Create Custom New Relic alerts
- Topics:
- Tools and External Services
- Security
CREATED FOR:
- Beginner
- Developer
Learn how to create custom alerts in New Relic for monitoring website traffic and setting thresholds for alert notifications.
Who is this video for?
- DevOps Engineers
- Site Reliability Engineers
- IT Operations Managers and security monitoring teams
Video Content
- Create custom New Relic alerts by accessing the alert interface through chart options or the query editor.
- Understanding how thresholds for alerts can be set based on data aggregation.
- Configure alerts with severity levels and assigned to existing policies to control notification workflows and destinations.
In this video, we’re going to take a look at how to create your own custom New Relic alerts. There’s a few ways to get to the alert interface. For many of the charts in New Relic, if you’re looking to create an alert from the data in the chart or similar data, you can click on the three dots and go to create alert condition. You can also come over on the left side here to query your data to open the query editor, and from there go to create alert. If you are in the alerts interface in conditions, you can click on new alert condition, and then for now we’re going to write our own query. I have this query ready to go. This is a query that’s going to be looking for the IPs that have hit our site and the counts excluding certain conditions. So as we run this, we’ll see the data show up, and we can go to next. This is where we’re going to set thresholds.
So for data aggregation, we can go to the window duration and say maybe every five minutes.
For streaming method, we’ll leave it at event flow for steady or frequently reporting data.
For the condition thresholds here, we’re going to leave it at static because we’re looking for an upper limit for when the data crosses that upper limit. For anomaly, that one you would use if you’re looking for thresholds that are crossed either above or below a standard threshold. We can leave the severity level either at critical or go to warning, and in this case, we don’t have a lot of IP hits, but for our purposes, we’re going to look for it’s been hit more than 10 in that five-minute duration. And so you’ll see our critical threshold update here, and you can see here’s our one pink bar where this is a time where the alert would have been sent. We’ll click next and name this. For policy, we’re going to choose an existing policy that we have. That is going to control where the alert is sent along with the workflows that have been set up in the destinations. We’ll save this condition, and now this alert is ready to go and ready to alert us when necessary.