Smart Alerts for service level objectives

For information about permissions and access requirements for Smart Alerts, see Accessing service levels.

Once created, the Smart Alerts list and each of the individual configurations display the teams that are associated with the underlying SLOs or Apdex configurations on which the Smart Alert is set up.

To create a Smart Alert for SLOs, complete the following steps:

1. From the Instana UI navigation menu, select Service levels.
2. Select the Smart Alerts tab.
3. Filter by Entityfor Service level objective. This should be the default view.
4. Click Create Smart Alert
   Note: Alternatively, you can create a Smart Alert from a specific SLOs dashboard by navigating to that dashboard.
5. The Smart Alert configuration dialog is displayed.
6. Define scope: Define the scope by selecting the SLOs to which this smart alert applies.
   • To make a selection, click Add SLOto open the SLO selection modal.
   • In the Add SLOsdialog, you can select on or multiple SLOs across different entity types.
   • The side panel on the right shows your selections grouped under each entity type (Applications, Infrastructure, Synthetic tests, Websites) with the count of selected items for each category.
   • Click Add to confirm your selections.
   • You can add more SLOs by clicking Add SLO + or remove existing ones by clicking the remove icon in the Action column.
7. Select the type of blueprint to be alerted on:

   • Burn rate: An alert is triggered based on the SLO burn rate and the specified threshold. Currently, two types of burn rate alert configurations are supported

     • Single alert window: Uses a single threshold and a single alerting window to monitor the Error Budget Burn Rate

     • Long and short alert window: Uses two distinct alerting windows (long and short), each with its respective threshold to examine data over different time periods. The long alerting window looks back over an extended period to evaluate whether the burn rate has exceeded the threshold. It helps identify issues that develop gradually and might pose future problems if not addressed. In contrast, the short alerting window analyzes a shorter time frame to quickly detect if the burn rate rises sharply. This is useful for identifying sudden spikes in errors or downtime that might indicate immediate and critical issues. Using both windows to determine the burn rate allows for a more balanced and accurate alerting mechanism helping ensure that alerts are triggered only when both short-term volatility and long-term patterns indicate a sustained problem.

     Note: Users can configure the durations and thresholds for both the long and short alert windows based on their monitoring needs. However, as a standard practice, the short alert window should typically be 1/12th the length of the long alert window to maintain meaningful contrast between short-term anomalies and long-term trends.

   • Error budget: An alert is sent when the error budget usage reaches the specified percentage.

   • Status: An alert is sent when the SLO status is less than the configured threshold.

8. Select whether to configure the error budget, SLO status, or burn rate:
   • Burn rate: Define alert window(s), threshold(s), time window, and grace period.
   • Error budget and SLO status: Define the threshold percentage, time window, and grace period.
9. Optional: Select alert channels: You can use an alert channel to send notifications about the alert. This is optional and can be added later.
   • The list of available alert channels is displayed.
   • Select one or more alert channels.
   • Click Add alert channels to add the selected alert channels.
   • Alert channels can also be created by clicking Create alert channel.
10. Configure the Alert properties:
    • Title: The title for identifying the Smart Alert.
    • Description: Enter a description to provide context for the alert.
    • To add dynamic placeholders to the title and description fields, click Insert placeholder. You can include supported placeholders such as the SLO name and alert severity. When an alert is triggered, the system resolves these placeholders to actual values and provides contextual information.
    • Alert level: Choose the style of the alert. You can set the alert to Critical or Warning.
    • Triggers incident: Choose whether the Smart Alert triggers an incident or not.
    Note: Incident triggering is only supported for application entities.
11. Optional: Add a custom payload to the Smart Alert by clicking Add Row and specify the key-value pairs for each added row.
12. Click Create.