Azure Kubernetes Service Cluster

Plugin: go.d.plugin Module: azure_monitor

Overview

Monitor AKS cluster health including API server and etcd resource usage, pod scheduling status and readiness, node capacity and conditions, cluster autoscaler behavior, and per-node CPU, memory, disk, and network utilization.

The collector uses Azure SDK clients for:

• Authentication via Entra ID (service principal, managed identity, or default credentials)
• Resource discovery via Azure Resource Graph queries
• Metrics collection via Azure Monitor Metrics batch API, grouped by region and time grain

This collector is supported on all platforms.

This collector supports collecting metrics from multiple instances of this integration, including remote instances.

The monitoring principal needs read access to Azure Resource Graph and Azure Monitor metrics for target resources.

Default Behavior

Auto-Detection

When profiles includes auto (the default), the collector queries Azure Resource Graph to discover which resource types exist in the subscription and enables matching built-in profiles automatically.

Limits

Azure Monitor metrics granularity is typically 1 minute. The collector enforces a minimum collection interval of 60 seconds.

Performance Impact

The collector uses bounded request concurrency and batches resources and metrics to minimize API calls. Default limits: 4 concurrent queries, 50 resources per batch, 20 metrics per query.

Metrics

Metrics grouped by scope.

The scope defines the instance that the metric belongs to. An instance is uniquely identified by a set of labels.

Per resource

These metrics refer to each monitored Azure resource.

Labels:

Label	Description
resource_name	The Azure resource name.
resource_group	The Azure resource group.
region	The Azure region where the resource is deployed.
resource_type	The Azure resource type identifier.
profile	The Azure Monitor profile id.
resource_uid	The unique Azure resource identifier.

Metrics:

Metric	Dimensions	Unit
azure_monitor.aks.apiserver_cpu	average, maximum	percentage
azure_monitor.aks.apiserver_memory	average, maximum	percentage
azure_monitor.aks.apiserver_inflight_requests	average	requests
azure_monitor.aks.etcd_cpu	average, maximum	percentage
azure_monitor.aks.etcd_memory	average, maximum	percentage
azure_monitor.aks.etcd_database	average, maximum	percentage
azure_monitor.aks.pod_status_phase	average	pods
azure_monitor.aks.pod_status_ready	average	pods
azure_monitor.aks.allocatable_cpu	average	cores
azure_monitor.aks.allocatable_memory	average	bytes
azure_monitor.aks.node_conditions	average	nodes
azure_monitor.aks.autoscaler_health	safe_to_autoscale, cooldown	state
azure_monitor.aks.autoscaler_unschedulable_pods	average	pods
azure_monitor.aks.autoscaler_unneeded_nodes	average	nodes
azure_monitor.aks.node_cpu_millicores	average, maximum	millicores
azure_monitor.aks.node_cpu_percentage	average, maximum	percentage
azure_monitor.aks.node_memory_working_set	average, maximum	bytes
azure_monitor.aks.node_memory_working_set_percentage	average, maximum	percentage
azure_monitor.aks.node_memory_rss	average, maximum	bytes
azure_monitor.aks.node_memory_rss_percentage	average, maximum	percentage
azure_monitor.aks.node_disk_usage	average, maximum	bytes
azure_monitor.aks.node_disk_percentage	average, maximum	percentage
azure_monitor.aks.node_network	in, out	bytes

Alerts

The following alerts are available:

Alert name	On metric	Description
am_aks_apiserver_cpu	azure_monitor.aks.apiserver_cpu	AKS API server CPU on ${label:resource_name}
am_aks_apiserver_memory	azure_monitor.aks.apiserver_memory	AKS API server memory on ${label:resource_name}
am_aks_apiserver_inflight_requests	azure_monitor.aks.apiserver_inflight_requests	AKS API server inflight requests on ${label:resource_name}
am_aks_etcd_cpu	azure_monitor.aks.etcd_cpu	AKS etcd CPU on ${label:resource_name}
am_aks_etcd_memory	azure_monitor.aks.etcd_memory	AKS etcd memory on ${label:resource_name}
am_aks_etcd_database	azure_monitor.aks.etcd_database	AKS etcd database usage on ${label:resource_name}
am_aks_autoscaler_safe_to_autoscale	azure_monitor.aks.autoscaler_health	AKS autoscaler unsafe on ${label:resource_name}
am_aks_autoscaler_unschedulable_pods	azure_monitor.aks.autoscaler_unschedulable_pods	AKS unschedulable pods on ${label:resource_name}
am_aks_node_cpu	azure_monitor.aks.node_cpu_percentage	AKS node CPU on ${label:resource_name}
am_aks_node_memory_working_set	azure_monitor.aks.node_memory_working_set_percentage	AKS node memory working set on ${label:resource_name}
am_aks_node_memory_rss	azure_monitor.aks.node_memory_rss_percentage	AKS node memory RSS on ${label:resource_name}
am_aks_node_disk	azure_monitor.aks.node_disk_percentage	AKS node disk usage on ${label:resource_name}

Setup

You can configure the azure_monitor collector in two ways:

Method	Best for	How to
UI	Fast setup without editing files	Go to Nodes → Configure this node → Collectors → Jobs, search for azure_monitor, then click + to add a job.
File	If you prefer configuring via file, or need to automate deployments (e.g., with Ansible)	Edit go.d/azure_monitor.conf and add a job.

important

UI configuration requires paid Netdata Cloud plan.

Prerequisites

Create an Azure monitoring principal

Create a service principal or use a managed identity with the following permissions:

1. Monitoring Reader role on the target subscription or resource groups (for Azure Monitor metrics access)
2. Reader role for Azure Resource Graph queries (for resource discovery)

For service principal authentication:

# Create the service principal
az ad sp create-for-rbac --name "netdata-monitor" --role "Monitoring Reader" \
  --scopes /subscriptions/<subscription-id>

# Note the appId (client_id), password (client_secret), and tenant

For managed identity (on Azure VMs, VMSS, or AKS):

# Assign Monitoring Reader role to the VM's managed identity
az role assignment create --assignee <managed-identity-principal-id> \
  --role "Monitoring Reader" --scope /subscriptions/<subscription-id>

Configuration

Options

The following options can be defined globally: update_every, autodetection_retry.

Profile files are loaded from:

• Stock: /usr/lib/netdata/conf.d/go.d/azure_monitor.profiles/default/
• User: /etc/netdata/go.d/azure_monitor.profiles/

User profile files with the same filename override stock profiles.

Config options

Group	Option	Description	Default	Required
Collection	update_every	Data collection interval (seconds). Must be at least 60.	60	no
	autodetection_retry	Autodetection retry interval (seconds). Set 0 to disable.	0	no
Target	subscription_id	Azure subscription ID.		yes
	cloud	Azure cloud environment: public, government, or china.	public	no
Collection	discovery_every	Resource discovery interval in seconds.	300	no
	query_offset	Offset in seconds for metric query windows. Increase if metrics appear incomplete.	180	no
	timeout	Timeout for Azure Resource Graph and Azure Monitor API requests, in seconds.	30	no
Limits	max_concurrency	Maximum concurrent batch queries to Azure Monitor.	4	no
	max_batch_resources	Maximum resources per Azure Monitor batch request.	50	no
	max_metrics_per_query	Maximum metrics per Azure Monitor batch request.	20	no
Profiles	profiles	Profile ids to enable. Use auto to discover resource types via Azure Resource Graph and enable matching profiles. Combine with explicit ids: [auto, custom_profile].	[auto]	no
Filters	resource_groups	Optional list of resource group names to restrict monitoring scope.	[]	no
Authentication	auth.mode	Authentication mode: service_principal, managed_identity, or default.		yes
	auth.mode_service_principal.tenant_id	Entra ID tenant ID (required for service_principal mode).		no
	auth.mode_service_principal.client_id	Entra ID application (client) ID (required for service_principal mode).		no
	auth.mode_service_principal.client_secret	Entra ID client secret (required for service_principal mode).		no
	auth.mode_managed_identity.client_id	Client ID for user-assigned managed identity. Leave empty for system-assigned.		no
Virtual Node	vnode	Associates this data collection job with a Virtual Node.		no

via UI

Configure the azure_monitor collector from the Netdata web interface:

1. Go to Nodes.
2. Select the node where you want the azure_monitor data-collection job to run and click the ⚙ (Configure this node). That node will run the data collection.
3. The Collectors → Jobs view opens by default.
4. In the Search box, type azure_monitor (or scroll the list) to locate the azure_monitor collector.
5. Click the + next to the azure_monitor collector to add a new job.
6. Fill in the job fields, then click Test to verify the configuration and Submit to save.
   • Test runs the job with the provided settings and shows whether data can be collected.
   • If it fails, an error message appears with details (for example, connection refused, timeout, or command execution errors), so you can adjust and retest.

via File

The configuration file name for this integration is go.d/azure_monitor.conf.

The file format is YAML. Generally, the structure is:

update_every: 1
autodetection_retry: 0
jobs:
  - name: some_name1
  - name: some_name2

You can edit the configuration file using the edit-config script from the Netdata config directory.

cd /etc/netdata 2>/dev/null || cd /opt/netdata/etc/netdata
sudo ./edit-config go.d/azure_monitor.conf

Examples

Service principal (auto-discover all resources)

Authenticate with a service principal and auto-discover all supported Azure resource types in the subscription.

jobs:
  - name: prod
    subscription_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
    auth:
      mode: service_principal
      mode_service_principal:
        tenant_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
        client_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
        client_secret: "your-client-secret"

Managed identity (Azure VM/VMSS/AKS)

Use the managed identity of the Azure VM, VMSS, or AKS node where Netdata is running.

Config

jobs:
  - name: prod
    subscription_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
    auth:
      mode: managed_identity

Specific profiles only

Monitor only specific Azure services instead of auto-discovering all resource types.

Config

jobs:
  - name: databases
    subscription_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
    profiles:
      - sql_database
      - postgres_flexible
      - redis_cache
    auth:
      mode: service_principal
      mode_service_principal:
        tenant_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
        client_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
        client_secret: "your-client-secret"

Filter by resource group

Only monitor resources in specific resource groups.

Config

jobs:
  - name: prod-rg
    subscription_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
    resource_groups:
      - production-rg
      - staging-rg
    auth:
      mode: default

Azure Government cloud

Connect to Azure Government cloud environment.

Config

jobs:
  - name: gov
    subscription_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
    cloud: government
    auth:
      mode: service_principal
      mode_service_principal:
        tenant_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
        client_id: "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
        client_secret: "your-client-secret"

Troubleshooting

Debug Mode

Important: Debug mode is not supported for data collection jobs created via the UI using the Dyncfg feature.

To troubleshoot issues with the azure_monitor collector, run the go.d.plugin with the debug option enabled. The output should give you clues as to why the collector isn't working.

• Navigate to the plugins.d directory, usually at /usr/libexec/netdata/plugins.d/. If that's not the case on your system, open netdata.conf and look for the plugins setting under [directories].

  cd /usr/libexec/netdata/plugins.d/

• Switch to the netdata user.

  sudo -u netdata -s

• Run the go.d.plugin to debug the collector:

  ./go.d.plugin -d -m azure_monitor

  To debug a specific job:

  ./go.d.plugin -d -m azure_monitor -j jobName

Getting Logs

If you're encountering problems with the azure_monitor collector, follow these steps to retrieve logs and identify potential issues:

• Run the command specific to your system (systemd, non-systemd, or Docker container).
• Examine the output for any warnings or error messages that might indicate issues. These messages should provide clues about the root cause of the problem.

System with systemd

Use the following command to view logs generated since the last Netdata service restart:

journalctl _SYSTEMD_INVOCATION_ID="$(systemctl show --value --property=InvocationID netdata)" --namespace=netdata --grep azure_monitor

System without systemd

Locate the collector log file, typically at /var/log/netdata/collector.log, and use grep to filter for collector's name:

grep azure_monitor /var/log/netdata/collector.log

Note: This method shows logs from all restarts. Focus on the latest entries for troubleshooting current issues.

Docker Container

If your Netdata runs in a Docker container named "netdata" (replace if different), use this command:

docker logs netdata 2>&1 | grep azure_monitor

No metrics are collected

Verify the following:

1. The service principal or managed identity has Monitoring Reader role on the subscription or resource group.
2. The subscription_id in the configuration matches the subscription containing the target resources.
3. Target resources are running and producing metrics (check Azure Portal > Metrics for the resource).
4. Check the Netdata error log for authentication or API errors: grep azure_monitor /var/log/netdata/error.log.

Missing metrics for some resource types

Azure Monitor profiles are matched by resource type. If a resource type exists but no metrics appear:

1. Ensure profiles: [auto] (default) is set, or the specific profile id is listed.
2. Verify the resource type matches a built-in profile. Run ls /usr/lib/netdata/conf.d/go.d/azure_monitor.profiles/default/ to see available profiles.
3. Some metrics require the resource to be actively processing data (e.g., IoT Hub telemetry metrics only appear when devices send messages).

Metrics appear delayed

Azure Monitor metrics have a built-in reporting delay of 1-3 minutes. The collector uses a query_offset (default: 180 seconds) to account for this. If metrics are missing or incomplete, try increasing query_offset to 240 or 300 seconds. Some metrics with longer time grains (e.g., PT5M) may take up to 5 minutes to appear.

Authentication errors in sovereign clouds

For Azure Government or Azure China clouds, set the cloud parameter:

• Azure Government: cloud: government
• Azure China (21Vianet): cloud: china

Ensure the service principal is registered in the correct cloud tenant.