fix: simplify logic and enable correct recording rule groups for prom extension

src/k8s-extension/HISTORY.rst

@@ -2,6 +2,13 @@
 
 Release History
 ===============
+1.6.5
+++++++++++++++++++
+* microsoft.azuremonitor.containers.metrics: fix: simplify logic and enable correct recording rule groups for managed prom
+
+1.6.4
+++++++++++++++++++
+* microsoft.azuremonitor.containers.metrics: Update default region for azure monitor metrics extension in mooncake
 
 1.6.3
 ++++++++++++++++++

src/k8s-extension/azext_k8s_extension/partner_extensions/DataProtectionKubernetes.py

@@ -9,7 +9,7 @@
 from azure.cli.core.azclierror import RequiredArgumentMissingError, InvalidArgumentValueError
 
 from .DefaultExtension import DefaultExtension
-from .._client_factory import cf_storage, cf_managed_clusters
+from .._client_factory import cf_managed_clusters
 from ..vendored_sdks.models import (Extension, PatchExtension, Scope, ScopeCluster)
 
 logger = get_logger(__name__)
@@ -20,13 +20,16 @@ def __init__(self):
         """Constants for configuration settings
            - Tenant Id (required)
            - Backup storage location (required)
+           - Resource Requests (optional)
            - Resource Limits (optional)
         """
         self.TENANT_ID = "credentials.tenantId"
         self.BACKUP_STORAGE_ACCOUNT_CONTAINER = "configuration.backupStorageLocation.bucket"
         self.BACKUP_STORAGE_ACCOUNT_NAME = "configuration.backupStorageLocation.config.storageAccount"
         self.BACKUP_STORAGE_ACCOUNT_RESOURCE_GROUP = "configuration.backupStorageLocation.config.resourceGroup"
         self.BACKUP_STORAGE_ACCOUNT_SUBSCRIPTION = "configuration.backupStorageLocation.config.subscriptionId"
+        self.RESOURCE_REQUEST_CPU = "resources.requests.cpu"
+        self.RESOURCE_REQUEST_MEMORY = "resources.requests.memory"
         self.RESOURCE_LIMIT_CPU = "resources.limits.cpu"
         self.RESOURCE_LIMIT_MEMORY = "resources.limits.memory"
         self.BACKUP_STORAGE_ACCOUNT_USE_AAD = "configuration.backupStorageLocation.config.useAAD"
@@ -36,6 +39,8 @@ def __init__(self):
         self.storage_account = "storageAccount"
         self.storage_account_resource_group = "storageAccountResourceGroup"
         self.storage_account_subscription = "storageAccountSubscriptionId"
+        self.cpu_request = "cpuRequest"
+        self.memory_request = "memoryRequest"
         self.cpu_limit = "cpuLimit"
         self.memory_limit = "memoryLimit"
         self.use_aad = "useAAD"
@@ -46,6 +51,8 @@ def __init__(self):
             self.storage_account.lower(): self.BACKUP_STORAGE_ACCOUNT_NAME,
             self.storage_account_resource_group.lower(): self.BACKUP_STORAGE_ACCOUNT_RESOURCE_GROUP,
             self.storage_account_subscription.lower(): self.BACKUP_STORAGE_ACCOUNT_SUBSCRIPTION,
+            self.cpu_request.lower(): self.RESOURCE_REQUEST_CPU,
+            self.memory_request.lower(): self.RESOURCE_REQUEST_MEMORY,
             self.cpu_limit.lower(): self.RESOURCE_LIMIT_CPU,
             self.memory_limit.lower(): self.RESOURCE_LIMIT_MEMORY,
             self.use_aad.lower(): self.BACKUP_STORAGE_ACCOUNT_USE_AAD,
@@ -83,7 +90,7 @@ def Create(
         plan_publisher,
         plan_product,
     ):
-        # Current scope of DataProtection Kubernetes Backup extension is 'cluster' #TODO: add TSGs when they are in place
+        # Current scope of DataProtection extension is 'cluster' #TODO: add TSGs when they are in place
         if scope == 'namespace':
             raise InvalidArgumentValueError(f"Invalid scope '{scope}'. This extension can only be installed at 'cluster' scope.")
 
@@ -106,7 +113,7 @@ def Create(
         configuration_settings[self.TENANT_ID] = tenant_id
 
         if configuration_settings.get(self.BACKUP_STORAGE_ACCOUNT_USE_AAD) is None:
-            logger.warning("useAAD flag is not specified. Setting it to 'true'. Please provide extension MSI Storage Blob Data Contributor role to the storage account.")
+            logger.warning("useAAD flag is not specified. Setting it to 'true'. Please provide extension MSI Storage Blob Data Contributor role on the storage account.")
             configuration_settings[self.BACKUP_STORAGE_ACCOUNT_USE_AAD] = "true"
 
         if configuration_settings.get(self.BACKUP_STORAGE_ACCOUNT_STORAGE_ACCOUNT_URI) is None:
@@ -149,18 +156,20 @@ def Update(
             self.__validate_and_map_config(configuration_settings, validate_bsl=bsl_specified)
             if bsl_specified:
                 self.__validate_backup_storage_account(cmd.cli_ctx, resource_group_name, cluster_name, configuration_settings)
-        # this step is for brownfield migrating to AAD
+
+        # This step is for brownfield migrating to AAD
         if configuration_settings.get(self.BACKUP_STORAGE_ACCOUNT_USE_AAD) is not None and configuration_settings.get(self.BACKUP_STORAGE_ACCOUNT_USE_AAD).lower() == "true":
-            logger.warning("useAAD flag is set to true. Please provide extension MSI Storage Blob Data Contributor role to the storage account.")
+            logger.warning("useAAD flag is set to true. Please provide extension MSI Storage Blob Data Contributor role on the storage account.")
 
         if configuration_settings.get(self.BACKUP_STORAGE_ACCOUNT_STORAGE_ACCOUNT_URI) is None:
-            # SA details provided in user inputs, but did not provide SA URI.
-            logger.warning("storageAccountURI is not populated. Setting it to the storage account URI of provided storage account")
+            # SA details provided in user input, but did not provide SA URI.
             if bsl_specified:
+                logger.warning("storageAccountURI is not populated. Setting it to the storage account URI of provided storage account.")
                 configuration_settings[self.BACKUP_STORAGE_ACCOUNT_STORAGE_ACCOUNT_URI] = self.__get_storage_account_uri(cmd.cli_ctx, configuration_settings)
                 logger.warning(f"storageAccountURI: {configuration_settings[self.BACKUP_STORAGE_ACCOUNT_STORAGE_ACCOUNT_URI]}")
             # SA details not provided in user input, SA Uri not provided in user input, and also not populated in the original extension, we populate it.
             elif not bsl_specified and original_extension.configuration_settings.get(self.BACKUP_STORAGE_ACCOUNT_STORAGE_ACCOUNT_URI) is None:
+                logger.warning("storageAccountURI is not populated. Setting it to the storage account URI of the storage account provided during extension installation.")
                 configuration_settings[self.BACKUP_STORAGE_ACCOUNT_STORAGE_ACCOUNT_URI] = self.__get_storage_account_uri(cmd.cli_ctx, original_extension.configuration_settings)
                 logger.warning(f"storageAccountURI: {configuration_settings[self.BACKUP_STORAGE_ACCOUNT_STORAGE_ACCOUNT_URI]}")
 
@@ -177,7 +186,7 @@ def __get_tenant_id(self, cli_ctx):
         return cli_ctx.data['tenant_id']
 
     def __validate_and_map_config(self, configuration_settings, validate_bsl=True):
-        """Validate and set configuration settings for Data Protection K8sBackup extension"""
+        """Validate and set configuration settings for DataProtection extension"""
         input_configuration_settings = dict(configuration_settings.items())
         input_configuration_keys = [key.lower() for key in configuration_settings]
 
@@ -189,14 +198,14 @@ def __validate_and_map_config(self, configuration_settings, validate_bsl=True):
         for key in input_configuration_settings:
             _key = key.lower()
             if _key in self.configuration_mapping:
-                configuration_settings[self.configuration_mapping[_key]] = configuration_settings.pop(key)
+                configuration_settings[self.configuration_mapping[_key]] = configuration_settings.pop(key).strip()
             else:
                 configuration_settings.pop(key)
                 logger.warning(f"Ignoring unrecognized configuration setting: {key}")
 
     def __validate_backup_storage_account(self, cli_ctx, resource_group_name, cluster_name, configuration_settings):
         """Validations performed on the backup storage account
-           - Existance of the storage account
+           - Existence of the storage account
            - Cluster and storage account are in the same location
         """
         storage_account = self.__get_storage_account(cli_ctx, configuration_settings)

src/k8s-extension/azext_k8s_extension/partner_extensions/EntraWorkloadIAM.py

@@ -27,6 +27,25 @@
 CONFIG_SETTINGS_HELM_TENANT_ID = 'global.workload-iam.tenantID'
 CONFIG_SETTINGS_HELM_JOIN_TOKEN = 'workload-iam-local-authority.localAuthorityArgs.joinToken'
 
+def settings_dict_to_lowercase(settings):
+    """
+    Create new dictionary where the keys of the known user settings are all lowercase (but leave the
+    others as they were in case they are specific settings that have to be passed to the Helm chart).
+    """
+    all_user_settings = [CONFIG_SETTINGS_USER_TRUST_DOMAIN, CONFIG_SETTINGS_USER_TENANT_ID,
+                         CONFIG_SETTINGS_USER_LOCAL_AUTHORITY, CONFIG_SETTINGS_USER_JOIN_TOKEN]
+
+    if settings is None:
+        return dict()
+
+    validated_settings = dict()
+    for key, value in settings.items():
+        if key.lower() in all_user_settings:
+            validated_settings[key.lower()] = value
+        else:
+            validated_settings[key] = value
+
+    return validated_settings
 
 def settings_dict_to_lowercase(settings):
     """

src/k8s-extension/azext_k8s_extension/partner_extensions/azuremonitormetrics/deaults.py

@@ -8,7 +8,7 @@
 def get_default_region(cmd):
     cloud_name = cmd.cli_ctx.cloud.name
     if cloud_name.lower() == 'azurechinacloud':
-        raise CLIError("Azure China Cloud is not supported for the Azure Monitor Metrics addon")
+        return "chinanorth3"
     if cloud_name.lower() == 'azureusgovernment':
         return "usgovvirginia"
     return "eastus"

src/k8s-extension/azext_k8s_extension/partner_extensions/azuremonitormetrics/recordingrules/create.py

@@ -3,8 +3,9 @@
 # Licensed under the MIT License. See License.txt in the project root for license information.
 # --------------------------------------------------------------------------------------------
 import json
-from ..constants import ALERTS_API, RULES_API
 from knack.util import CLIError
+from ..constants import ALERTS_API, RULES_API
+from azure.cli.command_modules.acs.azuremonitormetrics.recordingrules.common import truncate_rule_group_name
 
 
 # pylint: disable=line-too-long
@@ -15,7 +16,18 @@ def get_recording_rules_template(cmd, azure_monitor_workspace_resource_id):
     url = f"{armendpoint}{azure_monitor_workspace_resource_id}/providers/microsoft.alertsManagement/alertRuleRecommendations?api-version={ALERTS_API}"
     r = send_raw_request(cmd.cli_ctx, "GET", url, headers=headers)
     data = json.loads(r.text)
-    return data['value']
+
+    filtered_templates = [
+        template for template in data.get('value', [])
+        if template.get("properties", {}).get("alertRuleType", "").lower() == "microsoft.alertsmanagement/prometheusrulegroups" and isinstance(template.get("properties", {}).get("rulesArmTemplate", {}).get("resources"), list) and all(
+            isinstance(rule, dict) and "record" in rule and "expression" in rule
+            for resource in template["properties"]["rulesArmTemplate"]["resources"]
+            if resource.get("type", "").lower() == "microsoft.alertsmanagement/prometheusrulegroups"
+            for rule in resource.get("properties", {}).get("rules", [])
+        )
+    ]
+
+    return filtered_templates
 
 
 # pylint: disable=line-too-long
@@ -39,8 +51,7 @@ def put_rules(cmd, default_rule_group_id, default_rule_group_name, mac_region, a
     for _ in range(3):
         try:
             headers = ['User-Agent=arc-azuremonitormetrics.put_rules.' + default_rule_group_name]
-            send_raw_request(cmd.cli_ctx, "PUT", url,
-                             body=body, headers=headers)
+            send_raw_request(cmd.cli_ctx, "PUT", url, body=body, headers=headers)
             break
         except CLIError as e:
             error = e
@@ -49,30 +60,30 @@ def put_rules(cmd, default_rule_group_id, default_rule_group_name, mac_region, a
 
 
 # pylint: disable=line-too-long
-def create_rules(cmd, cluster_subscription, cluster_resource_group_name, cluster_name, azure_monitor_workspace_resource_id, mac_region):
+def create_rules(cmd, cluster_subscription, cluster_resource_group_name, cluster_name, azure_monitor_workspace_resource_id, mac_region, raw_parameters):
     default_rules_template = get_recording_rules_template(cmd, azure_monitor_workspace_resource_id)
-    default_rule_group_name = "NodeRecordingRulesRuleGroup-{0}".format(cluster_name)
-    default_rule_group_id = "/subscriptions/{0}/resourceGroups/{1}/providers/Microsoft.AlertsManagement/prometheusRuleGroups/{2}".format(
-        cluster_subscription,
-        cluster_resource_group_name,
-        default_rule_group_name
-    )
-    url = "{0}{1}?api-version={2}".format(
-        cmd.cli_ctx.cloud.endpoints.resource_manager,
-        default_rule_group_id,
-        RULES_API
-    )
-    put_rules(cmd, default_rule_group_id, default_rule_group_name, mac_region, azure_monitor_workspace_resource_id, cluster_name, default_rules_template, url, True, 0)
 
-    default_rule_group_name = "KubernetesRecordingRulesRuleGroup-{0}".format(cluster_name)
-    default_rule_group_id = "/subscriptions/{0}/resourceGroups/{1}/providers/Microsoft.AlertsManagement/prometheusRuleGroups/{2}".format(
-        cluster_subscription,
-        cluster_resource_group_name,
-        default_rule_group_name
-    )
-    url = "{0}{1}?api-version={2}".format(
-        cmd.cli_ctx.cloud.endpoints.resource_manager,
-        default_rule_group_id,
-        RULES_API
-    )
-    put_rules(cmd, default_rule_group_id, default_rule_group_name, mac_region, azure_monitor_workspace_resource_id, cluster_name, default_rules_template, url, True, 1)
+    enable_windows_recording_rules = raw_parameters.get("enable_windows_recording_rules", False)
+
+    for index, rule_template in enumerate(default_rules_template):
+        rule_name = rule_template["name"]
+        is_windows_rule = "win" in rule_name.lower()
+
+        enable_rules = not (is_windows_rule and not enable_windows_recording_rules)
+
+        rule_group_name = truncate_rule_group_name(f"{rule_template['name']}-{cluster_name}")
+        rule_group_id = f"/subscriptions/{cluster_subscription}/resourceGroups/{cluster_resource_group_name}/providers/Microsoft.AlertsManagement/prometheusRuleGroups/{rule_group_name}"
+        url = f"{cmd.cli_ctx.cloud.endpoints.resource_manager}{rule_group_id}?api-version={RULES_API}"
+
+        put_rules(
+            cmd,
+            rule_group_id,
+            rule_group_name,
+            mac_region,
+            azure_monitor_workspace_resource_id,
+            cluster_name,
+            default_rules_template,
+            url,
+            enable_rules,
+            index
+        )

src/k8s-extension/azext_k8s_extension/partner_extensions/azuremonitormetrics/recordingrules/delete.py

@@ -35,3 +35,9 @@ def delete_rules(cmd, cluster_subscription, cluster_resource_group_name, cluster
         cluster_resource_group_name,
         "KubernetesRecordingRulesRuleGroup-{0}".format(cluster_name)
     )
+    delete_rule(
+        cmd,
+        cluster_subscription,
+        cluster_resource_group_name,
+        truncate_rule_group_name("UXRecordingRulesRuleGroup - {0}".format(cluster_name))
+    )

src/k8s-extension/setup.py

@@ -33,7 +33,7 @@
 # TODO: Add any additional SDK dependencies here
 DEPENDENCIES = []
 
-VERSION = "1.6.3"
+VERSION = "1.6.5"
 
 with open("README.rst", "r", encoding="utf-8") as f:
     README = f.read()

testing/.gitignore

@@ -0,0 +1,9 @@
+settings.json
+tmp/
+bin/*
+!bin/connectedk8s-1.0.0-py3-none-any.whl
+!bin/k8s_extension-0.4.0-py3-none-any.whl
+!bin/k8s_extension_private-0.1.0-py3-none-any.whl
+!bin/k8s_configuration-1.0.0-py3-none-any.whl
+!bin/connectedk8s-values.yaml
+*.xml

testing/Bootstrap.ps1

@@ -0,0 +1,80 @@
+param (
+    [switch] $SkipInstall,
+    [switch] $CI
+)
+
+# Disable confirm prompt for script
+az config set core.disable_confirm_prompt=true
+
+# Configuring the environment
+$ENVCONFIG = Get-Content -Path $PSScriptRoot/settings.json | ConvertFrom-Json
+
+az account set --subscription $ENVCONFIG.subscriptionId
+
+if (-not (Test-Path -Path $PSScriptRoot/tmp)) {
+    New-Item -ItemType Directory -Path $PSScriptRoot/tmp
+}
+
+if (!$SkipInstall) {
+    Write-Host "Removing the old connnectedk8s extension..."
+    az extension remove -n connectedk8s
+    Write-Host "Installing connectedk8s..."
+    az extension add -n connectedk8s
+    if (!$?) {
+        Write-Host "Unable to install connectedk8s, exiting..."
+        exit 1
+    }
+}
+
+Write-Host "Onboard cluster to Azure...starting!"
+
+az group show --name $envConfig.resourceGroup
+if (!$?) {
+    Write-Host "Resource group does not exist, creating it now in region 'eastus2euap'"
+    az group create --name $envConfig.resourceGroup --location eastus2euap
+
+    if (!$?) {
+        Write-Host "Failed to create Resource Group - exiting!"
+        Exit 1
+    }
+}
+
+# Skip creating the AKS Cluster if this is CI
+if (!$CI) {
+    az aks show -g $ENVCONFIG.resourceGroup -n $ENVCONFIG.aksClusterName
+    if (!$?) {
+        Write-Host "Cluster does not exist, creating it now"
+        az aks create -g $ENVCONFIG.resourceGroup -n $ENVCONFIG.aksClusterName --generate-ssh-keys
+    } else {
+        Write-Host "Cluster already exists, no need to create it."
+    }
+
+    Write-Host "Retrieving credentials for your AKS cluster..."
+
+    az aks get-credentials -g $ENVCONFIG.resourceGroup -n $ENVCONFIG.aksClusterName -f tmp/KUBECONFIG
+    if (!$?) 
+    {
+        Write-Host "Cluster did not create successfully, exiting!" -ForegroundColor Red
+        Exit 1
+    }
+    Write-Host "Successfully retrieved the AKS kubectl credentials"
+} else {
+    Copy-Item $HOME/.kube/config -Destination $PSScriptRoot/tmp/KUBECONFIG 
+}
+
+az connectedk8s show -g $ENVCONFIG.resourceGroup -n $ENVCONFIG.arcClusterName
+if ($?)
+{
+    Write-Host "Cluster is already connected, no need to re-connect"
+    Exit 0
+}
+
+Write-Host "Connecting the cluster to Arc with connectedk8s..."
+$Env:KUBECONFIG="$PSScriptRoot/tmp/KUBECONFIG"
+az connectedk8s connect -g $ENVCONFIG.resourceGroup -n $ENVCONFIG.arcClusterName -l uksouth
+if (!$?)
+{
+    kubectl get pods -A
+    Exit 1
+}
+Write-Host "Successfully onboarded the cluster to Azure"