Amazon CloudWatch Monitoring: Feature Spotlight
This is the first of a new series that will highlight some key features and integrations we’ve launched. It’s a look at things we’ve been working hard to improve, as well as spotlighting some other features that customers have told us we’ve been a bit too quiet about up until now. First up, we’re looking at monitoring your Amazon Services with Hosted Graphite.
We’ve had a lot of great feedback on our AWS CloudWatch add on in the years since its release. It syncs your AWS metrics straight to your Hosted Graphite account so you can view your CloudWatch data on curated, interactive dashboards. As a result, the health of your CloudWatch services is very easy to check.
Where CloudWatch stores your five minute data points for 63 days by default, our maximum retention is 180. They store your one hour data points for 455 days, while we hold onto them for two years. We also give you access to a full alert history so you can track historic incidents and review anomalies.
Alerts and annotations
All teams have different sets of critical paths. Our alerts tell you when your metrics do something unexpected, if you need to act and how you might fix the problem. Flexible rules and notifications give you more control over what you hear and how you hear it. Choose to be alerted via email, PagerDuty, VictorOps, HipChat, Slack, OpsGenie or alternatively build your own integration with our webhooks. Whatever way you’re notified, for deeper context you can overlay automatic alert annotations on your historic graphs.
It’s easy to share dashboards with your team. You can hide or share specific dashboards with particular users or verify access using Active Directory.
All of the following CloudWatch services are covered:
Getting set up
It’s quick to get started. If you’re already a Hosted Graphite customer, you can start monitoring your Amazon Services immediately. Just give us the access key for a read-only IAM user and we’ll take care of the rest. What you share is up to you: you can send us everything or choose the metrics that matter most. AWS lets you tag most of their resource types, so if you select the Service Tagging option and send us a tag name (and value), we’ll only get metrics for resources with those specific tags.