Container Compute Service:Monitor ACS cluster status with Prometheus

Configure alerting for Prometheus monitoring

Create alerts to receive real-time notifications through email, SMS, or DingTalk when specific conditions are met, helping you proactively identify anomalies. When an alerting rule is triggered, the system sends a notification to a designated contact group. You must create contacts before you can create a contact group. When you create a contact, specify their phone number and email address, or provide a DingTalk chatbot URL for automated notifications.

Step 1: Create a contact

1. Log on to the Prometheus console. In the upper-left corner, select the region where your Kubernetes cluster is located.

2. In the left navigation pane, choose Managed Service for Prometheus > Notification Contacts.

3. On the Contact tab, click Create a contact

4. In the Create a contact dialog box, set the following parameters and click OK.

   Parameter	Description
   Name	The name of the contact.
   Identifier	The user ID for an enterprise IM tool, such as DingTalk, Lark, or WeCom. Examples include a DingTalk ID or WeCom ID. This field is used to @mention contacts in alert cards within a group chat. If you want to use this feature, enter the correct user ID for the corresponding IM tool. The @mention feature for Lark and WeCom requires this field, but the feature for DingTalk does not.
   Telephone	After you set a phone number, the contact can receive alert notifications by phone call and SMS. Note You can use only verified phone numbers for phone call notifications within a notification policy. To verify a phone number, see Verify a mobile phone number.
   E-mail	After you set an email address, the contact can receive alert notifications by email.

   Important

   • You must provide at least a mobile phone number or an email address. Each mobile phone number or email address must be unique to a single contact.

   • The new version of Contacts does not support setting up DingTalk chatbots. To create a DingTalk chatbot, create it on the DingTalk/Lark/WeCom tab. See DingTalk Robot. Existing DingTalk chatbots from the previous version are still available.

Step 2: Configure alerting rules

1. In the left navigation pane, click Access Management.

2. On the Integration Environments tab, click the name of the desired cluster to open its Integration Management page.

3. At bottom, click Alert Rule. In the Alert Rule list, find the rule to modify and click Edit in the Actions column. After you configure the Alert Rule, click Complete. For more information, see Prometheus alert rules (new console) or Prometheus alert rules (legacy console).

Custom Prometheus monitoring metrics

Managed Service for Prometheus also provides advanced methods for collecting and configuring monitoring metrics. For more information, see the following topics:

• Default service discovery

• Collect custom metrics using ServiceMonitor

• Collect custom metrics using PodMonitor

• Other Prometheus service discovery configurations

Enable cluster network monitoring

By default, ACS integrates with Application Real-Time Monitoring Service (ARMS) to collect core control plane and basic metrics (pods and containers) for free. However, the service does not collect metrics related to CoreDNS and cluster Ingress by default. You must configure these settings in the Prometheus console. Data collection for these two components is a paid service.

1. In the ACS console, go to Cluster List, and choose the cluster you want to enable cluster network monitoring for.

2. In the left navigation pane, go to Add-ons.

3. On the Core Components tab, under Networking, install the managed editions of the ALB Ingress Controller and CoreDNS components.

Alternatively, install the Community Edition of the software in your cluster. For detailed configuration instructions, see Integration overview.