("latin1").split("\r\n")
        method, path, _ = lines[0].split(" ", 2)
        headers = {}
        for ln in lines[1:]:
            if ":" in ln:
                k, v = ln.split(":", 1); headers[k.strip().lower()] = v.strip()
        cl = int(headers.get("content-length", "0") or 0)
        body = body_start
        while len(body) < cl and len(body) < MAX_BODY:
            chunk = await reader.read(min(65536, cl - len(body)))
            if not chunk: break
            body += chunk

        if path.split("?")[0] not in ("/hook", "/webhook"):
            writer.write(b"HTTP/1.1 404 Not Found\r\nContent-Length: 0\r\n\r\n"); await writer.drain(); return
        if method != "POST":
            writer.write(b"HTTP/1.1 405 Method Not Allowed\r\nContent-Length: 0\r\n\r\n"); await writer.drain(); return

        # 1) Authorization bearer token (checked FIRST).
        if not check_auth(headers.get("authorization", "")):
            log.warning("bad/missing Authorization from %s", writer.get_extra_info("peername"))
            writer.write(b"HTTP/1.1 401 Unauthorized\r\nContent-Length: 0\r\n\r\n"); await writer.drain(); return
        # 2) HMAC signature of body.
        if not verify(headers.get("x-gitea-signature", ""), body):
            log.warning("bad signature from %s", writer.get_extra_info("peername"))
            writer.write(b"HTTP/1.1 403 Forbidden\r\nContent-Length: 0\r\n\r\n"); await writer.drain(); return
        # 3) Only pushes to the watched ref.
        try:
            ref = json.loads(body).get("ref", "") if body else ""
        except Exception:
            ref = ""
        if ref and ref != REF_FILTER:
            writer.write(b'HTTP/1.1 200 OK\r\nContent-Type: text/plain\r\nContent-Length: 17\r\n\r\nignored: wrong ref')
            await writer.drain(); return

        asyncio.create_task(redeploy())
        writer.write(b'HTTP/1.1 200 OK\r\nContent-Type: text/plain\r\nContent-Length: 12\r\n\r\ndeploying...\n')
        await writer.drain()
    finally:
        writer.close()


async def main():
    server = await asyncio.start_server(handle, "0.0.0.0", PORT)
    log.info("webhook receiver listening on 0.0.0.0:%d (filter=%s auth=%s)", PORT, REF_FILTER, bool(AUTH_TOKEN))
    async with server:
        await server.serve_forever()

if __name__ == "__main__":
    asyncio.run(main())