The Database service provides database management features.
The Database service provides scalable and reliable cloud provisioning functionality for both relational and non-relational database engines. Users can quickly and easily use database features without the burden of handling complex administrative tasks. Cloud users and database administrators can provision and manage multiple database instances as needed.
The Database service provides resource isolation at high performance levels, and automates complex administrative tasks such as deployment, configuration, patching, backups, restores, and monitoring.
You can modify various cluster characteristics by editing the /etc/trove/trove.conf file. A comprehensive list of the Database service configuration options is described in the Database service chapter in the Configuration Reference.
An administrative user can create data stores for a variety of databases.
This section assumes you do not yet have a MySQL data store, and shows you how to create a MySQL data store and populate it with a MySQL 5.5 data store version.
To create a data store
Create a trove image
Create an image for the type of database you want to use, for example, MySQL, MongoDB, Cassandra.
This image must have the trove guest agent installed, and it must have the trove-guestagent.conf file configured to connect to your OpenStack environment. To configure trove-guestagent.conf, add the following lines to trove-guestagent.conf on the guest instance you are using to build your image:
rabbit_host = controller
rabbit_password = RABBIT_PASS
nova_proxy_admin_user = admin
nova_proxy_admin_pass = ADMIN_PASS
nova_proxy_admin_tenant_name = service
trove_auth_url = http://controller:35357/v2.0
This example assumes you have created a MySQL 5.5 image called mysql-5.5.qcow2.
Important
If you have a guest image that was created with an OpenStack version before Kilo, modify the guest agent init script for the guest image to read the configuration files from the directory /etc/trove/conf.d.
For a backwards compatibility with pre-Kilo guest instances, set the database service configuration options injected_config_location to /etc/trove and guest_info to /etc/guest_info.
Register image with Image service
You need to register your guest image with the Image service.
In this example, you use the glance image-create command to register a mysql-5.5.qcow2 image.
$ glance image-create --name mysql-5.5 --disk-format qcow2 --container-format bare --is-public True < mysql-5.5.qcow2
+------------------+--------------------------------------+
| Property | Value |
+------------------+--------------------------------------+
| checksum | d41d8cd98f00b204e9800998ecf8427e |
| container_format | bare |
| created_at | 2014-05-23T21:01:18 |
| deleted | False |
| deleted_at | None |
| disk_format | qcow2 |
| id | bb75f870-0c33-4907-8467-1367f8cb15b6 |
| is_public | True |
| min_disk | 0 |
| min_ram | 0 |
| name | mysql-5.5 |
| owner | 1448da1223124bb291f5ae8e9af4270d |
| protected | False |
| size | 0 |
| status | active |
| updated_at | 2014-05-23T21:01:22 |
| virtual_size | None |
+------------------+--------------------------------------+
Create the data store
Create the data store that will house the new image. To do this, use the trove-manage datastore_update command.
This example uses the following arguments:
Argument | Description | In this example: |
---|---|---|
config file | The configuration file to use. | --config-file=/etc/trove/trove.conf |
name | Name you want to use for this data store. | mysql |
default version | You can attach multiple versions/images to a data store. For example, you might have a MySQL 5.5 version and a MySQL 5.6 version. You can designate one version as the default, which the system uses if a user does not explicitly request a specific version. | "" At this point, you do not yet have a default version, so pass in an empty string. |
Example:
$ trove-manage --config-file=/etc/trove/trove.conf datastore_update mysql ""
Add a version to the new data store
Now that you have a MySQL data store, you can add a version to it, using the trove-manage datastore_version_update command. The version indicates which guest image to use.
This example uses the following arguments:
Argument | Description | In this example: |
---|---|---|
config file | The configuration file to use. | --config-file=/etc/trove/trove.conf |
data store | The name of the data store you just created via trove-manage datastore_update. | mysql |
version name | The name of the version you are adding to the data store. | mysql-5.5 |
data store manager | Which data store manager to use for this version. Typically, the data store manager is identified by one of the following strings, depending on the database:
|
mysql |
glance ID | The ID of the guest image you just added to the Image service. You can get this ID by using the glance image-show IMAGE_NAME command. | bb75f870-0c33-4907-8467-1367f8cb15b6 |
packages | If you want to put additional packages on each guest that you create with this data store version, you can list the package names here. | "" In this example, the guest image already contains all the required packages, so leave this argument empty. |
active |
|
1 |
Example:
$ trove-manage --config-file=/etc/trove/trove.conf datastore_version_update mysql mysql-5.5 mysql GLANCE_ID "" 1
Optional. Set your new version as the default version. To do this, use the trove-manage datastore_update command again, this time specifying the version you just created.
$ trove-manage --config-file=/etc/trove/trove.conf datastore_update mysql mysql-5.5
Load validation rules for configuration groups
Note
Applies only to MySQL and Percona data stores
Background. You can manage database configuration tasks by using configuration groups. Configuration groups let you set configuration parameters, in bulk, on one or more databases.
When you set up a configuration group using the trove configuration-create command, this command compares the configuration values you are setting against a list of valid configuration values that are stored in the validation-rules.json file.
Operating System | Location of validation-rules.json | Notes |
---|---|---|
Ubuntu 14.04 | /usr/lib/python2.7/dist-packages/trove/templates/DATASTORE_NAME | DATASTORE_NAME is the name of either the MySQL data store or the Percona data store. This is typically either mysql or percona. |
RHEL 7, CentOS 7, Fedora 20, and Fedora 21 | /usr/lib/python2.7/site-packages/trove/templates/DATASTORE_NAME | DATASTORE_NAME is the name of either the MySQL data store or the Percona data store. This is typically either mysql or percona. |
Therefore, as part of creating a data store, you need to load the validation-rules.json file, using the trove-manage db_load_datastore_config_parameters command. This command takes the following arguments:
This example loads the validation-rules.json file for a MySQL database on Ubuntu 14.04:
$ trove-manage db_load_datastore_config_parameters mysql mysql-5.5 /usr/lib/python2.7/dist-packages/trove/templates/mysql/validation-rules.json
Validate data store
To validate your new data store and version, start by listing the data stores on your system:
$ trove datastore-list
+--------------------------------------+--------------+
| id | name |
+--------------------------------------+--------------+
| 10000000-0000-0000-0000-000000000001 | Legacy MySQL |
| e5dc1da3-f080-4589-a4c2-eff7928f969a | mysql |
+--------------------------------------+--------------+
Take the ID of the MySQL data store and pass it in with the datastore-version-list command:
$ trove datastore-version-list DATASTORE_ID
+--------------------------------------+-----------+
| id | name |
+--------------------------------------+-----------+
| 36a6306b-efd8-4d83-9b75-8b30dd756381 | mysql-5.5 |
+--------------------------------------+-----------+
The Database service supports a variety of both relational and non-relational database engines, but to a varying degree of support for each data store. The Database service project has defined several classifications that indicate the quality of support for each data store. Data stores also implement different extensions. An extension is called a strategy and is classified similar to data stores.
Valid classifications for a data store and a strategy are:
Each classification builds on the previous one. This means that a data store that meets the technical preview requirements must also meet all the requirements for experimental, and a data store that meets the stable requirements must also meet all the requirements for technical preview.
Requirements
Experimental
A data store is considered to be experimental if it meets these criteria:
A strategy is considered experimental if:
Technical preview
A data store is considered to be a technical preview if it meets the requirements of experimental and further:
It implements APIs required to plant and start the capabilities of the data store as defined in the Datastore Compatibility Matrix.
Note
It is not required that the data store implements all features like resize, backup, replication, or clustering to meet this classification.
It provides a mechanism for building a guest image that allows you to exercise its capabilities.
It meets the other Documented Technical Requirements.
Important
A strategy is not normally considered to be technical preview.
Stable
A data store or a strategy is considered stable if:
Initial Classifications
The following table shows the current classification assignments for the different data stores.
Classification | Data store |
---|---|
Stable | MySQL |
Technical Preview | Cassandra, MongoDB |
Experimental | All others |
An administrative user can configure various characteristics of a MongoDB cluster.
Query routers and config servers
Background. Each cluster includes at least one query router and one config server. Query routers and config servers count against your quota. When you delete a cluster, the system deletes the associated query router(s) and config server(s).
Configuration. By default, the system creates one query router and one config server per cluster. You can change this by editing the /etc/trove/trove.conf file. These settings are in the mongodb section of the file:
Setting | Valid values are: |
---|---|
num_config_servers_per_cluster | 1 or 3 |
num_query_routers_per_cluster | 1 or 3 |
Except where otherwise noted, this document is licensed under Creative Commons Attribution 3.0 License. See all OpenStack Legal Documents.