如何为Logic App Standard工作流配置失败告警？

Hey there, I've dealt with this exact scenario before—Logic App Standard doesn't have that handy RunsFailed out-of-the-box metric like Consumption plans do, but you don't need to write custom logs to set up failure alerts. The key is leveraging the built-in telemetry that's automatically sent to your Log Analytics Workspace if you've configured it correctly. Here's how to set this up properly:

Option 1: Alert on Failed Workflow Runs (Broad Scope) #

First, make sure your Logic App Standard instance is linked to a Log Analytics Workspace:

• Go to your Logic App > Monitoring > Diagnostic settings
• Create a new setting, select your Log Analytics Workspace, and check the WorkflowRuntimeLogs category (this captures all workflow run and activity events)

Once that's set up, create the alert rule:

1. Navigate to Azure Monitor > Alerts > Create > Alert rule
2. Under Resource, select your Log Analytics Workspace
3. For Condition, choose "Custom log search" as the signal type, then paste this Kusto query to filter failed runs:

   WorkflowRuntimeLogs
   | where OperationName == "WorkflowRunCompleted"
   | where Status == "Failed"
   | project TimeGenerated, WorkflowName, RunId, Status
   

4. Set your alert threshold (e.g., "Number of results greater than 0") and evaluation frequency (e.g., check every 5 minutes, look back 5 minutes)
5. Configure your Action Group to send notifications (email, Teams, SMS) or trigger automated actions
6. Finish by naming your alert rule and adding a description for context

Option 2: Alert on Specific Failed Activities (Granular Control) #

If you need to target failures in individual activities rather than entire runs, adjust the Kusto query to focus on activity-level events:

WorkflowRuntimeLogs
| where OperationName == "ActivityCompleted"
| where Status == "Failed"
| project TimeGenerated, WorkflowName, ActivityName, RunId, ErrorMessage

This lets you get alerts only when a specific step (like an HTTP call or SQL query) fails, which is great for targeted troubleshooting.

Pro Tips for Optimization #

• Add dynamic details to notifications: Include fields like WorkflowName and RunId in your alert message so your team can jump straight to the failed run in the Logic App portal
• Set alert suppression: Prevent spammy alerts by configuring a suppression period (e.g., don't send another alert for the same workflow failure within 15 minutes)
• Automate remediation: Use Action Groups to trigger a Logic App or Azure Function that attempts to retry the failed run, or creates a ticket in your ITSM tool

内容的提问来源于stack exchange，提问作者AasthaK