test: add edge case tests for stuck job detection

test/towerops/job_monitoring_test.exs

@@ -119,5 +119,87 @@ defmodule Towerops.JobMonitoringTest do
       assert length(stuck) == 1
       assert hd(stuck).id == stuck_job.id
     end
+
+    test "returns stuck jobs from both worker types ordered by attempted_at" do
+      device = device_fixture()
+
+      # Stuck polling job (3 minutes ago)
+      stuck_poller = oban_job_fixture(%{
+        worker: "Towerops.Workers.DevicePollerWorker",
+        state: "executing",
+        args: %{"device_id" => device.id},
+        attempted_at: DateTime.add(DateTime.utc_now(), -180, :second)
+      })
+
+      # Stuck discovery job (6 minutes ago - older, should be first)
+      stuck_discovery = oban_job_fixture(%{
+        worker: "Towerops.Workers.DiscoveryWorker",
+        state: "executing",
+        args: %{"device_id" => device.id},
+        attempted_at: DateTime.add(DateTime.utc_now(), -360, :second)
+      })
+
+      # Recent jobs that should not be included
+      oban_job_fixture(%{
+        worker: "Towerops.Workers.DevicePollerWorker",
+        state: "executing",
+        args: %{"device_id" => device.id},
+        attempted_at: DateTime.add(DateTime.utc_now(), -30, :second)
+      })
+
+      stuck = JobMonitoring.list_stuck_jobs()
+
+      assert length(stuck) == 2
+      # Verify oldest first (discovery job at -360s)
+      assert Enum.at(stuck, 0).id == stuck_discovery.id
+      assert Enum.at(stuck, 1).id == stuck_poller.id
+    end
+
+    test "excludes jobs just under the threshold boundary" do
+      device = device_fixture()
+
+      # Polling job at 119 seconds - should NOT be stuck (under 120s threshold)
+      oban_job_fixture(%{
+        worker: "Towerops.Workers.DevicePollerWorker",
+        state: "executing",
+        args: %{"device_id" => device.id},
+        attempted_at: DateTime.add(DateTime.utc_now(), -119, :second)
+      })
+
+      # Discovery job at 299 seconds - should NOT be stuck (under 300s threshold)
+      oban_job_fixture(%{
+        worker: "Towerops.Workers.DiscoveryWorker",
+        state: "executing",
+        args: %{"device_id" => device.id},
+        attempted_at: DateTime.add(DateTime.utc_now(), -299, :second)
+      })
+
+      assert JobMonitoring.list_stuck_jobs() == []
+    end
+
+    test "includes jobs just over the threshold" do
+      device = device_fixture()
+
+      # Polling job at 125 seconds - should be stuck (over 120s threshold)
+      stuck_poller = oban_job_fixture(%{
+        worker: "Towerops.Workers.DevicePollerWorker",
+        state: "executing",
+        args: %{"device_id" => device.id},
+        attempted_at: DateTime.add(DateTime.utc_now(), -125, :second)
+      })
+
+      # Discovery job at 305 seconds - should be stuck (over 300s threshold)
+      stuck_discovery = oban_job_fixture(%{
+        worker: "Towerops.Workers.DiscoveryWorker",
+        state: "executing",
+        args: %{"device_id" => device.id},
+        attempted_at: DateTime.add(DateTime.utc_now(), -305, :second)
+      })
+
+      stuck = JobMonitoring.list_stuck_jobs()
+
+      assert length(stuck) == 2
+      assert stuck |> Enum.map(& &1.id) |> Enum.sort() == [stuck_discovery.id, stuck_poller.id] |> Enum.sort()
+    end
   end
 end