Container Service for Kubernetes:How to configure ACK One ArgoCD alerts

Prerequisites

Before you begin, make sure that you have:

• Fleet monitoring enabled. See Enable Fleet monitoring.

• A notification object configured. See Notification objects.

If your notification object is DingTalk, add Custom Keywords alerts in the security settings of the DingTalk chatbot before proceeding.

Create an Argo CD alert rule

1. Log on to the ACK One console. In the left-side navigation pane, choose Fleet > Fleet Observability > Fleet Monitoring.

2. In the upper-right corner of the Fleet Monitoring page, click Alert Settings to open the Prometheus Alert Rules page.

3. Click Create Prometheus Alert Rule and fill in the fields described in the following table.

   Parameter	Description	Default	Example
   Alert rule name	A name for the alert rule.	—	ACK One Argo CD pod memory alert
   Check type	Detection method. Static Threshold compares a metric against a fixed value. Custom PromQL lets you write a PromQL expression directly.	—	Static Threshold
   Prometheus instance	The ACK One Fleet instance to monitor.	—	text-XXXX
   Alert contact group	The Kubernetes application group to be monitored within your environment.	—	Kubernetes workload
   Alert metric	The metric to evaluate. For Argo CD pods, Container Memory Usage and Container CPU Utilization are the most important metrics to watch.	—	Container Memory Usage
   Alert condition	The threshold condition that triggers an alert event.	—	CPU utilization greater than 80%
   Filter conditions	Narrows the scope of the alert rule. See the filter condition types below.	Traverse	Namespace Equal argocd, Pod: Traverse
   Duration	Controls when an alert event fires. See the duration options below.	—	Alert condition continuously met for 2 minutes
   Alert level	Severity level. Default is the lowest; P1 is the highest. Valid values: Default, P4, P3, P2, P1.	Default	P1
   Alert message	The message sent to recipients when an alert fires. You can specify custom variables in the alert message based on the Go template syntax.	—	Namespace: {{$labels.namespace}} / Pod: {{$labels.pod_name}} / Container: {{$labels.container}} CPU utilization: {{$labels.metrics_params_opt_label_value}} {{$labels.metrics_params_value}}%. Current value: {{ printf "%.2f" $value }}%
   Alert notification	Notification format. Valid values: Simple Mode and Standard Mode.	—	Simple Mode
   Notification objects	The channels that receive alert messages, such as a DingTalk group.	—	DingTalk alert
   Notification period	The time window during which alert notifications are sent.	—	23:00–01:00
   Whether to resend notifications	How often to resend the alert if it is not cleared.	—	Every 10 minutes

   Filter condition types

   Type	Scope	Additional input required	Supports multiple values
   Traverse (default)	All resources in the Prometheus instance	No	N/A
   Equal	The specified resource only	Resource name	No
   Not equal	All resources except the specified one	Resource name	No
   Regex match	Resources whose names match the expression	Regular expression	Yes (via regex)
   Regex not match	Resources whose names do not match the expression	Regular expression	Yes (via regex)

   Duration options

   Option	When an alert fires
   If the alert condition is met	As soon as a single data point crosses the threshold
   If the alert condition is continuously met for N minutes	Only after the threshold has been exceeded for at least N minutes continuously

4. Click Completed to save the alert rule.

Verify the alert rule

After saving the rule, simulate the alert condition to confirm that notifications reach the intended recipients:

1. Temporarily lower the alert threshold to a value that your Argo CD pods currently exceed, or generate a load spike on the pods.

2. Wait for the configured Duration to elapse.

3. Check the notification object (for example, the DingTalk group) for an alert message.

4. Confirm that the message content matches the Alert message template you configured.

To review historical alert events, open the Prometheus console. See View historical alerts.

What's next

• Create and manage an alert rule template — reuse alert configurations across multiple Fleet instances.

• Enable Fleet monitoring — set up the monitoring foundation if you have not done so yet.