Cloud Databases FAQ

Cloud Databases is a standalone, API-based, relational database service built on OpenStack® cloud that enables Rackspace customers to easily provision and manage multiple MySQL database instances. Instances are provisioned in a single-tenant, container-based environment per account and are accessible through the Rackspace internal ServiceNet network. In addition, High Availability (HA) instances can also be provisioned with a public IP during creation to enable access via public IP. Each database instance is optimized for performance. You can run a database instance by using MySQL, Percona, or MariaDB as the database technology.

Cloud Databases provides a complete solution for customers who demand a high-performance, purpose-built infrastructure designed for relational databases that is backed and supported by engineers who specialize in MySQL workloads. Cloud Databases is a fully-managed service for customers who want to focus on developing their applications without having to manage the underlying infrastructure. The service offers high availability, scheduled backups, on-demand backups and restores, integrated monitoring, redundant storage, replication, scalability, and full control of your database.

Yes, High Availability (HA) instance groups enable both internal connections on the data center's internal service network (Service Net) and external networks through a public IP or hostname. Single instances and replica sets are provisioned only with network interfaces on Service Net. Connecting to one of these types of Cloud Databases instances remotely requires either a Cloud Server or Cloud Load Balancer to proxy the connection.

For more information about connecting to a Cloud Databases instance, see Connect to a Cloud Databases instance.

For more information on HA instance groups, see High Availability for Cloud Databases or the "High Availability" section of this FAQ.

Each Cloud Databases instance comes with an attached storage volume. Storage volumes are automatically provisioned on a shared Internet small computer system interface (iSCSI) storage area network (SAN) that enables increased performance, scalability, availability, and manageability. Applications with high input/output (I/O) demands are performance-optimized and data is protected through both local and network RAID-10. (RAID stands for redundant array of independent disks.) Additionally, network RAID provides synchronous replication of volumes with automatic failover of SAN nodes and load balancing across available storage clusters.

Every Cloud Databases instance is optimized for performance. Cloud Databases uses container-based virtualization, which eliminates the performance bottlenecks that are associated with traditional hardware virtualization and enables your database to run at near bare metal speeds. Cloud Databases also uses dedicated SAN storage and high-speed networking to give you faster access to your data.

For the most up-to-date information about available instance sizes, see the Cloud Databases website or the Cloud Databases Developer Guide.

Any US or UK customer with a Cloud account may provision multiple ServiceNet database instances and also manage multiple databases and users (within resource limits). This service is also available to RackConnected Cloud Servers that use RackConnect. Both First and Open Cloud Servers can connect to Cloud Databases, as well as any product with access to our internal ServiceNet network within the same regional data center.

No, Cloud Databases are available only to customers with Cloud account credentials. Managed Operations Service Level or dedicated customers with RackConnect (those customers who also have a Cloud account) have access but can use the service only with their Rackspace Cloud product resources.

Yes. Use the following steps to access Cloud Databases:

1. Log in to the Cloud Control Panel
2. In the top navigation bar, click Select a Product > Rackspace Cloud.
3. Select Databases.

Connecting to a Cloud Database instance remotely requires a high availability instance group with a public IP, or a single or replica instance connected either to a Cloud Server or a Cloud Load Balancer to proxy the connection.

Yes. An SSL certificate is installed for each Cloud Databases instance. This certificate enables a secure connection between your application and the instance. When an SSL connection is established, any data transfer between the instance and the application is encrypted.

The behavior of your instance during a backup depends on the storage that you use for tables. If you use only InnoDB, write access to your database instance is not suspended. However, if you have MyISAM tables, those databases are write-locked during the backup process.

MySQL supports several types of table engines, which are also known as *table types*. The tables on a Cloud Databases instance can use either a mix of different table engine types or use the same type. We currently support backups of databases that use InnoDB and MyISAM.

Backups are stored in your Cloud Files account, and you are charged for the storage that they use. Standard rates for Cloud Files storage apply. For current costs, see the OpenStack pricing page.

Backups are created by using Percona XtraBackup to create a hot copy of all of the databases on an instance. The resulting database files are streamed directly to your Cloud Files account for storage.

Scheduled backup, on-demand backup, and restore operations are currently supported by the Cloud Control Panel and the Cloud Databases API. For more information, see Scheduled backups for Cloud Databases and Manage backups for Cloud Databases. For details about using the Cloud Databases API, see the API Reference.

You can also use backup and restore features through the Trove command-line interface (CLI) tool. For information about using this tool, see Using the Trove client.

The duration of a backup depends on the size of your databases and any network saturation during the backup.

To restore a database backup, you must create a new database instance and specify the backup that you want to restore during the create request. Your backup is loaded to the new instance, and you receive a DNS endpoint for the new instance. After the restore operation is complete, update your application to use the new endpoint.

The original instance is not altered during a restore operation and might remain in use or be deleted through the API, the CLI, or the Cloud Control Panel.

There is no limit on the number of database backups that you can create.

Note: You may run only one backup at a time. Duplicate requests generate a 422 error.

Cloud Databases runs off of SAN storage by using a mount point. After an instance is deleted, the mount point is destroyed.

Backups are not deleted when a database instance is deleted. You must manually remove any stored backups.

Yes. Scheduled backups enable you to set a time of day to take a daily backup and a day of the week to take a full backup. You can also specify the maximum number of full backups to retain.

Support coverage information is available on the OpenStack Cloud Service Levels page.

The Cloud Service Level Agreement (SLA) is available on the Rackspace website.

Monitoring is available for all Cloud Databases instances through pre-configured Cloud Monitoring checks, including load average, CPU, memory, disk storage, network, and a number of MySQL metrics. You can monitor your Cloud Databases instances by using the Cloud Control Panel, the Cloud Monitoring API, or the Cloud Monitoring CLI.

You can also set up alarms that send email alerts to you that are based on thresholds that you define. An alert for disk space is set up by default for every instance.

In addition, you can use Rackspace Intelligence to observe usage patterns for any unexpected changes in your environment. For more information, see Getting started with Rackspace Intelligence for the cloud.

Yes. A Cloud Databases support ticket category is available in the Cloud Control Panel.

InnoDB is the default storage engine for Cloud Databases. InnoDB enforces Atomicity, Consistency, Isolation, Durability (ACID) transactions that enable commit, rollback, and crash recovery capabilities to protect user data.

During a backup, a hot copy process is used on all tables. InnoDB tables record all transactions during the copy in order to replay them during a restore operation.

In order to create a consistent backup, MyISAM tables are write-locked during the copy process. While the instance is being backed up, you cannot add or delete databases, add or delete users, or delete, stop, or reboot the instance.

For more information about these engine types, see the following MySQL documentation:

• The InnoDB Storage Engine
• The MyISAM Storage Engine

The following table provides details about maximum numbers of connections and access to `my.cnf` file settings per database size:

You can provision instances with up to 64 GB of memory and up to 500 GB of disk storage. You can increase storage up to the maximum by using the Cloud Control Panel. To increase storage beyond 500 GB, submit a support ticket. Note that disk storage cannot be decreased on a running instance.

Yes. However, all users created through the Cloud Control Panel, API, and CLI have full permissions by default.

To create read-only users, you first must enable the root user and then use that user to generate and manage additional users with read-only privileges.

Yes. Currently, the root user can only be enabled through the public API or the CLI. After root is enabled, it cannot be disabled.

Cloud Databases supports MySQL 5.6, Percona 5.6, and MariaDB 10. For all newly created Cloud Databases instances, MySQL 5.6 is the default. We will continue to support MySQL 5.1 for legacy instances, but we recommend that our customers use the latest version of MySQL, Percona, or MariaDB because they offer significant performance improvements and newer features. For more information that will help you choose the right database version for your application, see Choosing the right database with Rackspace Cloud Databases.

The following table shows bandwidth in megabits per second (Mbps), based on instance size.

For release notes, API documentation, and a _Getting started_ guide for Cloud Databases,

see Rackspace Cloud Databases API v1.0.

Yes. By default, all accounts have a preconfigured set of thresholds that manage capacity and prevent abuse of the system. The system recognizes two kinds of limits: rate limits and absolute limits. _Rate limits_ are thresholds that are reset after a certain amount of time passes. _Absolute limits_ are fixed at the account level. For the most up-to-date information about rate and absolute limits (which include instance and volume limits),

see the Rackspace Cloud Databases Developer Guide.

If you cannot access your Cloud Databases instance, your data is still protected on a redundant SAN.

Cloud Databases provides several options for connecting to your database, which gives you complete flexibility in how you access your database. You can connect to your database by using the following methods:

• Public and private access with High Availability Groups

• Public and private access for Cloud Databases Instances

• Connect to a Cloud Databases instance

You can also use the Cloud Control Panel, the API, or the CLI to manage your database instance. While some features are not available in the Cloud Control Panel, you can access these features through the API or the CLI. More information about the API and the CLI is available in the Cloud Databases API documentation.

You can set the default time zone for a Cloud Databases instance of MySQL by creating a configuration group that sets the `default\_time\_zone` parameter to the offset from UTC. (For example,`-6:00` for CST.)

For more information, see Set the time zone for a Cloud Databases instance.

Yes. You can use standard MySQL client tools to import data into and export data from your instance. For more information, see Import data into Cloud Databases and Export Data from MySQL.

Yes. Configuration settings for Cloud Databases instances can be stored and applied by using the Cloud Control Panel and the Cloud Databases API.

You can save your settings in configuration groups, and then apply each configuration group to multiple instances. You can maintain multiple configuration groups to account for different workloads.

Access to MySQL is allowed only over port 3306. Shell-level access is not available. Full MySQL access can be obtained by enabling the root user on the database instance.

The default storage engine is InnoDB. However, other storage engines such as MyISAM that are included with MySQL also work for certain use cases.

A Cloud Databases High Availability (HA) instance group includes a source database instance with one or two replicas. If the source database instance becomes unavailable, an automatic failover is initiated to one of the replicas. The automatic failover and promotion of the new instance is completed within a short downtime of approximately 10 to 30 seconds.

When you provision an HA instance group, you can choose any flavor from 1 GB to 64 GB.

Yes. Both on-demand and scheduled backups are available for HA instance groups.

Yes. However, resizes of HA instance groups can be applied only to the entire group and cannot be applied to individual instances in the HA group.

For details on the technical architecture used to create Cloud Databases HA instances, see High Availability for Cloud Databases.

Yes. Rackspace ensures that if the source database instance becomes unavailable, an automatic failover is initiated to the replicas within 10 to 30 seconds of downtime.

HA instances carry a small premium per instance over regular Cloud Databases instances and are charged per instance, similar to replica sets. (An HA group with a primary and one replica counts as two instances.) The additional cost per instance covers the load balancer containers that are added for HA instance groups. For the latest pricing information, see the Cloud Databases product page.

Rackspace currently supports MySQL 5.6, Percona 5.6, and MariaDB 10 for HA database instances.

Rackspace enables you to add a maximum of two replicas to the primary source database instance. You can have a maximum of three total instances in the HA group: one primary and two replicas. In the future, we will increase the number of replicas that you can add to an HA group.

For more information about HA, see High Availability for Cloud Databases.

Yes. You can convert replica sets to HA groups through the API or the Cloud Control Panel. We will add the ability to convert a single instance to an HA group at a later date.

Yes. You can monitor replication using the monitoring agent installed on the instance. For more information, see Database replication with Cloud Databases and Monitoring Read Replication in the API documentation.

No. You can only add replica of the same database type and version as your source database instance.

No. You can only add a replica in the same region as your source database instance.

When you add a replica to the source DB instance, the instance is restarted. This means that your database is unavailable until the instance restarts.

Each read replica added is charged the same way as a new instance.

Replication does not currently support automatic failover. If your database instance goes down and you want to use a replica instance to minimize downtime, you must do a manual failover to the replica instances. For manual failover, you must detach the replica from the source instance and change the application endpoint to this new source database instance.

If you're interested in automatic failover, take a look at our High Availability instances.

Replication is supported for MySQL 5.6, Percona 5.6, and MariaDB 10. We do not support replication for MySQL 5.1 because this is an older version of MySQL and there have been significant improvements for replication support in newer versions of MySQL.

We highly recommend all users to upgrade from MySQL 5.1 to MySQL 5.6.

Yes we do support primary-replica replication. You can add and manage replicas using Cloud Control Panel and Cloud Databases API. For more information about managing replication by using the API, see the API documentation for replication.

------------------------------------------------------------------------

Pricing information is available at the Cloud Databases pricing page. Standard charges apply for any Cloud Servers, Cloud Load Balancers, or Cloud Sites that are used to access your Cloud Database instances.