feat(netbird): auto-renew NetBird client-update API tokens before expiry

The tokens captured for central update control expire (NetBird enforces a
365-day max on Personal Access Tokens), and nothing was renewing them —
discovered that the 47 tokens created via the browser-automation bulk
onboarding were actually only 30-day tokens (left the UI's default
expiration field untouched instead of setting 365), so they would have
silently broken automatic-update control next month with no warning.

- Bumped all existing tokens to fresh 365-day ones via the API (using the
  still-valid old token as bearer — no re-login needed)
- Added netbird_api_token_renewed_at per deployment
- Scheduler now checks daily and renews any token older than 300 days
  automatically, so this never needs to be done by hand again

app/database.py

@@ -129,6 +129,7 @@ def _run_migrations() -> None:
         ("system_config", "auto_update_last_run_at", "TEXT"),
         # NetBird client (peer) automatic-updates master default + per-customer PAT
         ("deployments", "netbird_api_token_encrypted", "TEXT"),
+        ("deployments", "netbird_api_token_renewed_at", "TEXT"),
         ("system_config", "netbird_client_auto_update_version", "TEXT DEFAULT 'disabled'"),
         ("system_config", "netbird_client_auto_update_always", "BOOLEAN DEFAULT 0"),
     ]

app/models.py

@@ -89,6 +89,7 @@ class Deployment(Base):
     netbird_admin_email: Mapped[Optional[str]] = mapped_column(Text, nullable=True)
     netbird_admin_password: Mapped[Optional[str]] = mapped_column(Text, nullable=True)
     netbird_api_token_encrypted: Mapped[Optional[str]] = mapped_column(Text, nullable=True)
+    netbird_api_token_renewed_at: Mapped[Optional[datetime]] = mapped_column(DateTime, nullable=True)
     deployment_status: Mapped[str] = mapped_column(
         String(20), default="pending", nullable=False
     )
@@ -118,6 +119,9 @@ class Deployment(Base):
             "setup_url": self.setup_url,
             "has_credentials": bool(self.netbird_admin_email and self.netbird_admin_password),
             "has_netbird_api_token": bool(self.netbird_api_token_encrypted),
+            "netbird_api_token_renewed_at": (
+                self.netbird_api_token_renewed_at.isoformat() if self.netbird_api_token_renewed_at else None
+            ),
             "deployment_status": self.deployment_status,
             "deployed_at": self.deployed_at.isoformat() if self.deployed_at else None,
             "last_health_check": (

app/routers/deployments.py

@@ -1,6 +1,7 @@
 """Deployment management API — start, stop, restart, logs, health for customers."""
 
 import logging
+from datetime import datetime
 
 from fastapi import APIRouter, BackgroundTasks, Depends, HTTPException, Query, status
 from sqlalchemy.orm import Session
@@ -362,6 +363,7 @@ async def set_customer_netbird_api_token(
         )
 
     deployment.netbird_api_token_encrypted = encrypt_value(payload.token)
+    deployment.netbird_api_token_renewed_at = datetime.utcnow()
     db.commit()
     logger.info("NetBird API token registered for customer %d by %s.", customer_id, current_user.username)
     return {"ok": True}

app/services/netbird_client_update_service.py

@@ -52,6 +52,48 @@ async def get_current_settings(container_prefix: str, token: str) -> dict[str, A
         return {"ok": False, "error": str(exc)}
 
 
+async def renew_token(container_prefix: str, token: str) -> dict[str, Any]:
+    """Mint a fresh 365-day Personal Access Token using the current one.
+
+    The old token is left in place (it naturally expires and NetBird gives no
+    reliable way to identify "our" token among a user's other PATs by content
+    alone, only by name — which isn't safe to assume is unique). One
+    harmless unused token lingering until its own expiry is an acceptable
+    trade-off for not risking deleting a token that turns out to be in use.
+    """
+    base_url = _base_url(container_prefix)
+    headers = {"Authorization": f"Token {token}"}
+    try:
+        async with httpx.AsyncClient(timeout=_TIMEOUT) as client:
+            resp = await client.get(f"{base_url}/api/users", headers=headers)
+            if resp.status_code != 200:
+                return {"ok": False, "error": f"GET /api/users -> HTTP {resp.status_code}"}
+            users = resp.json()
+            me = next((u for u in users if u.get("is_current")), None)
+            if not me:
+                return {"ok": False, "error": "Could not identify current user from /api/users."}
+
+            create_resp = await client.post(
+                f"{base_url}/api/users/{me['id']}/tokens",
+                headers={**headers, "Content-Type": "application/json"},
+                content=json.dumps({"name": "MSP Central Management", "expires_in": 365}),
+            )
+            if create_resp.status_code not in (200, 201):
+                return {"ok": False, "error": f"POST tokens -> HTTP {create_resp.status_code}: {create_resp.text[:200]}"}
+            new_token = create_resp.json().get("plain_token")
+            if not new_token:
+                return {"ok": False, "error": "Token creation response had no plain_token."}
+
+            verify_resp = await client.get(f"{base_url}/api/accounts", headers={"Authorization": f"Token {new_token}"})
+            if verify_resp.status_code != 200:
+                return {"ok": False, "error": "New token failed verification."}
+
+            return {"ok": True, "token": new_token}
+    except Exception as exc:
+        logger.warning("Failed to renew NetBird API token for %s: %s", container_prefix, exc)
+        return {"ok": False, "error": str(exc)}
+
+
 async def push_auto_update_settings(
     container_prefix: str, token: str, version: str, always: bool
 ) -> dict[str, Any]:

app/services/netbird_service.py

@@ -348,9 +348,9 @@ async def deploy_customer(db: Session, customer_id: int) -> dict[str, Any]:
             deployment.setup_url = setup_url
             deployment.netbird_admin_email = encrypt_value(admin_email) if setup_ok else deployment.netbird_admin_email
             deployment.netbird_admin_password = encrypt_value(admin_password) if setup_ok else deployment.netbird_admin_password
-            deployment.netbird_api_token_encrypted = (
-                encrypt_value(netbird_api_token) if netbird_api_token else deployment.netbird_api_token_encrypted
-            )
+            if netbird_api_token:
+                deployment.netbird_api_token_encrypted = encrypt_value(netbird_api_token)
+                deployment.netbird_api_token_renewed_at = datetime.utcnow()
             deployment.deployment_status = "running"
             deployment.deployed_at = datetime.utcnow()
         else:
@@ -366,6 +366,7 @@ async def deploy_customer(db: Session, customer_id: int) -> dict[str, Any]:
                 netbird_admin_email=encrypt_value(admin_email) if setup_ok else None,
                 netbird_admin_password=encrypt_value(admin_password) if setup_ok else None,
                 netbird_api_token_encrypted=encrypt_value(netbird_api_token) if netbird_api_token else None,
+                netbird_api_token_renewed_at=datetime.utcnow() if netbird_api_token else None,
                 deployment_status="running",
                 deployed_at=datetime.utcnow(),
             )

app/services/scheduler_service.py

@@ -8,17 +8,23 @@ SystemConfig.auto_update_check_enabled / auto_update_check_time.
 
 import asyncio
 import logging
-from datetime import datetime
+from datetime import datetime, timedelta
 
 from app.database import SessionLocal
 from app.models import Deployment, SystemConfig
-from app.services import image_service
+from app.services import image_service, netbird_client_update_service
+from app.utils.security import decrypt_value, encrypt_value
 
 logger = logging.getLogger(__name__)
 
 _POLL_INTERVAL_SECONDS = 60
 _task: asyncio.Task | None = None
 
+# NetBird PATs we mint are issued for 365 days; renew well before that so a
+# missed tick or a slow rollout never risks the token actually expiring.
+_TOKEN_RENEW_AFTER_DAYS = 300
+_TOKEN_RENEW_CHECK_HOUR = 4  # run once per day, distinct from the image-check hour
+
 
 def start() -> None:
     """Start the background polling task. Safe to call once at app startup."""
@@ -36,15 +42,64 @@ def stop() -> None:
         _task = None
 
 
+_last_token_renewal_date = None
+
+
 async def _poll_loop() -> None:
     while True:
         try:
             await _tick()
         except Exception:
             logger.exception("Scheduler tick failed")
+        try:
+            await _token_renewal_tick()
+        except Exception:
+            logger.exception("Token renewal tick failed")
         await asyncio.sleep(_POLL_INTERVAL_SECONDS)
 
 
+async def _token_renewal_tick() -> None:
+    """Once a day, renew any NetBird client-update API token nearing its
+    365-day expiry — keeps central update control working indefinitely
+    without anyone needing to notice or act.
+    """
+    global _last_token_renewal_date
+    now = datetime.now()
+    if now.hour != _TOKEN_RENEW_CHECK_HOUR:
+        return
+    if _last_token_renewal_date == now.date():
+        return
+    _last_token_renewal_date = now.date()
+
+    db = SessionLocal()
+    try:
+        cutoff = now - timedelta(days=_TOKEN_RENEW_AFTER_DAYS)
+        deployments = (
+            db.query(Deployment)
+            .filter(Deployment.netbird_api_token_encrypted.isnot(None))
+            .all()
+        )
+        due = [
+            d for d in deployments
+            if d.netbird_api_token_renewed_at is None or d.netbird_api_token_renewed_at < cutoff
+        ]
+        if not due:
+            return
+        logger.info("Renewing NetBird API token for %d customer(s)...", len(due))
+        for d in due:
+            token = decrypt_value(d.netbird_api_token_encrypted)
+            result = await netbird_client_update_service.renew_token(d.container_prefix, token)
+            if result["ok"]:
+                d.netbird_api_token_encrypted = encrypt_value(result["token"])
+                d.netbird_api_token_renewed_at = now
+                db.commit()
+                logger.info("Renewed NetBird API token for %s.", d.container_prefix)
+            else:
+                logger.warning("Failed to renew NetBird API token for %s: %s", d.container_prefix, result.get("error"))
+    finally:
+        db.close()
+
+
 async def _tick() -> None:
     db = SessionLocal()
     try: