Our next-generation Cloud Servers powered by OpenStack® are here, offering an open, flexible cloud platform. With a responsive API and an improved Cloud Control Panel it's easier than ever to manage your server instances.
If you're considering moving your application from our first-generation cloud platform to our open cloud, weigh the pros and cons of the new platform before committing to a cloud migration. While we're excited about the possibilities an open cloud offers, the differences between the environments might make our first-generation cloud a better choice for your application for the time being.
Our next-generation cloud platform offers some benefits over the original.
The OpenStack API used by our open cloud is responsive and flexible, with room to grow. New features include management of server metadata and retrieval of bandwidth usage statistics.
Next-generation Cloud Servers have assigned IPv6 addresses, though at this time we only support one assigned IPv6 address per server.
You can choose the region that will house a created Cloud Server in our next-generation environment.
Regional control helps you manage backend availability between your servers and connectivity to services such as Cloud Databases and Cloud Load Balancers. You can also ensure that you have copies of your application in multiple regions to aid with disaster recovery.
Disk management is more flexible, letting you repartition your Cloud Server's virtual disk and switch file systems if desired.
The next-generation cloud's OpenStack underpinnings provide the potential for interoperability with OpenStack-based tools like python-novaclient, as well as compatibility with private and third-party clouds running OpenStack.
Before committing to a switch be sure your application isn't dependent on a feature or option that isn't available in our open cloud.
Not all features of first-generation Cloud Servers are ready for our new cloud environment yet. At this time that includes the ability to schedule image snapshots and creation of shared IP groups via the API.
For more information please see our article detailing feature comparisons between first-generation Cloud Servers and next-generation Cloud Servers.
Instances with 256 MB of memory are not available for next-generation Cloud Servers - server sizes start at 512 MB.
At this time all migration options involve moving data to a new server, and the new server will have a different IP address from the original instance. After a migration you will likely need to make DNS changes and application configuration changes to account for the new IP address.
Switching an existing application from the first-generation cloud to our open cloud requires the migration of your data to a new next-generation instance. In addition to the change in IP address, finalizing the move to the new server may require some application downtime.
To help you migrate existing data to our open cloud, the Cloud Control Panel offers the ability to take a special snapshot of a first-generation Cloud Server that is automatically converted to a next-generation snapshot. The new snapshot can be used to make a next-generation Cloud Server.
Using a snapshot to migrate to the open cloud can be a simpler and quicker process than migrating your data with rsync, but it will not work with all instances. Support for snapshot migration is determined by the base image used to initially create the instance.
Manual migration is available as an alternative for unsupported instances and is described further along in this article.
To create a next-generation snapshot from a first-generation server go to the Cloud Control Panel, click the action cog for the server you want to duplicate, and select the Create Image command. If the server is compatible with the conversion process you'll be offered a choice between making a first-generation or next-generation image.
For a converted image choose the Next-Generation Cloud Server Image option. Once the snapshot creation is complete you'll see it as an image choice when creating a new Cloud Server. Note that standard snapshot storage pricing on Cloud Files will apply.
If you don't see an option to create a next-generation image your instance may not be supported by the conversion process at this time. In that case you will want to use a manual migration approach as detailed in the next section.
If you have trouble creating the image please contact support and they will assist you.
Bear in mind that the IP address of an instance created from a snapshot will be different from the server used to create that snapshot. Depending on your application there may also be issues if both the original server and the new server are active simultaneously.
For instances that aren't supported by our snapshot-based migration process, or for users who prefer a more hands-on approach, we offer some articles describing how to manually migrate your data to a next-generation server.
Remember to take the size of the original server into account, as well as other considerations like the region of connected Cloud Database instances and Cloud Load Balancers when creating a new server for a manual migration.
Our article series on migrating a Linux server provides instructions for migrating your data via rsync, either with a manual process or with the assistance of a python script.
Our article on migrating a Windows server details how to use the Web Farm Framework to transfer IIS and MSSQL data between Windows Server 2008 servers.
After migrating to a new Cloud Server remember to make any necessary changes to DNS information or load balancers for your application.
Also keep in mind that existing first-generation snapshots can't be used to create next-generation servers. Don't forget to make backup arrangements for the new server.
© 2011-2013 Rackspace US, Inc.
Except where otherwise noted, content on this site is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License