Monitoring events on your cloud server¶
Automated monitoring of your cloud server configuration and its entire infrastructure stack enables you to know how your system is being used and respond to changes quickly and flexibly.
Monitoring operates on two levels:
- Remote monitoring, managed by Rackspace and based in our global data centers
- Agent-based monitoring, available for you to configure on your cloud
Rackspace Cloud Monitoring is included at no charge with every cloud account.
You can customize your monitoring configuration in several ways:
- Choose which servers to monitor. For example, you might not want to be notified when a test-only system fails.
- Choose which regions to monitor. For example, you might be particularly concerned about a new project based in the HKG region.
- Describe what kind of activity or inactivity is cause for alarm. For example, you might want to know when a monitored resource takes longer than 50 milliseconds to respond to a ping.
- Identify whom to notify and how to notify them. For example, you might want to send email to your DevOps team list when a resource reaches a warning state, and send a pager message to your Customer Support team when the resource reaches a failed state.
Because Rackspace Cloud Monitoring is closely integrated with Rackspace Auto Scale, your monitoring configuration can become the basis of your autoscaling configuration. Used in combination, Cloud Monitoring can report an event (for example, a workload spike on a server) and Auto Scale can respond to that report (for example, by adding a server to share the workload within a Cloud Load Balancer group).
You can interact with Cloud Monitoring and Auto Scale by using the same methods that you use to interact with Cloud Servers:
- The Cloud Control Panel, as explained in its technical documentation
- The raxmon command-line interface (CLI)
- Software development kits
- Application programming interfaces