Alert coverage gaps highlight areas in your stack where you're missing alert coverage. In our Alerts and AI section we've created a homepage where you can see all your uncovered entities so you can assess what information your team is lacking and address the coverage gap.
In an increasingly dynamic landscape, we know how important it is to stay on top of your stack and be proactively notified if anything goes wrong. But, as your team grows or changes, it can be hard to know if you're getting the alerts you need, when you need them. Alert coverage gaps gives your team a comprehensive list of all entities that your team is not currenty covering.
What does covering an entity mean? A covered entity means that you've set up an alert to notify you of how a particular entity is performing. For example, you might have created an alert condition that opens an incident if a host's CPU goes above 80%. An uncovered entity is part of your system that is unmonitored which means there could be unhealthy behavior that goes unchecked. We created alert coverage gaps to highlight uncovered entities so your team doesn't miss valuable data and you can prevent incidents before they happen.
On the Alerts and AI UI page, select alert coverage gaps in the naviagtion. There you will see a dashboard of all your entitites that are not currently covered.
When you select alert coverage gaps you will see a list of all your unmonitored entities.
Once you've selected an entity you'd like to monitor, then you can create an alert condition to notify your team of any abnormal behavior.
New Relic will dynamically suggest the conditions we believe should be monitored based on the selected entity and your specific system. Some recommended alerts we might suggest are error percentage, apdex, or response time.
If you're not in the EU, once you select an alert condition you'd like to monitor, you may get recommended alert conditions based off of a machine learning model. The model may set different recommended thresholds based on a comparison of your entities with other users. New Relic monitors what conditions those other users have set up for entities that look and behave similarly and makes recommendations to your team based on this information. This means that the way you get your data is constantly improving and is always rooted in a realtime understanding of the current observability landscape.
If you're not given recommended alert conditions, once you select an alert condition you'd like to monitor, then your team will be alerted when this signal deviates from its normal baseline, upper and lower, for at least 5 minutes by 3.00 standard deviations.
Creating an alert for an unmonitored entity.
Every alert condition needs to be routed through a policy so you will be given the option to either select an existing policy or create a new one.
When you create an alert you will select which policy the alert will be grouped with.
Once you've selected or created a policy attached to the alert condition, then you might need to add a destination. Destinations are where we send notifications about your New Relic data. You can link your notificaitions to third-party tools like Slack, Atlassian Jira, email and more.