A company has been running their application on Compute Engine.

To monitor and get notified in case the script causing the Compute Engine instance to crash is executed again, you should create an Alerting Policy in Stackdriver (now known as Google Cloud Monitoring). The Process Health condition can be set to monitor the number of executions of the script and ensure it remains below the desired threshold. By enabling notifications, you will be alerted if this threshold is exceeded.

Step-by-Step:

Log Script Executions: Ensure that the script execution is logged.

Create a User-Defined Metric: Go to Google Cloud Console > Logging > Logs-based Metrics, and create a new user-defined metric that counts the number of times the script executes.

Set Up Alerting Policy:

Navigate to Google Cloud Console > Monitoring > Alerting.

Click on “Create Policy”.

Add a condition and select “Logs-based Metric”.

Configure the condition to trigger when the number of script executions exceeds the threshold.

Configure Notifications: Add notification channels (email, SMS, etc.) to the alerting policy.

Save and Test: Save the policy and test to ensure notifications are received when the script is executed beyond the threshold.

Google Cloud Logging User-defined Metrics

Google Cloud Monitoring Alerting Policies