[AKS] Cluster health monitor addon (#9674)

Author: zman-ms

src/aks-preview/HISTORY.rst

@@ -12,6 +12,11 @@ To release a new version, please select a new version number (usually plus 1 to
 Pending
 +++++++
 
+19.0.0b25
++++++++
+* `az aks create`: Add `--enable-continuous-control-plane-and-addon-monitor` to enable continuous control plane and addon monitor.
+* `az aks update`: Add `--enable-continuous-control-plane-and-addon-monitor` and `--disable-continuous-control-plane-and-addon-monitor` to manage continuous control plane and addon monitor.
+
 19.0.0b24
 +++++++
 * Vendor new SDK and bump API version to 2026-01-02-preview.

src/aks-preview/azext_aks_preview/_params.py

@@ -1201,6 +1201,12 @@ def load_arguments(self, _):
             help="Enable managed installation of Gateway API CRDs from the standard release channel."
         )
         c.argument("enable_hosted_system", action="store_true", is_preview=True)
+        c.argument(
+            "enable_continuous_control_plane_and_addon_monitor",
+            action="store_true",
+            is_preview=True,
+            help="Enable continuous control plane and addon monitor for the cluster.",
+        )
 
     with self.argument_context("aks update") as c:
         # managed cluster paramerters
@@ -1779,6 +1785,18 @@ def load_arguments(self, _):
             is_preview=True,
             help="Disable Application Load Balancer (Application Gateway for Containers)."
         )
+        c.argument(
+            "enable_continuous_control_plane_and_addon_monitor",
+            action="store_true",
+            is_preview=True,
+            help="Enable continuous control plane and addon monitor for the cluster.",
+        )
+        c.argument(
+            "disable_continuous_control_plane_and_addon_monitor",
+            action="store_true",
+            is_preview=True,
+            help="Disable continuous control plane and addon monitor for the cluster.",
+        )
 
     with self.argument_context("aks upgrade") as c:
         c.argument("kubernetes_version", completer=get_k8s_upgrades_completion_list)

src/aks-preview/azext_aks_preview/custom.py

@@ -1160,7 +1160,9 @@ def aks_create(
     enable_upstream_kubescheduler_user_configuration=False,
     # managed gateway installation
     enable_gateway_api=False,
-    enable_hosted_system=False
+    enable_hosted_system=False,
+    # health monitor
+    enable_continuous_control_plane_and_addon_monitor=False,
 ):
     # DO NOT MOVE: get all the original parameters and save them as a dictionary
     raw_parameters = locals()
@@ -1404,6 +1406,9 @@ def aks_update(
     # application load balancer
     enable_application_load_balancer=False,
     disable_application_load_balancer=False,
+    # health monitor
+    enable_continuous_control_plane_and_addon_monitor=False,
+    disable_continuous_control_plane_and_addon_monitor=False,
 ):
     # DO NOT MOVE: get all the original parameters and save them as a dictionary
     raw_parameters = locals()

src/aks-preview/azext_aks_preview/managed_cluster_decorator.py

@@ -3767,6 +3767,20 @@ def get_enable_hosted_system(self) -> bool:
             raise RequiredArgumentMissingError('"--enable-hosted-system" requires "--sku automatic".')
         return enable_hosted_system
 
+    def get_enable_continuous_control_plane_and_addon_monitor(self) -> bool:
+        """Obtain the value of enable_continuous_control_plane_and_addon_monitor.
+
+        :return: bool
+        """
+        return self.raw_param.get("enable_continuous_control_plane_and_addon_monitor")
+
+    def get_disable_continuous_control_plane_and_addon_monitor(self) -> bool:
+        """Obtain the value of disable_continuous_control_plane_and_addon_monitor.
+
+        :return: bool
+        """
+        return self.raw_param.get("disable_continuous_control_plane_and_addon_monitor")
+
 
 # pylint: disable=too-many-public-methods
 class AKSPreviewManagedClusterCreateDecorator(AKSManagedClusterCreateDecorator):
@@ -4796,6 +4810,22 @@ def set_up_bootstrap_profile(self, mc: ManagedCluster) -> ManagedCluster:
 
         return mc
 
+    def set_up_health_monitor_profile(self, mc: ManagedCluster) -> ManagedCluster:
+        """Set up health monitor profile for the ManagedCluster object.
+
+        :return: the ManagedCluster object
+        """
+        self._ensure_mc(mc)
+
+        if self.context.get_enable_continuous_control_plane_and_addon_monitor():
+            if mc.health_monitor_profile is None:
+                mc.health_monitor_profile = (
+                    self.models.ManagedClusterHealthMonitorProfile()  # pylint: disable=no-member
+                )
+            mc.health_monitor_profile.enable_continuous_control_plane_and_addon_monitor = True
+
+        return mc
+
     def set_up_static_egress_gateway(self, mc: ManagedCluster) -> ManagedCluster:
         self._ensure_mc(mc)
 
@@ -4920,6 +4950,8 @@ def construct_mc_profile_preview(self, bypass_restore_defaults: bool = False) ->
         mc = self.set_up_bootstrap_profile(mc)
         # set up static egress gateway profile
         mc = self.set_up_static_egress_gateway(mc)
+        # set up health monitor profile
+        mc = self.set_up_health_monitor_profile(mc)
         # set up imds restriction(a property in network profile)
         mc = self.set_up_imds_restriction(mc)
         # set up user-defined scheduler configuration for kube-scheduler upstream
@@ -7132,6 +7164,32 @@ def update_static_egress_gateway(self, mc: ManagedCluster) -> ManagedCluster:
             mc.network_profile.static_egress_gateway_profile.enabled = False
         return mc
 
+    def update_health_monitor_profile(self, mc: ManagedCluster) -> ManagedCluster:
+        """Update health monitor profile for the ManagedCluster object.
+
+        :return: the ManagedCluster object
+        """
+        self._ensure_mc(mc)
+
+        enable = self.context.get_enable_continuous_control_plane_and_addon_monitor()
+        disable = self.context.get_disable_continuous_control_plane_and_addon_monitor()
+
+        if not enable and not disable:
+            return mc
+        if enable and disable:
+            raise MutuallyExclusiveArgumentError(
+                "Cannot specify --enable-continuous-control-plane-and-addon-monitor and "
+                "--disable-continuous-control-plane-and-addon-monitor at the same time."
+            )
+
+        if mc.health_monitor_profile is None:
+            mc.health_monitor_profile = (
+                self.models.ManagedClusterHealthMonitorProfile()  # pylint: disable=no-member
+            )
+        mc.health_monitor_profile.enable_continuous_control_plane_and_addon_monitor = bool(enable)
+
+        return mc
+
     def update_imds_restriction(self, mc: ManagedCluster) -> ManagedCluster:
         """Update imds restriction for the ManagedCluster object.
 
@@ -7552,6 +7610,8 @@ def update_mc_profile_preview(self) -> ManagedCluster:
         mc = self.update_bootstrap_profile(mc)
         # update static egress gateway
         mc = self.update_static_egress_gateway(mc)
+        # update health monitor profile
+        mc = self.update_health_monitor_profile(mc)
         # update imds restriction
         mc = self.update_imds_restriction(mc)
         # update VMAS to VMS