Handling host-related issues

A cloud server is a virtual server; it resides on a physical host. The physical host is designed with redundancies for power, network, and disk to prevent failures to the underlying virtual servers. It might sometimes be necessary to restart a physical host for maintenance or other issues. When it is practical to do so, virtual servers are not disrupted when their physical host is restarted

In many cases, a virtual server can be live-migrated away from its current physical host to a new one. Live migration to a new physical host is transparent to the running virtual server and transparent to users of the running virtual server. If a physical host is in a state where the virtual servers are down and live migration cannot be used, Rackspace sends regular communication until the virtual server in question is up and running.

See also

Using flavors to create a new cloud server Changing the characteristics of a cloud server Monitoring events on your cloud server Creating a cloud server in a region Handling host-related issues Operating a cloud server with nova-agent Configuring disks for a cloud server Managing access with SSH keys Synchronizing time with NTP Describing a cloud server with metadata: Setting and updating metadata Showing metadata

Understanding flavor classes: Flavor classes for virtual cloud servers Flavor classes for OnMetal cloud servers Choosing a flavor class Checking for regional availability of flavor classes Controlling cloud server initiation: Controlling boot behavior with cloud-init Controlling boot behavior with a config drive Controlling boot behavior with personality

Understanding Cloud Servers introduces key ideas. To learn how to put these ideas to work, start at Actions for Cloud Servers.