Key Alerts Activation and Notifications

  • Release version: Washingtondc
  • Updated February 1, 2024
  • 2 minutes to read
    • Summarize
    Summarized using AI
    This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete.

    Summary of Key Alerts Activation and Notifications

    The Key Alerts Activation functionality within the Instance Observer allows ServiceNow customers to configure alerts based on historical thresholds tailored to their business needs. This feature provides proactive notifications for critical metrics, helping teams respond to potential issues before they escalate.

    Show full answer Show less

    Key Features

    • Critical Metrics: Alerts are based on moving averages for six key metrics:
      • Transaction (increase) Instance: Alerts for increases in transaction moving averages.
      • Transaction (drop) Instance: Alerts for decreases in transaction moving averages.
      • Response Time - Instance: Alerts for response times exceeding historical medians.
      • Database Response Time: Alerts for database response times exceeding historical medians.
      • Default Semaphore Mean: Alerts when semaphore means approach concurrency limits.
      • Default Semaphore Queue Depth: Alerts when queue depths approach maximum limits.
    • P-Score Assessment: The P-Score evaluates instance usage, determining the eligibility for alerts based on transaction stability.
    • Automated Threshold Refresh: Thresholds are refreshed weekly for eligible instances, ensuring alerts remain relevant.

    Key Outcomes

    By implementing Key Alerts Activation, customers can expect to receive timely notifications that help them understand and address unusual data patterns. This proactive approach minimizes the risk of future issues, enhances operational efficiency, and allows teams to make informed decisions based on real-time metrics.

    Use the Instance Observer Key Alerts Activation functionality for flexibility to set alerts with historical thresholds from four week increments specific to your business needs and configure who on your team receives notifications.

    Overview of critical metrics

    Keys Alerts Activation will provide you with proactive alerts about moving averages, a statistic that captures the average change in a data series over time, for six critical metrics:

    1. Transaction (increase) Instance: Indicates transaction moving average increases by a certain percentage of it's historical median for a certain time period.
    2. Transaction (drop) Instance: Indicates transaction moving average drops by a certain percentage of it's historical median for a certain time period.
    3. Response Time - Instance: Indicates the moving average of response time is greater than the historical median in a certain time period.
    4. Database Response Time: Indicates the moving average of database response time is greater than the historical median in a certain time period.
    5. Default Semaphore Mean: Indicates the default semaphore mean is approaching limits of max concurrency by a certain factor in a certain time period.
    6. Default Semaphore Queue Depth: Indicates the default semaphore queue depth is approaching limits of maximum queue depth by a certain factor in a certain time period.
    Note:
    It is important to note that each notification may not represent an issue, moreover signals about a data point that is unusual or does not conform to the expected patterns for that specific metric. Users can be proactive, where necessary, and act on a notification accordingly that may avoid a future issue.

    To set up alerts, see Configure Key Alerts.

    Instance Observer utilizes the P-Score to assess instance usage, then calculate the thresholds exclusively for those instances. The P-score is a rating determined by the number of transactions on an instance, providing a higher score when the instance's transactions remain consistently stable.

    • When an instance meets the required P-Score for eligibility, users will see a banner to opt-in for the alerts activation feature. Users will receive intelligent thresholds under the IO recommended condition for all six metrics derived from historical data.
    • If an instance has a low P-Score, the banner to opt-in for alerts activation feature will not be visible. The P-Score is regularly refreshed, and as instances grow in the future, you will see the option to opt-in for the alerts activation feature. Thresholds are refreshed automatically weekly for instances meeting the minimum P-Score criteria.