Running a self-managed Prometheus agent inside your Kubernetes cluster consumes resources (3 CPU cores and 4 GB memory by default) and limits metric retention to 7 days. Container Monitoring Pro Edition replaces the in-cluster agent with a fully managed collector, extends metric retention to 90 days, and provides comprehensive prebuilt dashboards with a 99.95% SLA.
After you enable Pro Edition, you cannot downgrade to Basic Edition.
Why upgrade to Pro Edition
| Capability | Basic Edition | Pro Edition |
|---|---|---|
| Metric retention for basic container cluster metrics | 7 days | 90 days |
| Prometheus collector | Self-managed agent in your cluster (3 CPU cores, 4 GB memory by default) | Fully managed agent -- no in-cluster resource cost, production-level SLA of 99.95% |
| Dashboards | Basic monitoring dashboards | Comprehensive monitoring dashboards |
Supported cluster types
Pro Edition supports the following Container Service for Kubernetes (ACK) cluster types:
ACK managed Pro cluster
ACK Lingjun cluster
ACK dedicated cluster
Billing
Container Monitoring Pro Edition has two billing components.
Cluster scale fee
This fee is based on Observability Capacity Unit (OCU) usage, calculated from your cluster node count. OCU is a billing unit introduced by Alibaba Cloud Native Observability that automatically calculates usage based on hourly resource usage.
| Item | Detail |
|---|---|
| Conversion | Every 10 cluster nodes = 1 OCU (rounded up) |
| Unit price | 0.023 USD per OCU per hour |
| Billing method | Pay-as-you-go |
| Billing cycle | Hourly, aggregated into a daily charge |
How it works: Each hour, the system records the maximum number of nodes in your cluster and converts that number to OCUs. At the end of each day, it sums the hourly OCU values and multiplies by the unit price.
Example: For a cluster with 35 nodes:
Hourly OCU = ceil(35 / 10) = 4 OCUs
Daily cost = 4 OCUs x 24 hours x 0.023 USD = 2.21 USD/day
Prometheus instance fee
Prometheus instance fees are billed separately. For details, see Prometheus instance billing.
Prerequisites
Before you enable Pro Edition, complete the following steps:
Activate Managed Service for Prometheus with one of these pay-as-you-go billing modes:
Enable Pro Edition
Choose the method that matches your situation:
New cluster? Select Pro Edition during integration. See Select Pro Edition during integration.
Existing Basic Edition? Upgrade your current integration. See Upgrade from Basic Edition to Pro Edition.
Select Pro Edition during integration
Go to the Integration Center page and select Kubernetes Cluster Monitoring.
In the Kubernetes Cluster Monitoring panel, select the cluster to integrate, choose Container Monitoring Pro Edition, and click OK.
Upgrade from Basic Edition to Pro Edition
Go to the Integration Management page and choose Integrated Environments > Container Service.
Find your cluster and click Upgrade to Pro Edition in the Actions column. In the dialog box, click OK.
Supported dashboards
All dashboards are automatically available after you enable Pro Edition.
| Category | Dashboard |
|---|---|
| Monitoring overview | Cluster monitoring overview |
| Cluster namespace dashboard | |
| Cluster core components | ACK Pro API server |
| ACK Pro ETCD | |
| ACK Pro Scheduler | |
| ACK Pro Cloud Controller Manager | |
| ACK Pro Kube Controller Manager | |
| Node monitoring | Node pool overview |
| Cluster node monitoring details | |
| Application monitoring | StatefulSet monitoring |
| Deployment monitoring | |
| Daemon process set application monitoring | |
| Cluster Pod monitoring | |
| Network monitoring | CoreDNS component monitoring |
| Cluster Ingress traffic monitoring | |
| Storage monitoring | CSI storage component monitoring-cluster level |
| CSI storage component monitoring-node level | |
| Pod IO Monitoring (Pod Level) | |
| Frontend Storage IO Monitoring (Cluster Level) | |
| GPU monitoring | Cluster GPU monitoring-cluster level |
| Cluster GPU monitoring-node level | |
| Cluster GPU monitoring-application Pod dimension | |
| Cost analysis/Resource optimization | Resource profile |
| Others | Backend Storage IO Monitoring (Cluster Level) |
| k8s-reclaimed-resource | |
| Cluster Prometheus self-monitoring | |
| Virtual Node(ECI) Overview |
Default alert rules
The following alert rules are active by default after you enable Pro Edition.
Node alerts
| Alert rule | Alert template |
|---|---|
| Node CPU usage greater than 75% | Node {{ $labels.instance }} CPU usage greater than 75%, current CPU usage {{ printf "%.2f" $value }}% |
| Node CPU usage greater than 85% | Node {{ $labels.instance }} CPU usage greater than 85%, current CPU usage {{ printf "%.2f" $value }}% |
| Node memory usage greater than 75% | Node {{ $labels.instance }} memory usage greater than 75%, current memory usage {{ printf "%.2f" $value }}% |
| Node memory usage greater than 85% | Node {{ $labels.instance }} memory usage greater than 85%, current memory usage {{ printf "%.2f" $value }}% |
| Node anomalies | Node {{$labels.node}} has been in unavailable status for more than 10 minutes |
| Disk usage greater than 95% | Node {{ $labels.instance }} disk {{ $labels.device }} usage exceeds 95%, current disk usage {{ printf "%.2f" $value }}% |
Workload alerts
| Alert rule | Alert template |
|---|---|
| Deployment Pod availability less than 50% | Namespace: {{$labels.namespace}} / Deployment: {{$labels.deployment}} Pod availability less than 50%, current unavailable Pod count {{ $value }} |
| Job execution failed | Namespace: {{$labels.namespace}}/Job: {{$labels.job_name}} execution failed |
| Pod startup timeout failure | Namespace: {{$labels.namespace}}/Pod: {{$labels.pod_name}} has not started successfully for more than 15 minutes, waiting reason {{$labels.reason}} |
| Pod status abnormal | Namespace: {{$labels.namespace}}/Pod: {{$labels.pod_name}} has been in {{$labels.phase}} status for more than 10 minutes |
| Pod frequent restart | Namespace: {{$labels.namespace}}/Pod: {{$labels.pod_name}} restarted more than {{ $labels.metrics_params_value}} times within {{$labels.metrics_params_time}} minutes, current restart count {{ $value }} |
| Container CPU usage exceeds 85% | Namespace: {{$labels.namespace}} / Pod: {{$labels.pod_name}} / Container: {{$labels.container}} CPU usage greater than 85%, current value {{ printf "%.2f" $value }}% |
| Container CPU usage exceeds 75% | Namespace: {{$labels.namespace}} / Pod: {{$labels.pod_name}} / Container: {{$labels.container}} CPU usage greater than 75%, current value {{ printf "%.2f" $value }}% |
| Container memory usage exceeds 75% | Namespace: {{$labels.namespace}} / Pod: {{$labels.pod_name}} / Container: {{$labels.container}} memory usage greater than 75%, current value {{ printf "%.2f" $value }}% |
| Container memory usage exceeds 85% | Namespace: {{$labels.namespace}} / Pod: {{$labels.pod_name}} / Container: {{$labels.container}} memory usage greater than 85%, current value {{ printf "%.2f" $value }}% |