Realtime Compute for Apache Flink:Configure alert rules

System event alerts

Subscribe to system events for single or batches of jobs.

1. Log on to the Cloud Monitor console.

2. In the left navigation pane, choose Event Center > Event Subscription.

3. On the Subscription Policy tab, click Create Subscription Policy.

4. On the Create Subscription Policy page, configure the subscription details. For configuring parameters not listed below, see Manage event subscriptions (Recommended).

   

   • Subscription Type: Select System Events.

   • Products: Select Flink.

   • Event name: Supported events include JOB_FAILED, ECS.SystemFailure, and ECS.SystemMaintenance.  

   • Event Content: Fill in the following information as needed.

     • Workspace ID: Enter a workspace ID, and the alert rule applies to all jobs across all namespaces within the target workspace. For viewing the workspace ID, see How do I view information such as the workspace ID?.

     • Namespace name: Enter a namespace name, and the alert rule applies to all jobs in the target namespace.

     • Deployment name: Enter a job deployment name, and the alert rule applies to specific job deployments. Separate multiple names using commas (,). When duplicate deployment names exist, use the deployment ID.

     • Deployment ID: Enter a job deployment ID, and the alert rule applies to specific job deployments. Separate multiple IDs using commas (,). To get a deployment ID, go to your deployment details page

   Note

   Leaving Application group, Event Content, or Event Resources empty will apply the subscription to all workspaces within your account.

Workflow event alerts

Subscribe to workflow events. For workflow details, see Manage workflows.

1. Obtain the resource ID of the workflow task.

   1. Log on to the Cloud Monitor console.

   2. In the left navigation pane, choose Event Center > System Event.

   3. On the Event Monitoring tab, set SelectProduct to Flink, set SelectEvent Name to flink:Workflow:TaskStateChange, and click Search.

   4. Filter for the workflow task's resource ID.

      

      The resource ID's format: acs:flink:cn-hangzhou:<AlibabaCloudAccountID>:resourceId/workspaceId/<workspaceId-namespaceId>#workflowDefinitionName/<workflowDefinitionName>#taskDefinitionName/<taskDefinitionName>. You can also use this format to directly construct the resource ID for your workflow task.

      Parameter	Description
      <AlibabaCloudAccountID>	The ID of the Alibaba Cloud account that owns the Flink workspace.
      <workspaceId-namespaceId>	Consists of the workspaceId and namespaceId, joined by a hyphen (-). workspaceId: Your workspace ID. For more information, see How do I view the information about a workspace, such as the workspace ID?. namespaceId: Your namespace name.
      <workflowDefinitionName>	The workflow name.
      <taskDefinitionName>	The name of the workflow task.

      Note

      State changes for workflow tasks typically appear in Cloud Monitor with a delay of a few minutes.

2. Subscribe to event notifications.

   1. In the left navigation pane of the Cloud Monitor console, choose Event Center > Event Subscription.

   2. On the Subscription Policy tab, click Create Subscription Policy.

   3. On the Create Subscription Policy page, configure the policy. For details, see Manage event subscriptions (Recommended).

      • Subscription Type: Select System Events.

      • Products: Select Flink.

      • Event name: Select flink:Workflow:TaskStateChange.

      • Event Content: Filter by the desired transition state, e.g., toState: FAILED, toState: SUCCESS, and fromState: SCHEDULED, toState: RUNNING.

      • Event Resources: Enter the resource ID obtained in the previous step. Separate multiple IDs with commas (,).

      • Event Type, Event Level, and Application group: Leave these parameters unset.

Single job configuration (Development Console)

You can create alert rules for target jobs from ground up or templates.

1. Go to the alert configuration page.

   1. Log on to Realtime Compute for Apache Flink's Management Console and click Console in the Actions column of your workspace.

   2. In the left navigation menu, choose O&M > Deployments. Click your job deployment.

   3. Select the Alarm tab.

2. Switch to the Alarm Rules subtab, and choose Add Rule > Custom Rule.

   Alternatively, choose Add Rule > Create Rule by Template.

3. Enter the alert rule information.

   Section	Label	Description
   Rule	Name	Enter a rule name. It must start with a letter and can contain lowercase letters, digits, and underscores (_). Length: 3 to 64 characters.
   	Description	Enter your remarks about the rule.
   	Content	Configure the conditions that trigger the alert. Realtime Compute for Apache Flink periodically compares actual metric values against the threshold and triggers an alert if a condition is met. Configure the alert logic: Metric: Restart Count in 1 Minute Checkpoint Count in 5 Minutes Emit Delay: Business latency from data generation to source egress, in seconds. Important The accuracy of Emit Delay depends on accurate timestamps from external systems. If external systems lack timestamps or record incorrect ones, Emit Delay will be inaccurate, failing to reflect true latency. To reflect actual latency more accurately, configure alert rules for composite metrics. For details, see Best practices for monitoring and alerting. IN RPS: Input records per second. OUT RPS: Output records per second. Source Idle Time: Duration the source operator was not processing data, in milliseconds. Job Failed Time Interval: The length of the historical data time window the system queries backwards from the current time during each check, in minutes. Comparator: Supports >= and <=. Thresholds: The value compared against the metric. Comparator (>=): Compares the maximum metric value against the threshold. Alert if max_value >= threshold. Comparator (<=): Compares the minimum metric value against the threshold. Alert if min_value <= threshold. Assume you are monitoring the Checkpoint Count in 5 Minutes metric, with a time interval of 10 minutes and a threshold of 2, using the <= comparator. Flink checks every minute, evaluating the metric over a 10-minute window. An alert is triggered if the minimum successful checkpoint count within any 5-minute period drops to 2 or fewer.
   	Effective Time	The time frame when the alert rule is enabled. Defaults to all day. You can restrict alerts to specific times, like during business hours (9 AM to 6 PM).
   	Alarm Rate	The interval at which alerts are sent within a specified number of minutes. Supports 1 minute to 1440 minutes (24 hours).
   Notification	Notification	Select one or more notification methods. Options include: DingTalk Email SMS Webhook Phone Ensure the recipient's phone number is verified; otherwise, the notification will not be effective. You can click Notification object management below. If the Unverified tag appears in the Phone column for the target contact on the Contacts tab, click it to complete verification. Important Ensure your chosen notification method has a corresponding notification object configured. For DingTalk, this means adding a DingTalk robot notification object after selecting DingTalk as the method.
   	Notification object	Choose one or more notification objects. Click Notification object management to create notification objects. For details, see Workspace and namespace management.
   	Alarm Noise Reduction	Click Advanced Settings and switch on Alarm Noise Reduction. When this option is enabled, alerts are sent only when the threshold is met continuously. This prevents alerts for temporary issues, such as brief failovers caused by cluster scheduling or auto-tuning.
   	No Data Alarms	Click Advanced Settings, turn on the No Data Alarms switch, and specify the duration. Triggers an alert if no monitoring data is reported within the selected time period. Typical causes: JobManager exceptions, job stops, or reporting link failures.

4. Click OK.

   After you save an alert rule, it is automatically enabled and appear in the alert rule list. You can choose to stop, edit, or delete them.