[AKS] az aks upgrade: Skip Machines mode agent pools during node image and Kubernetes version upgrade

src/azure-cli/HISTORY.rst

@@ -16,6 +16,7 @@ Release History
 
 **AKS**
 
+* `az aks upgrade`: Skip Machines mode agent pools during `--node-image-only` and Kubernetes version upgrade, which do not support these operations, instead of surfacing the raw RP OperationNotAllowed error
 * `az aks nodepool upgrade`: Fix `--max-unavailable` being silently ignored (#33215)
 * `az aks maintenanceconfiguration add/update`: Add support for maintenanceWindow format in default maintenance configuration (#33431)
 * `az aks check-acr`: Support national/sovereign clouds where nodes report `cloud=AzureStackCloud` by pointing canipull at the on-node `akscustom.json` environment file (#33551)

src/azure-cli/azure/cli/command_modules/acs/_consts.py

@@ -26,6 +26,7 @@
 CONST_NODEPOOL_MODE_SYSTEM = "System"
 CONST_NODEPOOL_MODE_USER = "User"
 CONST_NODEPOOL_MODE_GATEWAY = "Gateway"
+CONST_NODEPOOL_MODE_MACHINES = "Machines"
 
 # os type
 CONST_DEFAULT_NODE_OS_TYPE = "Linux"

src/azure-cli/azure/cli/command_modules/acs/custom.py

@@ -61,6 +61,7 @@
     CONST_MONITORING_ADDON_NAME,
     CONST_MONITORING_LOG_ANALYTICS_WORKSPACE_RESOURCE_ID,
     CONST_MONITORING_USING_AAD_MSI_AUTH,
+    CONST_NODEPOOL_MODE_MACHINES,
     CONST_NODEPOOL_MODE_USER,
     CONST_OPEN_SERVICE_MESH_ADDON_NAME,
     CONST_ROTATION_POLL_INTERVAL,
@@ -1320,6 +1321,10 @@ def aks_upgrade(cmd,
             if vmas_cluster:
                 raise CLIError('This cluster is using AvailabilitySet. Node image upgrade only operation '
                                'can only be applied on VirtualMachineScaleSets or VirtualMachines cluster.')
+            # Skip Machines mode pools to avoid a known client-side error: these pools are containers of individual machines and do not support node image version upgrade.
+            if agent_pool_profile.mode == CONST_NODEPOOL_MODE_MACHINES:
+                logger.warning("Skipping node image upgrade for agent pool '%s': Machines mode pools do not support node image version upgrade.", agent_pool_profile.name)
+                continue
             agent_pool_client = cf_agent_pools(cmd.cli_ctx)
             _upgrade_single_nodepool_image_version(True, agent_pool_client,
                                                    resource_group_name, name, agent_pool_profile.name)
@@ -1381,6 +1386,10 @@ def aks_upgrade(cmd,
 
     if upgrade_all:
         for agent_profile in (instance.agent_pool_profiles or []):
+            # Skip Machines mode pools to avoid a known client-side error: these pools are containers of individual machines and do not support Kubernetes version upgrade.
+            if agent_profile.mode == CONST_NODEPOOL_MODE_MACHINES:
+                logger.warning("Skipping Kubernetes version upgrade for agent pool '%s': Machines mode pools do not support Kubernetes version upgrade.", agent_profile.name)
+                continue
             agent_profile.orchestrator_version = kubernetes_version
             agent_profile.creation_data = None
 

src/azure-cli/azure/cli/command_modules/acs/tests/latest/test_custom.py

@@ -30,6 +30,7 @@
     aks_agentpool_upgrade,
     aks_enable_addons,
     aks_stop,
+    aks_upgrade,
     is_monitoring_addon_enabled,
     k8s_install_kubectl,
     k8s_install_kubelogin,
@@ -971,6 +972,51 @@ def test_aks_stop(self):
         )
         self.assertEqual(aks_stop(self.cmd, self.client, "rg", "name", False), None)
 
+    def test_aks_upgrade_node_image_only_skips_machines_mode_pool(self):
+        """Machines mode pools must be skipped during --node-image-only to avoid a known client-side error."""
+        machines_pool = self.models.ManagedClusterAgentPoolProfile(name="machinespool", mode="Machines", type="VirtualMachines")
+        vmss_pool = self.models.ManagedClusterAgentPoolProfile(name="nodepool1", mode="User", type="VirtualMachineScaleSets")
+        mc = self.models.ManagedCluster(location="test_location")
+        mc.agent_pool_profiles = [machines_pool, vmss_pool]
+        mc.pod_identity_profile = None
+        mc.kubernetes_version = "1.24.0"
+        mc.provisioning_state = "Succeeded"
+        mc.max_agent_pools = 10
+
+        self.client.get = mock.Mock(return_value=mc)
+
+        with mock.patch("azure.cli.command_modules.acs.custom.cf_agent_pools") as mock_cf, \
+             mock.patch("azure.cli.command_modules.acs.custom._upgrade_single_nodepool_image_version") as mock_upgrade:
+            mock_cf.return_value = mock.Mock()
+
+            aks_upgrade(self.cmd, self.client, "rg", "name", node_image_only=True, yes=True)
+
+            # Only the VMSS pool should be upgraded; the Machines mode pool must be skipped.
+            upgraded_pools = [call.args[4] for call in mock_upgrade.call_args_list]
+            self.assertNotIn("machinespool", upgraded_pools)
+            self.assertIn("nodepool1", upgraded_pools)
+
+    def test_aks_upgrade_kubernetes_version_skips_machines_mode_pool(self):
+        """Machines mode pools must be skipped during Kubernetes version upgrade to avoid a known client-side error."""
+        machines_pool = self.models.ManagedClusterAgentPoolProfile(name="machinespool", mode="Machines", type="VirtualMachines")
+        vmss_pool = self.models.ManagedClusterAgentPoolProfile(name="nodepool1", mode="User", type="VirtualMachineScaleSets")
+        mc = self.models.ManagedCluster(location="test_location")
+        mc.agent_pool_profiles = [machines_pool, vmss_pool]
+        mc.pod_identity_profile = None
+        mc.kubernetes_version = "1.24.0"
+        mc.provisioning_state = "Succeeded"
+        mc.max_agent_pools = 10
+        mc.service_principal_profile = None
+
+        self.client.get = mock.Mock(return_value=mc)
+        self.client.begin_create_or_update = mock.Mock(return_value=None)
+
+        aks_upgrade(self.cmd, self.client, "rg", "name", kubernetes_version="1.25.0", yes=True)
+
+        # Machines mode pool must not have orchestrator_version set; VMSS pool must be upgraded.
+        self.assertIsNone(machines_pool.orchestrator_version)
+        self.assertEqual(vmss_pool.orchestrator_version, "1.25.0")
+
 
 class TestRunCommand(unittest.TestCase):
     def test_get_command_context_invalid_file(self):