Add ping and TCP monitor types

Adds ping and TCP monitor creation APIs, worker collectors, network checks UI, dashboard monitor status support, and progress documentation.

backend/app/api/monitors.py

@@ -7,7 +7,7 @@ from sqlalchemy.orm import Session
 from app.auth.dependencies import get_current_user, require_role
 from app.db.session import get_db
 from app.models import AlertRule, Asset, CheckResult, Incident, Monitor, User
-from app.schemas.core import CheckResultRead, MonitorCreate, MonitorRead, MonitorUpdate, WebsiteMonitorCreate
+from app.schemas.core import CheckResultRead, MonitorCreate, MonitorRead, MonitorUpdate, PingMonitorCreate, TcpMonitorCreate, WebsiteMonitorCreate
 
 router = APIRouter(prefix="/monitors", tags=["monitors"])
 
@@ -80,6 +80,93 @@ def create_website_monitor(
     return monitor
 
 
+@router.post("/ping", response_model=MonitorRead)
+def create_ping_monitor(
+    payload: PingMonitorCreate,
+    _: User = Depends(require_role("admin")),
+    db: Session = Depends(get_db),
+) -> Monitor:
+    asset_id: int | None = None
+    if payload.create_asset:
+        asset = Asset(name=payload.name, asset_type="host", address=payload.host, status="unknown", extra={})
+        db.add(asset)
+        db.flush()
+        asset_id = asset.id
+
+    monitor = Monitor(
+        asset_id=asset_id,
+        name=payload.name,
+        monitor_type="ping",
+        target=payload.host,
+        config={"timeout_seconds": payload.timeout_seconds},
+        interval_seconds=payload.interval_seconds,
+        status="unknown",
+    )
+    db.add(monitor)
+    db.flush()
+
+    if payload.alert_enabled:
+        db.add(
+            AlertRule(
+                monitor_id=monitor.id,
+                name=f"{payload.name} ping failure",
+                severity=payload.alert_severity,
+                condition={"type": "status_not_up"},
+                failure_threshold=payload.failure_threshold,
+                cooldown_seconds=300,
+                is_enabled=True,
+            )
+        )
+
+    db.commit()
+    db.refresh(monitor)
+    return monitor
+
+
+@router.post("/tcp", response_model=MonitorRead)
+def create_tcp_monitor(
+    payload: TcpMonitorCreate,
+    _: User = Depends(require_role("admin")),
+    db: Session = Depends(get_db),
+) -> Monitor:
+    asset_id: int | None = None
+    target = f"{payload.host}:{payload.port}"
+    if payload.create_asset:
+        asset = Asset(name=payload.name, asset_type="tcp_service", address=target, status="unknown", extra={})
+        db.add(asset)
+        db.flush()
+        asset_id = asset.id
+
+    monitor = Monitor(
+        asset_id=asset_id,
+        name=payload.name,
+        monitor_type="tcp",
+        target=target,
+        config={"host": payload.host, "port": payload.port, "timeout_seconds": payload.timeout_seconds},
+        interval_seconds=payload.interval_seconds,
+        status="unknown",
+    )
+    db.add(monitor)
+    db.flush()
+
+    if payload.alert_enabled:
+        db.add(
+            AlertRule(
+                monitor_id=monitor.id,
+                name=f"{payload.name} TCP connection failure",
+                severity=payload.alert_severity,
+                condition={"type": "status_not_up"},
+                failure_threshold=payload.failure_threshold,
+                cooldown_seconds=300,
+                is_enabled=True,
+            )
+        )
+
+    db.commit()
+    db.refresh(monitor)
+    return monitor
+
+
 @router.get("/{monitor_id}", response_model=MonitorRead)
 def get_monitor(monitor_id: int, _: User = Depends(get_current_user), db: Session = Depends(get_db)) -> Monitor:
     monitor = db.get(Monitor, monitor_id)
@@ -110,7 +197,7 @@ def update_monitor(
 @router.delete("/{monitor_id}", status_code=204)
 def delete_monitor(
     monitor_id: int,
-    cleanup_orphan_website_asset: bool = True,
+    cleanup_orphan_asset: bool = True,
     _: User = Depends(require_role("admin")),
     db: Session = Depends(get_db),
 ) -> None:
@@ -129,10 +216,10 @@ def delete_monitor(
     db.delete(monitor)
     db.flush()
 
-    if cleanup_orphan_website_asset and asset_id is not None:
+    if cleanup_orphan_asset and asset_id is not None:
         remaining = db.scalar(select(func.count(Monitor.id)).where(Monitor.asset_id == asset_id))
         asset = db.get(Asset, asset_id)
-        if remaining == 0 and asset is not None and asset.asset_type == "website":
+        if remaining == 0 and asset is not None and asset.asset_type in {"website", "host", "tcp_service"}:
             db.delete(asset)
 
     db.commit()

backend/app/schemas/core.py

@@ -73,6 +73,29 @@ class WebsiteMonitorCreate(BaseModel):
     failure_threshold: int = Field(default=3, ge=1, le=20)
 
 
+class PingMonitorCreate(BaseModel):
+    name: str = Field(min_length=1, max_length=160)
+    host: str = Field(min_length=1, max_length=255)
+    timeout_seconds: int = Field(default=5, ge=1, le=60)
+    interval_seconds: int = Field(default=60, ge=10)
+    create_asset: bool = True
+    alert_enabled: bool = True
+    alert_severity: str = "warning"
+    failure_threshold: int = Field(default=3, ge=1, le=20)
+
+
+class TcpMonitorCreate(BaseModel):
+    name: str = Field(min_length=1, max_length=160)
+    host: str = Field(min_length=1, max_length=255)
+    port: int = Field(ge=1, le=65535)
+    timeout_seconds: int = Field(default=5, ge=1, le=60)
+    interval_seconds: int = Field(default=60, ge=10)
+    create_asset: bool = True
+    alert_enabled: bool = True
+    alert_severity: str = "warning"
+    failure_threshold: int = Field(default=3, ge=1, le=20)
+
+
 class CheckResultRead(BaseModel):
     id: int
     monitor_id: int