add service checks REST API, GraphQL API, and ping executor (#52)

- REST CRUD at /api/v1/checks for HTTP, TCP, DNS, ping checks
- GraphQL queries (checks, check) and mutations (create/update/delete)
- ping executor using system ping with macOS/Linux parsing
- check config validation for ping type (requires host)
- API documentation updated with checks resource section

Reviewed-on: graham/towerops-web#52

CHANGELOG.txt

@@ -1,3 +1,21 @@
+2026-03-16
+feat: add service checks REST API, GraphQL API, and ping executor
+  - New REST API: full CRUD at /api/v1/checks for HTTP, TCP, DNS, and ping checks
+  - New GraphQL API: checks/check queries and createCheck/updateCheck/deleteCheck mutations
+  - New ping executor: replaces stub with real ICMP ping via system ping command
+  - Ping executor parses both macOS and Linux output for packet loss and RTT
+  - Check config validation added for ping type (requires "host")
+  - API docs updated with Checks resource section
+  Files: lib/towerops/monitoring/executors/ping_executor.ex (new),
+         lib/towerops/workers/check_executor_worker.ex,
+         lib/towerops/monitoring/check.ex,
+         lib/towerops_web/controllers/api/v1/checks_controller.ex (new),
+         lib/towerops_web/router.ex,
+         lib/towerops_web/graphql/types/check.ex (new),
+         lib/towerops_web/graphql/resolvers/check.ex (new),
+         lib/towerops_web/graphql/schema.ex,
+         lib/towerops_web/controllers/api_docs_html/index.html.heex
+
 2026-03-14
 fix: network map label size, site layout, and missing links
   - Increased node label font from 11px to 13px and compound node label to 14px

lib/towerops/monitoring/check.ex

@@ -119,6 +119,7 @@ defmodule Towerops.Monitoring.Check do
       "http" -> validate_http_config(changeset, config)
       "tcp" -> validate_tcp_config(changeset, config)
       "dns" -> validate_dns_config(changeset, config)
+      "ping" -> validate_ping_config(changeset, config)
       _ -> changeset
     end
   end
@@ -152,6 +153,14 @@ defmodule Towerops.Monitoring.Check do
     end
   end
 
+  defp validate_ping_config(changeset, config) do
+    if Map.has_key?(config, "host") do
+      changeset
+    else
+      add_error(changeset, :config, "Ping check requires 'host' in config")
+    end
+  end
+
   defp validate_source_type(changeset) do
     source_type = get_field(changeset, :source_type)
 

lib/towerops/monitoring/executors/ping_executor.ex

@@ -0,0 +1,125 @@
+defmodule Towerops.Monitoring.Executors.PingExecutor do
+  @moduledoc """
+  Executes ICMP ping checks using the system `ping` command.
+
+  Config format:
+  %{
+    "host" => "10.0.0.1",
+    "count" => 3  # optional, default: 3
+  }
+  """
+
+  require Logger
+
+  @default_count 3
+  @default_timeout 5000
+
+  @doc """
+  Executes a ping check.
+
+  Returns:
+  - {:ok, response_time_ms, output} on success
+  - {:error, reason} on failure
+  """
+  def execute(config, timeout_ms \\ @default_timeout) do
+    host = Map.fetch!(config, "host")
+    count = config |> Map.get("count", @default_count) |> max(1) |> min(10)
+
+    with :ok <- validate_host(host) do
+      args = build_args(host, count, timeout_ms)
+
+      # Use timeout_ms plus buffer for the system call
+      system_timeout = timeout_ms + count * 1000 + 2000
+
+      case System.cmd("ping", args, stderr_to_stdout: true, timeout: system_timeout) do
+        {output, 0} ->
+          parse_output(output)
+
+        {output, _exit_code} ->
+          # Non-zero exit — try to parse anyway (partial loss still has stats)
+          case parse_output(output) do
+            {:ok, _, _} = success -> success
+            {:error, _} = error -> error
+          end
+      end
+    end
+  rescue
+    e ->
+      Logger.error("Ping check exception: #{inspect(e)}")
+      {:error, "Exception: #{Exception.message(e)}"}
+  catch
+    :exit, {:timeout, _} ->
+      {:error, "Ping command timed out"}
+  end
+
+  @doc """
+  Parses ping command output to extract RTT and packet loss.
+  """
+  def parse_output(output) when is_binary(output) and byte_size(output) > 0 do
+    with {:ok, loss_pct, packets_info} <- parse_packet_loss(output),
+         {:ok, avg_ms} <- parse_rtt(output) do
+      if loss_pct >= 100.0 do
+        {:error, "100% packet loss"}
+      else
+        summary = "#{packets_info}, #{format_loss(loss_pct)} loss, avg #{avg_ms}ms"
+        {:ok, avg_ms, summary}
+      end
+    end
+  end
+
+  def parse_output(_), do: {:error, "No ping output"}
+
+  defp validate_host(host) do
+    # Only allow valid hostnames and IP addresses — no shell metacharacters
+    if Regex.match?(~r/^[a-zA-Z0-9\.\-\:]+$/, host) do
+      :ok
+    else
+      {:error, "Invalid host: contains disallowed characters"}
+    end
+  end
+
+  defp build_args(host, count, timeout_ms) do
+    count_str = Integer.to_string(count)
+
+    case :os.type() do
+      {:unix, :darwin} ->
+        # macOS: -W is in milliseconds
+        ["-c", count_str, "-W", Integer.to_string(timeout_ms), host]
+
+      {:unix, _} ->
+        # Linux: -w is in seconds (total deadline)
+        deadline = max(div(timeout_ms, 1000), count + 1)
+        ["-c", count_str, "-w", Integer.to_string(deadline), host]
+    end
+  end
+
+  defp parse_packet_loss(output) do
+    # Match patterns like "3 packets transmitted, 3 packets received, 0.0% packet loss"
+    # or "3 packets transmitted, 3 received, 0% packet loss"
+    case Regex.run(~r/(\d+) packets? transmitted, (\d+) (?:packets? )?received, ([\d.]+)% packet loss/, output) do
+      [_, transmitted, _received, loss_str] ->
+        {loss_pct, _} = Float.parse(loss_str)
+        {:ok, loss_pct, "#{transmitted} packets"}
+
+      nil ->
+        {:error, "Could not parse packet loss from ping output"}
+    end
+  end
+
+  defp parse_rtt(output) do
+    # macOS: "round-trip min/avg/max/stddev = 0.042/0.059/0.068/0.012 ms"
+    # Linux: "rtt min/avg/max/mdev = 1.120/1.266/1.450/0.140 ms"
+    case Regex.run(~r/(?:round-trip|rtt) min\/avg\/max\/(?:std|m)dev = [\d.]+\/([\d.]+)\//, output) do
+      [_, avg_str] ->
+        {avg_ms, _} = Float.parse(avg_str)
+        {:ok, avg_ms}
+
+      nil ->
+        # No RTT stats (100% loss case)
+        {:error, "100% packet loss"}
+    end
+  end
+
+  defp format_loss(loss_pct) when loss_pct == 0.0, do: "0%"
+  defp format_loss(loss_pct), do: "#{Float.round(loss_pct, 1)}%"
+end

lib/towerops/workers/check_executor_worker.ex

@@ -28,6 +28,7 @@ defmodule Towerops.Workers.CheckExecutorWorker do
   alias Towerops.Monitoring
   alias Towerops.Monitoring.Executors.DnsExecutor
   alias Towerops.Monitoring.Executors.HttpExecutor
+  alias Towerops.Monitoring.Executors.PingExecutor
   alias Towerops.Monitoring.Executors.SnmpInterfaceExecutor
   alias Towerops.Monitoring.Executors.SnmpProcessorExecutor
   alias Towerops.Monitoring.Executors.SnmpSensorExecutor
@@ -86,7 +87,7 @@ defmodule Towerops.Workers.CheckExecutorWorker do
   defp dispatch_executor(%{check_type: "http"} = check), do: execute_http_check(check)
   defp dispatch_executor(%{check_type: "tcp"} = check), do: execute_tcp_check(check)
   defp dispatch_executor(%{check_type: "dns"} = check), do: execute_dns_check(check)
-  defp dispatch_executor(%{check_type: "ping"}), do: {:error, "Ping executor not yet implemented"}
+  defp dispatch_executor(%{check_type: "ping"} = check), do: execute_ping_check(check)
   defp dispatch_executor(%{check_type: type}), do: {:error, "Unknown check type: #{type}"}
 
   defp record_result({:ok, %{value: value, status: status, output: output, response_time_ms: time}}, check) do
@@ -171,6 +172,23 @@ defmodule Towerops.Workers.CheckExecutorWorker do
     end
   end
 
+  # Adapter for Ping executor which returns different format
+  defp execute_ping_check(check) do
+    case PingExecutor.execute(check.config, check.timeout_ms) do
+      {:ok, response_time, output} ->
+        {:ok,
+         %{
+           value: nil,
+           status: 0,
+           output: output,
+           response_time_ms: response_time
+         }}
+
+      {:error, reason} ->
+        {:error, reason}
+    end
+  end
+
   defp schedule_next_check(check) do
     # Calculate staggered offset based on check ID
     offset = PollingOffset.calculate_offset(check.id, check.interval_seconds)

lib/towerops_web/controllers/api/v1/checks_controller.ex

@@ -0,0 +1,199 @@
+defmodule ToweropsWeb.Api.V1.ChecksController do
+  @moduledoc """
+  API controller for managing service checks (HTTP, TCP, DNS, ping).
+
+  All endpoints require API token authentication and operations are scoped
+  to the organization associated with the token. Only service check types
+  (http, tcp, dns, ping) are exposed — SNMP checks are managed internally.
+  """
+  use ToweropsWeb, :controller
+
+  import ToweropsWeb.Api.ErrorHelpers, only: [translate_errors: 1]
+
+  alias Towerops.Monitoring
+  alias Towerops.Monitoring.Check
+  alias ToweropsWeb.ScopedResource
+
+  @service_check_types ~w(http tcp dns ping)
+
+  @doc """
+  GET /api/v1/checks
+
+  Lists service checks for the authenticated organization.
+  Optional query params: `check_type`, `device_id`.
+  """
+  def index(conn, params) do
+    organization_id = conn.assigns.current_organization_id
+
+    opts =
+      []
+      |> maybe_add_opt(:device_id, params["device_id"])
+      |> maybe_add_opt(:check_type, params["check_type"])
+
+    checks =
+      organization_id
+      |> Monitoring.list_checks(opts)
+      |> Enum.filter(&(&1.check_type in @service_check_types))
+      |> Enum.map(&format_check/1)
+
+    json(conn, %{checks: checks})
+  end
+
+  @doc """
+  POST /api/v1/checks
+
+  Creates a new service check for the authenticated organization.
+  """
+  def create(conn, %{"check" => check_params}) do
+    organization_id = conn.assigns.current_organization_id
+
+    check_type = Map.get(check_params, "check_type", "")
+
+    if check_type in @service_check_types do
+      attrs = Map.put(check_params, "organization_id", organization_id)
+
+      case Monitoring.create_check(attrs) do
+        {:ok, check} ->
+          if check.enabled, do: Monitoring.schedule_check(check)
+
+          conn
+          |> put_status(:created)
+          |> json(format_check(check))
+
+        {:error, %Ecto.Changeset{} = changeset} ->
+          conn
+          |> put_status(:unprocessable_entity)
+          |> json(%{errors: translate_errors(changeset)})
+      end
+    else
+      conn
+      |> put_status(:bad_request)
+      |> json(%{error: "Invalid check type '#{check_type}'. Must be one of: #{Enum.join(@service_check_types, ", ")}"})
+    end
+  end
+
+  def create(conn, _params) do
+    conn
+    |> put_status(:bad_request)
+    |> json(%{error: "Missing 'check' parameter"})
+  end
+
+  @doc """
+  GET /api/v1/checks/:id
+
+  Gets a single check by ID.
+  """
+  def show(conn, %{"id" => id}) do
+    organization_id = conn.assigns.current_organization_id
+
+    case ScopedResource.fetch(Check, id, organization_id) do
+      {:ok, check} ->
+        json(conn, format_check(check))
+
+      {:error, :forbidden} ->
+        conn
+        |> put_status(:forbidden)
+        |> json(%{error: "Access denied to this check"})
+
+      {:error, :not_found} ->
+        conn
+        |> put_status(:not_found)
+        |> json(%{error: "Check not found"})
+    end
+  end
+
+  @doc """
+  PATCH /api/v1/checks/:id
+
+  Updates a check.
+  """
+  def update(conn, %{"id" => id, "check" => check_params}) do
+    organization_id = conn.assigns.current_organization_id
+
+    case ScopedResource.fetch(Check, id, organization_id) do
+      {:ok, check} ->
+        case Monitoring.update_check(check, check_params) do
+          {:ok, updated_check} ->
+            json(conn, format_check(updated_check))
+
+          {:error, %Ecto.Changeset{} = changeset} ->
+            conn
+            |> put_status(:unprocessable_entity)
+            |> json(%{errors: translate_errors(changeset)})
+        end
+
+      {:error, :forbidden} ->
+        conn
+        |> put_status(:forbidden)
+        |> json(%{error: "Access denied to this check"})
+
+      {:error, :not_found} ->
+        conn
+        |> put_status(:not_found)
+        |> json(%{error: "Check not found"})
+    end
+  end
+
+  def update(conn, _params) do
+    conn
+    |> put_status(:bad_request)
+    |> json(%{error: "Missing 'check' parameter"})
+  end
+
+  @doc """
+  DELETE /api/v1/checks/:id
+
+  Deletes a check and cancels any scheduled jobs.
+  """
+  def delete(conn, %{"id" => id}) do
+    organization_id = conn.assigns.current_organization_id
+
+    case ScopedResource.fetch(Check, id, organization_id) do
+      {:ok, check} ->
+        case Monitoring.delete_check(check) do
+          {:ok, _check} ->
+            json(conn, %{success: true})
+
+          {:error, %Ecto.Changeset{} = changeset} ->
+            conn
+            |> put_status(:unprocessable_entity)
+            |> json(%{errors: translate_errors(changeset)})
+        end
+
+      {:error, :forbidden} ->
+        conn
+        |> put_status(:forbidden)
+        |> json(%{error: "Access denied to this check"})
+
+      {:error, :not_found} ->
+        conn
+        |> put_status(:not_found)
+        |> json(%{error: "Check not found"})
+    end
+  end
+
+  defp format_check(check) do
+    %{
+      id: check.id,
+      name: check.name,
+      check_type: check.check_type,
+      description: check.description,
+      enabled: check.enabled,
+      device_id: check.device_id,
+      agent_token_id: check.agent_token_id,
+      interval_seconds: check.interval_seconds,
+      timeout_ms: check.timeout_ms,
+      retry_interval_seconds: check.retry_interval_seconds,
+      max_check_attempts: check.max_check_attempts,
+      config: check.config,
+      current_state: check.current_state,
+      current_state_type: check.current_state_type,
+      last_check_at: check.last_check_at,
+      inserted_at: check.inserted_at
+    }
+  end
+
+  defp maybe_add_opt(opts, _key, nil), do: opts
+  defp maybe_add_opt(opts, _key, ""), do: opts
+  defp maybe_add_opt(opts, key, value), do: Keyword.put(opts, key, value)
+end

lib/towerops_web/controllers/api_docs_html/index.html.heex

@@ -130,6 +130,15 @@
                 Integrations
               </a>
             </li>
+            <li>
+              <a
+                href="#checks"
+                data-nav-link="checks"
+                class="nav-link block py-1 pl-4 pr-3 text-sm text-gray-600 transition hover:text-gray-900 dark:text-gray-400 dark:hover:text-white"
+              >
+                Checks
+              </a>
+            </li>
             <li>
               <a
                 href="#check-results"
@@ -3100,6 +3109,445 @@ curl -X POST https://towerops.net/api/v1/integrations/intg-uuid-1/test \\
 
       <hr class="my-16 border-gray-200 dark:border-white/10" />
       
+<!-- Checks Resource -->
+      <section id="checks" class="scroll-mt-24 mb-16">
+        <h2 class="text-3xl font-bold tracking-tight text-zinc-900 dark:text-white">
+          Checks
+        </h2>
+        <p class="mt-4 text-base text-gray-600 dark:text-gray-400">
+          Service checks monitor the availability and responsiveness of network services. Supported check types: HTTP, TCP, DNS, and ping. SNMP checks are managed internally and are not exposed via this API.
+        </p>
+
+<!-- Check Model -->
+        <div class="mt-12">
+          <h3 class="text-xl font-semibold text-zinc-900 dark:text-white">The check model</h3>
+          <div class="mt-6">
+            <dl class="divide-y divide-gray-200 dark:divide-white/10">
+              <div class="py-4">
+                <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                  id
+                  <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                    string
+                  </span>
+                </dt>
+                <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                  Unique identifier for the check (UUID).
+                </dd>
+              </div>
+              <div class="py-4">
+                <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                  name
+                  <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                    string
+                  </span>
+                </dt>
+                <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                  The name of the check.
+                </dd>
+              </div>
+              <div class="py-4">
+                <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                  check_type
+                  <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                    string
+                  </span>
+                </dt>
+                <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                  The type of check: <code>http</code>, <code>tcp</code>, <code>dns</code>, or <code>ping</code>.
+                </dd>
+              </div>
+              <div class="py-4">
+                <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                  enabled
+                  <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                    boolean
+                  </span>
+                </dt>
+                <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                  Whether the check is enabled and scheduled to run.
+                </dd>
+              </div>
+              <div class="py-4">
+                <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                  config
+                  <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                    object
+                  </span>
+                </dt>
+                <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                  Type-specific configuration. HTTP: <code>url</code>, <code>method</code>, <code>expected_status</code>, <code>verify_ssl</code>, <code>follow_redirects</code>, <code>regex</code>. TCP: <code>host</code>, <code>port</code>, <code>send</code>, <code>expect</code>. DNS: <code>hostname</code>, <code>record_type</code>, <code>server</code>, <code>expected</code>. Ping: <code>host</code>, <code>count</code>.
+                </dd>
+              </div>
+              <div class="py-4">
+                <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                  interval_seconds
+                  <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                    integer
+                  </span>
+                </dt>
+                <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                  How often the check runs, in seconds. Default: 60.
+                </dd>
+              </div>
+              <div class="py-4">
+                <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                  current_state
+                  <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                    integer
+                  </span>
+                </dt>
+                <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                  Current check state: 0=OK, 1=WARNING, 2=CRITICAL, 3=UNKNOWN.
+                </dd>
+              </div>
+              <div class="py-4">
+                <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                  current_state_type
+                  <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                    string
+                  </span>
+                </dt>
+                <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                  Current state type: <code>soft</code> or <code>hard</code>.
+                </dd>
+              </div>
+            </dl>
+          </div>
+        </div>
+
+        <hr class="my-12 border-gray-200 dark:border-white/10" />
+
+<!-- List all checks -->
+        <div class="mt-12">
+          <div class="flex items-center gap-x-3">
+            <h3 class="text-xl font-semibold text-zinc-900 dark:text-white">List all checks</h3>
+            <span class="inline-flex items-center rounded-md bg-emerald-50 px-2 py-1 text-xs font-medium text-emerald-700 ring-1 ring-inset ring-emerald-600/20 dark:bg-emerald-400/10 dark:text-emerald-400 dark:ring-emerald-400/30">
+              GET
+            </span>
+            <span class="text-sm font-mono text-gray-600 dark:text-gray-400">/api/v1/checks</span>
+          </div>
+
+          <div class="mt-6 grid grid-cols-1 gap-x-12 gap-y-8 lg:grid-cols-2">
+            <div>
+              <p class="text-sm text-gray-600 dark:text-gray-400">
+                Lists all service checks for the authenticated organization. Only returns HTTP, TCP, DNS, and ping checks.
+              </p>
+
+              <h4 class="mt-6 text-sm font-semibold text-zinc-900 dark:text-white">
+                Optional query parameters
+              </h4>
+              <dl class="mt-3 divide-y divide-gray-200 dark:divide-white/10">
+                <div class="py-3">
+                  <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                    check_type
+                    <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                      string
+                    </span>
+                  </dt>
+                  <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                    Filter by check type: <code>http</code>, <code>tcp</code>, <code>dns</code>, or <code>ping</code>.
+                  </dd>
+                </div>
+                <div class="py-3">
+                  <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                    device_id
+                    <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                      string
+                    </span>
+                  </dt>
+                  <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                    Filter checks by device UUID.
+                  </dd>
+                </div>
+              </dl>
+            </div>
+
+            <div>
+              <div class="overflow-hidden rounded-lg bg-gray-900 dark:bg-gray-950">
+                <div class="border-b border-zinc-800 px-4 py-2">
+                  <p class="text-xs font-medium text-zinc-400">Request</p>
+                </div>
+                <pre class="p-4 text-sm text-zinc-100 overflow-x-auto"><code><%= raw(~S"""
+curl -G https://towerops.net/api/v1/checks \
+  -H "Authorization: Bearer {token}" \
+  -d check_type=http
+""") %></code></pre>
+              </div>
+
+              <div class="mt-4 overflow-hidden rounded-lg bg-gray-900 dark:bg-gray-950">
+                <div class="border-b border-zinc-800 px-4 py-2">
+                  <p class="text-xs font-medium text-zinc-400">Response</p>
+                </div>
+                <pre class="p-4 text-sm text-zinc-100 overflow-x-auto"><code><%= raw(~S"""
+{
+  "checks": [
+    {
+      "id": "7c9e6679-7425-40de-944b-e07fc1f90ae7",
+      "name": "Web Health Check",
+      "check_type": "http",
+      "enabled": true,
+      "config": {"url": "https://example.com/health", "expected_status": 200},
+      "interval_seconds": 60,
+      "current_state": 0,
+      "current_state_type": "hard",
+      "last_check_at": "2026-03-16T12:00:00Z",
+      "inserted_at": "2026-03-15T10:00:00Z"
+    }
+  ]
+}
+""") %></code></pre>
+              </div>
+            </div>
+          </div>
+        </div>
+
+        <hr class="my-12 border-gray-200 dark:border-white/10" />
+
+<!-- Create a check -->
+        <div class="mt-12">
+          <div class="flex items-center gap-x-3">
+            <h3 class="text-xl font-semibold text-zinc-900 dark:text-white">Create a check</h3>
+            <span class="inline-flex items-center rounded-md bg-blue-50 px-2 py-1 text-xs font-medium text-blue-700 ring-1 ring-inset ring-blue-600/20 dark:bg-blue-400/10 dark:text-blue-400 dark:ring-blue-400/30">
+              POST
+            </span>
+            <span class="text-sm font-mono text-gray-600 dark:text-gray-400">/api/v1/checks</span>
+          </div>
+
+          <div class="mt-6 grid grid-cols-1 gap-x-12 gap-y-8 lg:grid-cols-2">
+            <div>
+              <p class="text-sm text-gray-600 dark:text-gray-400">
+                Creates a new service check for the authenticated organization. The check is automatically scheduled if enabled.
+              </p>
+
+              <h4 class="mt-6 text-sm font-semibold text-zinc-900 dark:text-white">
+                Required parameters
+              </h4>
+              <dl class="mt-3 divide-y divide-gray-200 dark:divide-white/10">
+                <div class="py-3">
+                  <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                    name
+                    <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                      string
+                    </span>
+                  </dt>
+                  <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                    The name of the check.
+                  </dd>
+                </div>
+                <div class="py-3">
+                  <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                    check_type
+                    <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                      string
+                    </span>
+                  </dt>
+                  <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                    One of: <code>http</code>, <code>tcp</code>, <code>dns</code>, <code>ping</code>.
+                  </dd>
+                </div>
+                <div class="py-3">
+                  <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                    config
+                    <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                      object
+                    </span>
+                  </dt>
+                  <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                    Type-specific configuration object. See check model for available fields.
+                  </dd>
+                </div>
+              </dl>
+
+              <h4 class="mt-6 text-sm font-semibold text-zinc-900 dark:text-white">
+                Optional parameters
+              </h4>
+              <dl class="mt-3 divide-y divide-gray-200 dark:divide-white/10">
+                <div class="py-3">
+                  <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                    enabled
+                    <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                      boolean
+                    </span>
+                  </dt>
+                  <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                    Whether the check is enabled. Default: <code>true</code>.
+                  </dd>
+                </div>
+                <div class="py-3">
+                  <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                    device_id
+                    <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                      string
+                    </span>
+                  </dt>
+                  <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                    Associate the check with a device.
+                  </dd>
+                </div>
+                <div class="py-3">
+                  <dt class="text-sm font-mono font-medium text-zinc-900 dark:text-white">
+                    interval_seconds
+                    <span class="ml-2 font-sans text-xs font-normal text-gray-500 dark:text-gray-400">
+                      integer
+                    </span>
+                  </dt>
+                  <dd class="mt-1 text-sm text-gray-600 dark:text-gray-400">
+                    How often the check runs. Default: <code>60</code>.
+                  </dd>
+                </div>
+              </dl>
+            </div>
+
+            <div>
+              <div class="overflow-hidden rounded-lg bg-gray-900 dark:bg-gray-950">
+                <div class="border-b border-zinc-800 px-4 py-2">
+                  <p class="text-xs font-medium text-zinc-400">Request</p>
+                </div>
+                <pre class="p-4 text-sm text-zinc-100 overflow-x-auto"><code><%= raw(~S"""
+curl https://towerops.net/api/v1/checks \
+  -H "Authorization: Bearer {token}" \
+  -H "Content-Type: application/json" \
+  -d '{"check": {"name": "Web Health", "check_type": "http", "config": {"url": "https://example.com/health", "expected_status": 200}}}'
+""") %></code></pre>
+              </div>
+
+              <div class="mt-4 overflow-hidden rounded-lg bg-gray-900 dark:bg-gray-950">
+                <div class="border-b border-zinc-800 px-4 py-2">
+                  <p class="text-xs font-medium text-zinc-400">Response <span class="text-zinc-500">201 Created</span></p>
+                </div>
+                <pre class="p-4 text-sm text-zinc-100 overflow-x-auto"><code><%= raw(~S"""
+{
+  "id": "7c9e6679-7425-40de-944b-e07fc1f90ae7",
+  "name": "Web Health",
+  "check_type": "http",
+  "enabled": true,
+  "config": {"url": "https://example.com/health", "expected_status": 200},
+  "interval_seconds": 60,
+  "timeout_ms": 5000,
+  "current_state": null,
+  "inserted_at": "2026-03-16T10:00:00Z"
+}
+""") %></code></pre>
+              </div>
+            </div>
+          </div>
+        </div>
+
+        <hr class="my-12 border-gray-200 dark:border-white/10" />
+
+<!-- Get a check -->
+        <div class="mt-12">
+          <div class="flex items-center gap-x-3">
+            <h3 class="text-xl font-semibold text-zinc-900 dark:text-white">Retrieve a check</h3>
+            <span class="inline-flex items-center rounded-md bg-emerald-50 px-2 py-1 text-xs font-medium text-emerald-700 ring-1 ring-inset ring-emerald-600/20 dark:bg-emerald-400/10 dark:text-emerald-400 dark:ring-emerald-400/30">
+              GET
+            </span>
+            <span class="text-sm font-mono text-gray-600 dark:text-gray-400">/api/v1/checks/:id</span>
+          </div>
+
+          <div class="mt-6 grid grid-cols-1 gap-x-12 gap-y-8 lg:grid-cols-2">
+            <div>
+              <p class="text-sm text-gray-600 dark:text-gray-400">
+                Retrieves a single check by its UUID.
+              </p>
+            </div>
+
+            <div>
+              <div class="overflow-hidden rounded-lg bg-gray-900 dark:bg-gray-950">
+                <div class="border-b border-zinc-800 px-4 py-2">
+                  <p class="text-xs font-medium text-zinc-400">Request</p>
+                </div>
+                <pre class="p-4 text-sm text-zinc-100 overflow-x-auto"><code><%= raw(~S"""
+curl https://towerops.net/api/v1/checks/7c9e6679-7425-40de-944b-e07fc1f90ae7 \
+  -H "Authorization: Bearer {token}"
+""") %></code></pre>
+              </div>
+            </div>
+          </div>
+        </div>
+
+        <hr class="my-12 border-gray-200 dark:border-white/10" />
+
+<!-- Update a check -->
+        <div class="mt-12">
+          <div class="flex items-center gap-x-3">
+            <h3 class="text-xl font-semibold text-zinc-900 dark:text-white">Update a check</h3>
+            <span class="inline-flex items-center rounded-md bg-amber-50 px-2 py-1 text-xs font-medium text-amber-700 ring-1 ring-inset ring-amber-600/20 dark:bg-amber-400/10 dark:text-amber-400 dark:ring-amber-400/30">
+              PATCH
+            </span>
+            <span class="text-sm font-mono text-gray-600 dark:text-gray-400">/api/v1/checks/:id</span>
+          </div>
+
+          <div class="mt-6 grid grid-cols-1 gap-x-12 gap-y-8 lg:grid-cols-2">
+            <div>
+              <p class="text-sm text-gray-600 dark:text-gray-400">
+                Updates an existing check. Only the provided fields will be updated.
+              </p>
+            </div>
+
+            <div>
+              <div class="overflow-hidden rounded-lg bg-gray-900 dark:bg-gray-950">
+                <div class="border-b border-zinc-800 px-4 py-2">
+                  <p class="text-xs font-medium text-zinc-400">Request</p>
+                </div>
+                <pre class="p-4 text-sm text-zinc-100 overflow-x-auto"><code><%= raw(~S"""
+curl -X PATCH https://towerops.net/api/v1/checks/7c9e6679-7425-40de-944b-e07fc1f90ae7 \
+  -H "Authorization: Bearer {token}" \
+  -H "Content-Type: application/json" \
+  -d '{"check": {"name": "Updated Name", "interval_seconds": 120}}'
+""") %></code></pre>
+              </div>
+            </div>
+          </div>
+        </div>
+
+        <hr class="my-12 border-gray-200 dark:border-white/10" />
+
+<!-- Delete a check -->
+        <div class="mt-12">
+          <div class="flex items-center gap-x-3">
+            <h3 class="text-xl font-semibold text-zinc-900 dark:text-white">Delete a check</h3>
+            <span class="inline-flex items-center rounded-md bg-red-50 px-2 py-1 text-xs font-medium text-red-700 ring-1 ring-inset ring-red-600/20 dark:bg-red-400/10 dark:text-red-400 dark:ring-red-400/30">
+              DELETE
+            </span>
+            <span class="text-sm font-mono text-gray-600 dark:text-gray-400">/api/v1/checks/:id</span>
+          </div>
+
+          <div class="mt-6 grid grid-cols-1 gap-x-12 gap-y-8 lg:grid-cols-2">
+            <div>
+              <p class="text-sm text-gray-600 dark:text-gray-400">
+                Deletes a check and cancels any scheduled executions.
+              </p>
+            </div>
+
+            <div>
+              <div class="overflow-hidden rounded-lg bg-gray-900 dark:bg-gray-950">
+                <div class="border-b border-zinc-800 px-4 py-2">
+                  <p class="text-xs font-medium text-zinc-400">Request</p>
+                </div>
+                <pre class="p-4 text-sm text-zinc-100 overflow-x-auto"><code><%= raw(~S"""
+curl -X DELETE https://towerops.net/api/v1/checks/7c9e6679-7425-40de-944b-e07fc1f90ae7 \
+  -H "Authorization: Bearer {token}"
+""") %></code></pre>
+              </div>
+
+              <div class="mt-4 overflow-hidden rounded-lg bg-gray-900 dark:bg-gray-950">
+                <div class="border-b border-zinc-800 px-4 py-2">
+                  <p class="text-xs font-medium text-zinc-400">Response</p>
+                </div>
+                <pre class="p-4 text-sm text-zinc-100 overflow-x-auto"><code><%= raw(~S"""
+{
+  "success": true
+}
+""") %></code></pre>
+              </div>
+            </div>
+          </div>
+        </div>
+      </section>
+
+      <hr class="my-16 border-gray-200 dark:border-white/10" />
+
 <!-- Check Results & Metrics Resource -->
       <section id="check-results" class="scroll-mt-24 mb-16">
         <h2 class="text-3xl font-bold tracking-tight text-zinc-900 dark:text-white">

lib/towerops_web/graphql/resolvers/check.ex

@@ -0,0 +1,90 @@
+defmodule ToweropsWeb.GraphQL.Resolvers.Check do
+  @moduledoc "GraphQL resolvers for service check queries and mutations."
+
+  alias Towerops.Monitoring
+  alias Towerops.Monitoring.Check
+  alias ToweropsWeb.GraphQL.Resolvers.Helpers
+  alias ToweropsWeb.ScopedResource
+
+  @service_check_types ~w(http tcp dns ping)
+
+  def list(_parent, args, %{context: %{organization_id: org_id}}) do
+    opts =
+      []
+      |> maybe_add_opt(:device_id, args[:device_id])
+      |> maybe_add_opt(:check_type, args[:check_type])
+
+    checks =
+      org_id
+      |> Monitoring.list_checks(opts)
+      |> Enum.filter(&(&1.check_type in @service_check_types))
+
+    {:ok, checks}
+  end
+
+  def list(_parent, _args, _resolution), do: Helpers.authentication_error()
+
+  def get(_parent, %{id: id}, %{context: %{organization_id: org_id}}) do
+    fetch_org_check(id, org_id)
+  end
+
+  def get(_parent, _args, _resolution), do: Helpers.authentication_error()
+
+  def create(_parent, %{input: input}, %{context: %{organization_id: org_id}}) do
+    attrs =
+      input
+      |> Map.new(fn {k, v} -> {to_string(k), v} end)
+      |> Map.put("organization_id", org_id)
+
+    check_type = Map.get(attrs, "check_type", "")
+
+    if check_type in @service_check_types do
+      case Monitoring.create_check(attrs) do
+        {:ok, check} ->
+          if check.enabled, do: Monitoring.schedule_check(check)
+          {:ok, check}
+
+        {:error, changeset} ->
+          {:error, Helpers.format_changeset_errors(changeset)}
+      end
+    else
+      {:error, "Invalid check type '#{check_type}'. Must be one of: #{Enum.join(@service_check_types, ", ")}"}
+    end
+  end
+
+  def create(_parent, _args, _resolution), do: Helpers.authentication_error()
+
+  def update(_parent, %{id: id, input: input}, %{context: %{organization_id: org_id}}) do
+    with {:ok, check} <- fetch_org_check(id, org_id) do
+      attrs = Map.new(input, fn {k, v} -> {to_string(k), v} end)
+
+      case Monitoring.update_check(check, attrs) do
+        {:ok, updated} -> {:ok, updated}
+        {:error, changeset} -> {:error, Helpers.format_changeset_errors(changeset)}
+      end
+    end
+  end
+
+  def update(_parent, _args, _resolution), do: Helpers.authentication_error()
+
+  def delete(_parent, %{id: id}, %{context: %{organization_id: org_id}}) do
+    with {:ok, check} <- fetch_org_check(id, org_id) do
+      case Monitoring.delete_check(check) do
+        {:ok, _} -> {:ok, %{success: true, message: "Check deleted"}}
+        {:error, _} -> {:ok, %{success: false, message: "Could not delete check"}}
+      end
+    end
+  end
+
+  def delete(_parent, _args, _resolution), do: Helpers.authentication_error()
+
+  defp fetch_org_check(id, org_id) do
+    case ScopedResource.fetch(Check, id, org_id) do
+      {:ok, check} -> {:ok, check}
+      {:error, _reason} -> {:error, "Check not found"}
+    end
+  end
+
+  defp maybe_add_opt(opts, _key, nil), do: opts
+  defp maybe_add_opt(opts, key, value), do: Keyword.put(opts, key, value)
+end

lib/towerops_web/graphql/schema.ex

@@ -24,6 +24,7 @@ defmodule ToweropsWeb.GraphQL.Schema do
   import_types(ToweropsWeb.GraphQL.Types.Schedule)
   import_types(ToweropsWeb.GraphQL.Types.EscalationPolicy)
   import_types(ToweropsWeb.GraphQL.Types.MaintenanceWindow)
+  import_types(ToweropsWeb.GraphQL.Types.Check)
   import_types(ToweropsWeb.GraphQL.Types.TimeSeries)
 
   query do
@@ -80,6 +81,22 @@ defmodule ToweropsWeb.GraphQL.Schema do
       resolve(&ToweropsWeb.GraphQL.Resolvers.Alert.get/3)
     end
 
+    # Service Checks
+    field :checks, list_of(:check) do
+      arg(:organization_id, :id)
+      arg(:device_id, :id)
+      arg(:check_type, :string)
+      middleware(OrganizationScope)
+      resolve(&ToweropsWeb.GraphQL.Resolvers.Check.list/3)
+    end
+
+    field :check, :check do
+      arg(:organization_id, :id)
+      arg(:id, non_null(:id))
+      middleware(OrganizationScope)
+      resolve(&ToweropsWeb.GraphQL.Resolvers.Check.get/3)
+    end
+
     # Agents
     field :agents, list_of(:agent) do
       arg(:organization_id, :id)
@@ -260,6 +277,23 @@ defmodule ToweropsWeb.GraphQL.Schema do
       resolve(&ToweropsWeb.GraphQL.Resolvers.Site.delete/3)
     end
 
+    # Check CRUD
+    field :create_check, :check do
+      arg(:input, non_null(:check_input))
+      resolve(&ToweropsWeb.GraphQL.Resolvers.Check.create/3)
+    end
+
+    field :update_check, :check do
+      arg(:id, non_null(:id))
+      arg(:input, non_null(:check_input))
+      resolve(&ToweropsWeb.GraphQL.Resolvers.Check.update/3)
+    end
+
+    field :delete_check, :delete_result do
+      arg(:id, non_null(:id))
+      resolve(&ToweropsWeb.GraphQL.Resolvers.Check.delete/3)
+    end
+
     # Alert actions
     field :acknowledge_alert, :alert do
       arg(:id, non_null(:id))

lib/towerops_web/graphql/types/check.ex

@@ -0,0 +1,37 @@
+defmodule ToweropsWeb.GraphQL.Types.Check do
+  @moduledoc "GraphQL types for service checks."
+  use Absinthe.Schema.Notation
+
+  object :check do
+    field :id, :id
+    field :name, :string
+    field :check_type, :string
+    field :description, :string
+    field :enabled, :boolean
+    field :device_id, :id
+    field :agent_token_id, :id
+    field :interval_seconds, :integer
+    field :timeout_ms, :integer
+    field :retry_interval_seconds, :integer
+    field :max_check_attempts, :integer
+    field :config, :json
+    field :current_state, :integer
+    field :current_state_type, :string
+    field :last_check_at, :string
+    field :inserted_at, :string
+  end
+
+  input_object :check_input do
+    field :name, :string
+    field :check_type, :string
+    field :description, :string
+    field :enabled, :boolean
+    field :device_id, :id
+    field :agent_token_id, :id
+    field :interval_seconds, :integer
+    field :timeout_ms, :integer
+    field :retry_interval_seconds, :integer
+    field :max_check_attempts, :integer
+    field :config, :json
+  end
+end

lib/towerops_web/router.ex

@@ -156,6 +156,7 @@ defmodule ToweropsWeb.Router do
 
     resources "/sites", SitesController, except: [:new, :edit]
     resources "/devices", DevicesController, except: [:new, :edit]
+    resources "/checks", ChecksController, except: [:new, :edit]
 
     resources "/agents", AgentsController, except: [:new, :edit]
 

priv/static/changelog.txt

@@ -1,3 +1,9 @@
+2026-03-16 — Service Checks API
+* New REST API for managing service checks (HTTP, TCP, DNS, ping)
+* New GraphQL queries and mutations for service check management
+* Ping monitoring now fully operational with packet loss and RTT tracking
+* API documentation updated with service checks endpoints
+
 2026-03-14 — Network Map Improvements
 * Node labels are now larger and easier to read
 * Sites no longer overlap — each site group is laid out in its own section

test/towerops/monitoring/executors/ping_executor_test.exs

@@ -0,0 +1,137 @@
+defmodule Towerops.Monitoring.Executors.PingExecutorTest do
+  use ExUnit.Case, async: true
+
+  alias Towerops.Monitoring.Executors.PingExecutor
+
+  describe "execute/2" do
+    test "returns success with valid ping output" do
+      config = %{"host" => "127.0.0.1", "count" => 1}
+
+      case PingExecutor.execute(config, 10_000) do
+        {:ok, response_time, output} ->
+          assert is_number(response_time)
+          assert response_time >= 0
+          assert is_binary(output)
+          assert output =~ "packet"
+
+        {:error, reason} ->
+          # Ping to localhost might fail in some CI environments
+          assert is_binary(reason)
+      end
+    end
+
+    test "defaults count to 3 when not provided" do
+      config = %{"host" => "127.0.0.1"}
+
+      # Should use default count of 3 — just verify it doesn't crash
+      result = PingExecutor.execute(config, 10_000)
+      assert match?({:ok, _, _}, result) or match?({:error, _}, result)
+    end
+
+    test "returns error for unreachable host" do
+      # RFC 5737 TEST-NET address — should be unreachable
+      config = %{"host" => "192.0.2.1", "count" => 1}
+
+      case PingExecutor.execute(config, 6000) do
+        {:error, reason} ->
+          assert is_binary(reason)
+
+        {:ok, _time, output} ->
+          # Some networks may route this, that's fine
+          assert is_binary(output)
+      end
+    end
+
+    test "returns error for invalid host" do
+      config = %{"host" => "definitely-not-a-real-host-12345.invalid", "count" => 1}
+
+      assert {:error, reason} = PingExecutor.execute(config, 6000)
+      assert is_binary(reason)
+    end
+
+    test "sanitizes host to prevent command injection" do
+      config = %{"host" => "127.0.0.1; rm -rf /", "count" => 1}
+
+      assert {:error, reason} = PingExecutor.execute(config, 5000)
+      assert reason =~ "Invalid host"
+    end
+
+    test "sanitizes count to positive integer" do
+      config = %{"host" => "127.0.0.1", "count" => -1}
+
+      # Should clamp or reject negative count
+      result = PingExecutor.execute(config, 10_000)
+      assert match?({:ok, _, _}, result) or match?({:error, _}, result)
+    end
+  end
+
+  describe "parse_output/1" do
+    test "parses macOS ping output" do
+      output = """
+      PING 127.0.0.1 (127.0.0.1): 56 data bytes
+      64 bytes from 127.0.0.1: icmp_seq=0 ttl=64 time=0.042 ms
+      64 bytes from 127.0.0.1: icmp_seq=1 ttl=64 time=0.068 ms
+      64 bytes from 127.0.0.1: icmp_seq=2 ttl=64 time=0.066 ms
+
+      --- 127.0.0.1 ping statistics ---
+      3 packets transmitted, 3 packets received, 0.0% packet loss
+      round-trip min/avg/max/stddev = 0.042/0.059/0.068/0.012 ms
+      """
+
+      assert {:ok, avg_ms, summary} = PingExecutor.parse_output(output)
+      assert_in_delta avg_ms, 0.059, 0.001
+      assert summary =~ "3 packets"
+      assert summary =~ "0% loss"
+    end
+
+    test "parses Linux ping output" do
+      output = """
+      PING 10.0.0.1 (10.0.0.1) 56(84) bytes of data.
+      64 bytes from 10.0.0.1: icmp_seq=1 ttl=64 time=1.23 ms
+      64 bytes from 10.0.0.1: icmp_seq=2 ttl=64 time=1.45 ms
+      64 bytes from 10.0.0.1: icmp_seq=3 ttl=64 time=1.12 ms
+
+      --- 10.0.0.1 ping statistics ---
+      3 packets transmitted, 3 received, 0% packet loss, time 2003ms
+      rtt min/avg/max/mdev = 1.120/1.266/1.450/0.140 ms
+      """
+
+      assert {:ok, avg_ms, summary} = PingExecutor.parse_output(output)
+      assert_in_delta avg_ms, 1.266, 0.001
+      assert summary =~ "3 packets"
+      assert summary =~ "0% loss"
+    end
+
+    test "returns error for 100% packet loss" do
+      output = """
+      PING 192.0.2.1 (192.0.2.1): 56 data bytes
+
+      --- 192.0.2.1 ping statistics ---
+      3 packets transmitted, 0 packets received, 100.0% packet loss
+      """
+
+      assert {:error, reason} = PingExecutor.parse_output(output)
+      assert reason =~ "100"
+      assert reason =~ "packet loss"
+    end
+
+    test "returns error for partial packet loss" do
+      output = """
+      PING 10.0.0.1 (10.0.0.1): 56 data bytes
+      64 bytes from 10.0.0.1: icmp_seq=0 ttl=64 time=1.23 ms
+
+      --- 10.0.0.1 ping statistics ---
+      3 packets transmitted, 1 packets received, 66.7% packet loss
+      round-trip min/avg/max/stddev = 1.230/1.230/1.230/0.000 ms
+      """
+
+      # Partial loss should still report as warning with the data
+      result = PingExecutor.parse_output(output)
+      assert match?({:ok, _, _}, result) or match?({:error, _}, result)
+    end
+
+    test "returns error for empty output" do
+      assert {:error, _reason} = PingExecutor.parse_output("")
+    end
+  end
+end

test/towerops_web/controllers/api/v1/checks_controller_test.exs

@@ -0,0 +1,425 @@
+defmodule ToweropsWeb.Api.V1.ChecksControllerTest do
+  use ToweropsWeb.ConnCase, async: true
+
+  import Towerops.AccountsFixtures
+  import Towerops.OrganizationsFixtures
+
+  alias Towerops.ApiTokens
+  alias Towerops.Monitoring
+
+  setup do
+    user = user_fixture()
+    organization = organization_fixture(user.id)
+
+    {:ok, {_token, raw_token}} =
+      ApiTokens.create_api_token(%{
+        organization_id: organization.id,
+        user_id: user.id,
+        name: "Test Token"
+      })
+
+    conn =
+      build_conn()
+      |> put_req_header("authorization", "Bearer #{raw_token}")
+      |> put_req_header("accept", "application/json")
+
+    %{
+      conn: conn,
+      user: user,
+      organization: organization,
+      raw_token: raw_token
+    }
+  end
+
+  defp create_check(organization_id, attrs \\ %{}) do
+    defaults = %{
+      name: "Test Check",
+      check_type: "http",
+      organization_id: organization_id,
+      config: %{"url" => "https://example.com"}
+    }
+
+    {:ok, check} = Monitoring.create_check(Map.merge(defaults, attrs))
+    check
+  end
+
+  describe "index/2" do
+    test "lists service checks for authenticated organization", %{
+      conn: conn,
+      organization: organization
+    } do
+      check1 =
+        create_check(organization.id, %{name: "HTTP Check", check_type: "http", config: %{"url" => "https://example.com"}})
+
+      check2 =
+        create_check(organization.id, %{
+          name: "TCP Check",
+          check_type: "tcp",
+          config: %{"host" => "10.0.0.1", "port" => 80}
+        })
+
+      conn = get(conn, ~p"/api/v1/checks")
+
+      assert %{"checks" => checks} = json_response(conn, 200)
+      assert length(checks) == 2
+
+      check_ids = Enum.map(checks, & &1["id"])
+      assert check1.id in check_ids
+      assert check2.id in check_ids
+
+      check_json = List.first(checks)
+      assert Map.has_key?(check_json, "id")
+      assert Map.has_key?(check_json, "name")
+      assert Map.has_key?(check_json, "check_type")
+      assert Map.has_key?(check_json, "config")
+      assert Map.has_key?(check_json, "enabled")
+      assert Map.has_key?(check_json, "inserted_at")
+    end
+
+    test "excludes SNMP checks from listing", %{conn: conn, organization: organization} do
+      _http =
+        create_check(organization.id, %{name: "HTTP Check", check_type: "http", config: %{"url" => "https://example.com"}})
+
+      # Create an SNMP check — should not appear
+      {:ok, _snmp} =
+        Monitoring.create_check(%{
+          name: "SNMP Sensor",
+          check_type: "snmp_sensor",
+          organization_id: organization.id,
+          config: %{"oid" => "1.3.6.1.2.1.1.3.0"}
+        })
+
+      conn = get(conn, ~p"/api/v1/checks")
+
+      assert %{"checks" => checks} = json_response(conn, 200)
+      assert length(checks) == 1
+      assert List.first(checks)["check_type"] == "http"
+    end
+
+    test "filters by check_type query param", %{conn: conn, organization: organization} do
+      _http =
+        create_check(organization.id, %{name: "HTTP Check", check_type: "http", config: %{"url" => "https://example.com"}})
+
+      _tcp =
+        create_check(organization.id, %{
+          name: "TCP Check",
+          check_type: "tcp",
+          config: %{"host" => "10.0.0.1", "port" => 80}
+        })
+
+      conn = get(conn, ~p"/api/v1/checks?check_type=tcp")
+
+      assert %{"checks" => checks} = json_response(conn, 200)
+      assert length(checks) == 1
+      assert List.first(checks)["check_type"] == "tcp"
+    end
+
+    test "returns empty list when organization has no checks", %{conn: conn} do
+      conn = get(conn, ~p"/api/v1/checks")
+
+      assert %{"checks" => []} = json_response(conn, 200)
+    end
+
+    test "does not return checks from other organizations", %{conn: conn} do
+      other_user = user_fixture()
+      other_org = organization_fixture(other_user.id)
+      _other_check = create_check(other_org.id)
+
+      conn = get(conn, ~p"/api/v1/checks")
+
+      assert %{"checks" => checks} = json_response(conn, 200)
+      assert Enum.empty?(checks)
+    end
+  end
+
+  describe "create/2" do
+    test "creates HTTP check with valid attributes", %{conn: conn, organization: organization} do
+      params = %{
+        "check" => %{
+          "name" => "Web Health",
+          "check_type" => "http",
+          "config" => %{
+            "url" => "https://example.com/health",
+            "expected_status" => 200
+          }
+        }
+      }
+
+      conn = post(conn, ~p"/api/v1/checks", params)
+
+      assert %{
+               "id" => id,
+               "name" => "Web Health",
+               "check_type" => "http",
+               "enabled" => true
+             } = json_response(conn, 201)
+
+      assert is_binary(id)
+
+      check = Monitoring.get_check!(id)
+      assert check.organization_id == organization.id
+    end
+
+    test "creates TCP check with valid attributes", %{conn: conn} do
+      params = %{
+        "check" => %{
+          "name" => "MySQL Port",
+          "check_type" => "tcp",
+          "config" => %{
+            "host" => "10.0.0.5",
+            "port" => 3306
+          }
+        }
+      }
+
+      conn = post(conn, ~p"/api/v1/checks", params)
+
+      assert %{
+               "id" => _id,
+               "name" => "MySQL Port",
+               "check_type" => "tcp"
+             } = json_response(conn, 201)
+    end
+
+    test "creates DNS check with valid attributes", %{conn: conn} do
+      params = %{
+        "check" => %{
+          "name" => "DNS Resolution",
+          "check_type" => "dns",
+          "config" => %{
+            "hostname" => "google.com",
+            "dns_server" => "8.8.8.8",
+            "record_type" => "A"
+          }
+        }
+      }
+
+      conn = post(conn, ~p"/api/v1/checks", params)
+
+      assert %{
+               "id" => _id,
+               "name" => "DNS Resolution",
+               "check_type" => "dns"
+             } = json_response(conn, 201)
+    end
+
+    test "creates ping check with valid attributes", %{conn: conn} do
+      params = %{
+        "check" => %{
+          "name" => "Gateway Ping",
+          "check_type" => "ping",
+          "config" => %{
+            "host" => "10.0.0.1",
+            "count" => 3
+          }
+        }
+      }
+
+      conn = post(conn, ~p"/api/v1/checks", params)
+
+      assert %{
+               "id" => _id,
+               "name" => "Gateway Ping",
+               "check_type" => "ping"
+             } = json_response(conn, 201)
+    end
+
+    test "returns 422 when config is invalid for check type", %{conn: conn} do
+      params = %{
+        "check" => %{
+          "name" => "Bad HTTP Check",
+          "check_type" => "http",
+          "config" => %{}
+        }
+      }
+
+      conn = post(conn, ~p"/api/v1/checks", params)
+
+      assert %{"errors" => _errors} = json_response(conn, 422)
+    end
+
+    test "returns 422 when name is missing", %{conn: conn} do
+      params = %{
+        "check" => %{
+          "check_type" => "http",
+          "config" => %{"url" => "https://example.com"}
+        }
+      }
+
+      conn = post(conn, ~p"/api/v1/checks", params)
+
+      assert %{"errors" => errors} = json_response(conn, 422)
+      assert Map.has_key?(errors, "name")
+    end
+
+    test "returns 400 when check parameter is missing", %{conn: conn} do
+      conn = post(conn, ~p"/api/v1/checks", %{})
+
+      assert %{"error" => "Missing 'check' parameter"} = json_response(conn, 400)
+    end
+
+    test "rejects SNMP check types", %{conn: conn} do
+      params = %{
+        "check" => %{
+          "name" => "SNMP Check",
+          "check_type" => "snmp_sensor",
+          "config" => %{"oid" => "1.3.6.1.2.1.1.3.0"}
+        }
+      }
+
+      conn = post(conn, ~p"/api/v1/checks", params)
+
+      assert %{"error" => _} = json_response(conn, 400)
+    end
+  end
+
+  describe "show/2" do
+    test "returns check when it belongs to organization", %{
+      conn: conn,
+      organization: organization
+    } do
+      check = create_check(organization.id, %{name: "Show Me"})
+
+      conn = get(conn, ~p"/api/v1/checks/#{check.id}")
+
+      assert %{
+               "id" => id,
+               "name" => "Show Me",
+               "check_type" => "http"
+             } = json_response(conn, 200)
+
+      assert id == check.id
+    end
+
+    test "returns 404 when check does not exist", %{conn: conn} do
+      conn = get(conn, ~p"/api/v1/checks/#{Ecto.UUID.generate()}")
+
+      assert %{"error" => "Check not found"} = json_response(conn, 404)
+    end
+
+    test "returns 403 when check belongs to different organization", %{conn: conn} do
+      other_user = user_fixture()
+      other_org = organization_fixture(other_user.id)
+      other_check = create_check(other_org.id)
+
+      conn = get(conn, ~p"/api/v1/checks/#{other_check.id}")
+
+      assert %{"error" => "Access denied to this check"} = json_response(conn, 403)
+    end
+  end
+
+  describe "update/2" do
+    test "updates check with valid attributes", %{conn: conn, organization: organization} do
+      check = create_check(organization.id, %{name: "Original"})
+
+      params = %{
+        "check" => %{
+          "name" => "Updated",
+          "interval_seconds" => 120
+        }
+      }
+
+      conn = patch(conn, ~p"/api/v1/checks/#{check.id}", params)
+
+      assert %{
+               "id" => id,
+               "name" => "Updated",
+               "interval_seconds" => 120
+             } = json_response(conn, 200)
+
+      assert id == check.id
+    end
+
+    test "returns 404 when check does not exist", %{conn: conn} do
+      params = %{
+        "check" => %{
+          "name" => "Updated"
+        }
+      }
+
+      conn = patch(conn, ~p"/api/v1/checks/#{Ecto.UUID.generate()}", params)
+
+      assert %{"error" => "Check not found"} = json_response(conn, 404)
+    end
+
+    test "returns 403 when check belongs to different organization", %{conn: conn} do
+      other_user = user_fixture()
+      other_org = organization_fixture(other_user.id)
+      other_check = create_check(other_org.id)
+
+      params = %{"check" => %{"name" => "Hacked"}}
+
+      conn = patch(conn, ~p"/api/v1/checks/#{other_check.id}", params)
+
+      assert %{"error" => "Access denied to this check"} = json_response(conn, 403)
+    end
+
+    test "returns 422 with invalid params", %{conn: conn, organization: organization} do
+      check = create_check(organization.id)
+
+      params = %{"check" => %{"name" => ""}}
+
+      conn = patch(conn, ~p"/api/v1/checks/#{check.id}", params)
+
+      assert %{"errors" => errors} = json_response(conn, 422)
+      assert Map.has_key?(errors, "name")
+    end
+
+    test "returns 400 when check parameter is missing", %{conn: conn, organization: organization} do
+      check = create_check(organization.id)
+
+      conn = patch(conn, ~p"/api/v1/checks/#{check.id}", %{})
+
+      assert %{"error" => "Missing 'check' parameter"} = json_response(conn, 400)
+    end
+  end
+
+  describe "delete/2" do
+    test "deletes check successfully", %{conn: conn, organization: organization} do
+      check = create_check(organization.id, %{name: "Doomed"})
+
+      conn = delete(conn, ~p"/api/v1/checks/#{check.id}")
+
+      assert %{"success" => true} = json_response(conn, 200)
+
+      assert Monitoring.get_check(check.id) == nil
+    end
+
+    test "returns 404 when check does not exist", %{conn: conn} do
+      conn = delete(conn, ~p"/api/v1/checks/#{Ecto.UUID.generate()}")
+
+      assert %{"error" => "Check not found"} = json_response(conn, 404)
+    end
+
+    test "returns 403 when check belongs to different organization", %{conn: conn} do
+      other_user = user_fixture()
+      other_org = organization_fixture(other_user.id)
+      other_check = create_check(other_org.id)
+
+      conn = delete(conn, ~p"/api/v1/checks/#{other_check.id}")
+
+      assert %{"error" => "Access denied to this check"} = json_response(conn, 403)
+    end
+  end
+
+  describe "authentication" do
+    test "returns 401 without authorization header" do
+      conn =
+        build_conn()
+        |> put_req_header("accept", "application/json")
+        |> get(~p"/api/v1/checks")
+
+      assert %{"error" => _message} = json_response(conn, 401)
+    end
+
+    test "returns 401 with invalid token" do
+      conn =
+        build_conn()
+        |> put_req_header("authorization", "Bearer towerops_invalid_token")
+        |> put_req_header("accept", "application/json")
+        |> get(~p"/api/v1/checks")
+
+      assert %{"error" => _message} = json_response(conn, 401)
+    end
+  end
+end

test/towerops_web/graphql/resolvers/check_test.exs

@@ -0,0 +1,275 @@
+defmodule ToweropsWeb.GraphQL.Resolvers.CheckTest do
+  use ToweropsWeb.ConnCase, async: true
+
+  import Towerops.AccountsFixtures
+  import Towerops.OrganizationsFixtures
+
+  alias Towerops.ApiTokens
+  alias Towerops.Monitoring
+
+  setup do
+    user = user_fixture()
+    organization = organization_fixture(user.id)
+
+    {:ok, {_token, raw_token}} =
+      ApiTokens.create_api_token(%{
+        organization_id: organization.id,
+        user_id: user.id,
+        name: "Test Token"
+      })
+
+    conn =
+      build_conn()
+      |> put_req_header("authorization", "Bearer #{raw_token}")
+      |> put_req_header("content-type", "application/json")
+
+    %{conn: conn, organization: organization}
+  end
+
+  defp graphql_query(conn, query, variables \\ %{}) do
+    conn
+    |> post("/api/graphql", Jason.encode!(%{query: query, variables: variables}))
+    |> json_response(200)
+  end
+
+  defp create_check(organization_id, attrs) do
+    defaults = %{
+      name: "Test Check",
+      check_type: "http",
+      organization_id: organization_id,
+      config: %{"url" => "https://example.com"}
+    }
+
+    {:ok, check} = Monitoring.create_check(Map.merge(defaults, attrs))
+    check
+  end
+
+  describe "checks query" do
+    test "lists service checks", %{conn: conn, organization: org} do
+      _http = create_check(org.id, %{name: "HTTP Check"})
+      _tcp = create_check(org.id, %{name: "TCP Check", check_type: "tcp", config: %{"host" => "10.0.0.1", "port" => 80}})
+
+      query = """
+      {
+        checks {
+          id
+          name
+          checkType
+          enabled
+        }
+      }
+      """
+
+      result = graphql_query(conn, query)
+
+      assert %{"data" => %{"checks" => checks}} = result
+      assert length(checks) == 2
+
+      names = Enum.map(checks, & &1["name"])
+      assert "HTTP Check" in names
+      assert "TCP Check" in names
+    end
+
+    test "filters by check_type", %{conn: conn, organization: org} do
+      _http = create_check(org.id, %{name: "HTTP Check"})
+      _tcp = create_check(org.id, %{name: "TCP Check", check_type: "tcp", config: %{"host" => "10.0.0.1", "port" => 80}})
+
+      query = """
+      query($type: String) {
+        checks(checkType: $type) {
+          name
+          checkType
+        }
+      }
+      """
+
+      result = graphql_query(conn, query, %{"type" => "tcp"})
+
+      assert %{"data" => %{"checks" => [check]}} = result
+      assert check["name"] == "TCP Check"
+    end
+
+    test "excludes SNMP checks", %{conn: conn, organization: org} do
+      _http = create_check(org.id, %{name: "HTTP Check"})
+
+      {:ok, _snmp} =
+        Monitoring.create_check(%{
+          name: "SNMP Sensor",
+          check_type: "snmp_sensor",
+          organization_id: org.id,
+          config: %{"oid" => "1.3.6.1.2.1.1.3.0"}
+        })
+
+      query = """
+      {
+        checks {
+          name
+          checkType
+        }
+      }
+      """
+
+      result = graphql_query(conn, query)
+
+      assert %{"data" => %{"checks" => checks}} = result
+      assert length(checks) == 1
+      assert List.first(checks)["checkType"] == "http"
+    end
+  end
+
+  describe "check query" do
+    test "returns a single check", %{conn: conn, organization: org} do
+      check = create_check(org.id, %{name: "Single Check"})
+
+      query = """
+      query($id: ID!) {
+        check(id: $id) {
+          id
+          name
+          checkType
+          config
+          enabled
+          intervalSeconds
+        }
+      }
+      """
+
+      result = graphql_query(conn, query, %{"id" => check.id})
+
+      assert %{"data" => %{"check" => data}} = result
+      assert data["id"] == check.id
+      assert data["name"] == "Single Check"
+      assert data["checkType"] == "http"
+      assert data["enabled"] == true
+    end
+
+    test "returns error for non-existent check", %{conn: conn} do
+      query = """
+      query($id: ID!) {
+        check(id: $id) {
+          id
+          name
+        }
+      }
+      """
+
+      result = graphql_query(conn, query, %{"id" => Ecto.UUID.generate()})
+
+      assert %{"errors" => [%{"message" => "Check not found"}]} = result
+    end
+  end
+
+  describe "createCheck mutation" do
+    test "creates an HTTP check", %{conn: conn, organization: org} do
+      mutation = """
+      mutation($input: CheckInput!) {
+        createCheck(input: $input) {
+          id
+          name
+          checkType
+          enabled
+        }
+      }
+      """
+
+      input = %{
+        "name" => "New HTTP Check",
+        "check_type" => "http",
+        "config" => Jason.encode!(%{"url" => "https://example.com/health"})
+      }
+
+      result = graphql_query(conn, mutation, %{"input" => input})
+
+      assert %{"data" => %{"createCheck" => check}} = result
+      assert check["name"] == "New HTTP Check"
+      assert check["checkType"] == "http"
+      assert check["enabled"] == true
+
+      # Verify it exists in DB
+      assert Monitoring.get_check(check["id"]).organization_id == org.id
+    end
+
+    test "returns error for invalid check type", %{conn: conn} do
+      mutation = """
+      mutation($input: CheckInput!) {
+        createCheck(input: $input) {
+          id
+        }
+      }
+      """
+
+      input = %{
+        "name" => "Bad Check",
+        "check_type" => "snmp_sensor",
+        "config" => Jason.encode!(%{"oid" => "1.3.6.1.2.1.1.3.0"})
+      }
+
+      result = graphql_query(conn, mutation, %{"input" => input})
+
+      assert %{"errors" => [%{"message" => message}]} = result
+      assert message =~ "Invalid check type"
+    end
+  end
+
+  describe "updateCheck mutation" do
+    test "updates an existing check", %{conn: conn, organization: org} do
+      check = create_check(org.id, %{name: "Original"})
+
+      mutation = """
+      mutation($id: ID!, $input: CheckInput!) {
+        updateCheck(id: $id, input: $input) {
+          id
+          name
+          intervalSeconds
+        }
+      }
+      """
+
+      input = %{"name" => "Updated", "interval_seconds" => 120}
+
+      result = graphql_query(conn, mutation, %{"id" => check.id, "input" => input})
+
+      assert %{"data" => %{"updateCheck" => updated}} = result
+      assert updated["name"] == "Updated"
+      assert updated["intervalSeconds"] == 120
+    end
+  end
+
+  describe "deleteCheck mutation" do
+    test "deletes a check", %{conn: conn, organization: org} do
+      check = create_check(org.id, %{name: "Doomed"})
+
+      mutation = """
+      mutation($id: ID!) {
+        deleteCheck(id: $id) {
+          success
+          message
+        }
+      }
+      """
+
+      result = graphql_query(conn, mutation, %{"id" => check.id})
+
+      assert %{"data" => %{"deleteCheck" => %{"success" => true}}} = result
+      assert Monitoring.get_check(check.id) == nil
+    end
+  end
+
+  describe "authentication" do
+    test "returns 401 without auth" do
+      conn = put_req_header(build_conn(), "content-type", "application/json")
+
+      query = """
+      {
+        checks {
+          id
+        }
+      }
+      """
+
+      resp = post(conn, "/api/graphql", Jason.encode!(%{query: query}))
+
+      assert resp.status == 401
+    end
+  end
+end