Integration test for bluechi-is-online node <name> --monitor

tests/bluechi_test/bluechi_is_online.py

@@ -57,3 +57,12 @@ def system_is_online(self) -> bool:
             0,
         )
         return result == 0
+
+    def monitor_node(self, node_name: str) -> bool:
+        result, output = self.run(
+            f"Monitoring status of node {node_name}.",
+            f"node {node_name} --monitor",
+            False,
+            0,
+        )
+        return result == 0

tests/tests/tier0/bluechi-is-online-node-monitor/main.fmf

@@ -0,0 +1,2 @@
+summary: --monitor keeps monitoring as long as node is online and exits if it detects an offline state.
+id: 929f11ba-00ab-4c4f-8614-e1aebcbaa96d

tests/tests/tier0/bluechi-is-online-node-monitor/test_bluechi_is_online_node_monitor.py

@@ -0,0 +1,81 @@
+#
+# Copyright Contributors to the Eclipse BlueChi project
+#
+# SPDX-License-Identifier: LGPL-2.1-or-later
+
+import logging
+import threading
+from typing import Dict
+
+from bluechi_test.config import BluechiAgentConfig, BluechiControllerConfig
+from bluechi_test.machine import BluechiAgentMachine, BluechiControllerMachine
+from bluechi_test.test import BluechiTest
+from bluechi_test.util import Timeout
+
+LOGGER = logging.getLogger(__name__)
+
+NODE_FOO = "node-foo"
+AGENT_ONE = "agent-one"
+
+
+class MonitorResult:
+    def __init__(self):
+        self.result = None
+
+
+def monitor_command(
+    ctrl: BluechiControllerMachine, node_name: str, monitor_result: MonitorResult
+):
+    monitor_result.result = ctrl.bluechi_is_online.monitor_node(node_name)
+
+
+def exec(ctrl: BluechiControllerMachine, nodes: Dict[str, BluechiAgentMachine]):
+    node_foo = nodes[NODE_FOO]
+
+    # # Test 1: Agent and node are running, no monitor output expected
+    LOGGER.debug("Starting NODE_FOO.")
+    monitor_result_test_one = MonitorResult()
+    with Timeout(2, f"Timeout while monitoring {NODE_FOO}"):
+        monitor_thread = threading.Thread(
+            target=monitor_command, args=(ctrl, NODE_FOO, monitor_result_test_one)
+        )
+
+    monitor_thread.start()
+    assert (
+        monitor_result_test_one.result is None
+    ), "Monitor command should not produce output when node is running."
+
+    # Test 2: Stop NODE_FOO and verify monitoring detects the failure
+    LOGGER.debug("Starting monitor thread before stopping NODE_FOO.")
+    monitor_result_test_two = MonitorResult()
+    monitor_thread = threading.Thread(
+        target=monitor_command, args=(ctrl, NODE_FOO, monitor_result_test_two)
+    )
+    monitor_thread.start()
+
+    LOGGER.debug("Stopping NODE_FOO.")
+    node_foo.systemctl.stop_unit("bluechi-agent")
+    assert node_foo.wait_for_unit_state_to_be("bluechi-agent", "inactive")
+    assert (
+        monitor_result_test_two.result is not None
+    ), "Monitor command should produce output when NODE_FOO is stopped."
+
+
+def test_bluechi_is_online_node_monitor(
+    bluechi_test: BluechiTest,
+    bluechi_node_default_config: BluechiAgentConfig,
+    bluechi_ctrl_default_config: BluechiControllerConfig,
+):
+    node_foo_cfg = bluechi_node_default_config.deep_copy()
+    node_foo_cfg.node_name = NODE_FOO
+
+    agent_one_cfg = bluechi_node_default_config.deep_copy()
+    agent_one_cfg.node_name = AGENT_ONE
+
+    bluechi_ctrl_default_config.allowed_node_names = [NODE_FOO, AGENT_ONE]
+
+    bluechi_test.set_bluechi_controller_config(bluechi_ctrl_default_config)
+    bluechi_test.add_bluechi_agent_config(node_foo_cfg)
+    bluechi_test.add_bluechi_agent_config(agent_one_cfg)
+
+    bluechi_test.run(exec)