Database Cookbook

The main highlight of this cookbook is the database and database_user resources for managing databases and database users in a RDBMS. Providers for MySQL, PostgreSQL and SQL Server are also provided, see usage documentation below.

This cookbook also contains recipes to configure mysql database masters and slaves and uses EBS for storage, integrating together with the application cookbook utilizing data bags for application related information. These recipes are written primarily to use MySQL and the Opscode mysql cookbook. Other RDBMS may be supported at a later date. This cookbook does not automatically restore database dumps, but does install tools to help with that.

Requirements

Chef version 0.10.10+.

Platforms

• Debian, Ubuntu
• Red Hat, CentOS, Scientific, Fedora, Amazon

Cookbooks

The following Opscode cookbooks are dependencies:

• mysql
• postgresql
• xfs
• aws

Resources/Providers

These resources aim to expose an abstraction layer for interacting with different RDBMS in a general way. Currently the cookbook ships with providers for MySQL, PostgreSQL and SQL Server. Please see specific usage in the Example sections below. The providers use specific Ruby gems installed under Chef's Ruby environment to execute commands and carry out actions. These gems will need to be installed before the providers can operate correctly. Specific notes for each RDBS flavor:

• MySQL: leverages the mysql gem which is installed as part of the mysql::ruby recipe. You must declare include_recipe "database::mysql" to include this in your recipe.
• PostgreSQL: leverages the pg gem which is installed as part of the postgresql::ruby recipe. You must declare include_recipe "database::postgresql" to include this.
• SQL Server: leverages the tiny_tds gem which is installed as part of the sql_server::client recipe.

This cookbook is not in charge of installing the Database Management System itself. Therefore, if you want to install MySQL, for instance, you should add include_recipe "mysql::server" in your recipe, or include mysql::server in the node run_list.

database

Manage databases in a RDBMS. Use the proper shortcut resource depending on your RDBMS: mysql_database, postgresql_database or sql_server_database.

Actions

• :create: create a named database
• :drop: drop a named database
• :query: execute an arbitrary query against a named database

Attribute Parameters

• database_name: name attribute. Name of the database to interact with
• connection: hash of connection info. valid keys include :host, :port, :username, :password and :socket (only for MySQL DB*)
• sql: string of sql or a block that executes to a string of sql, which will be executed against the database. used by :query action only

* The database cookbook uses the mysql gem, which uses the real_connect() function from mysql API to connect to the server.

"The value of host may be either a host name or an IP address. If host is NULL or the string "localhost", a connection to the local host is assumed. For Windows, the client connects using a shared-memory connection, if the server has shared-memory connections enabled. Otherwise, TCP/IP is used. For Unix, the client connects using a Unix socket file. For local connections, you can also influence the type of connection to use with the MYSQL_OPT_PROTOCOL or MYSQL_OPT_NAMED_PIPE options to mysql_options(). The type of connection must be supported by the server. For a host value of "." on Windows, the client connects using a named pipe, if the server has named-pipe connections enabled. If named-pipe connections are not enabled, an error occurs."

If you set the :host key to "localhost" or if you leave it blank, a socket will be used. By default real_connect() function will look for socket in /var/lib/mysql/mysql.sock. If your socket file in non-default location - you can use :socket key to specify that location.

Providers

• Chef::Provider::Database::Mysql: shortcut resource mysql_database
• Chef::Provider::Database::Postgresql: shortcut resource postgresql_database
• Chef::Provider::Database::SqlServer: shortcut resource sql_server_database

Examples

# Create a mysql database
mysql_database 'oracle_rules' do
  connection(
    :host     => 'localhost',
    :username => 'root',
    :password => node['mysql']['server_root_password']
  )
  action :create
end

# Create a sql server database
sql_server_database 'mr_softie' do
  connection(
    :host     => '127.0.0.1',
    :port     => node['sql_server']['port'],
    :username => 'sa',
    :password => node['sql_server']['server_sa_password']
  )
  action :create
end

# create a postgresql database
postgresql_database 'mr_softie' do
  connection(
    :host      => '127.0.0.1'
    :port      => 5432,
    :username  => 'postgres',
    :password  => node['postgresql']['password']['postgres']
  )
  action :create
end

# create a postgresql database with additional parameters
postgresql_database 'mr_softie' do
  connection(
    :host     => '127.0.0.1',
    :port     => 5432,
    :username => 'postgres',
    :password => node['postgresql']['password']['postgres']
  )
  template 'DEFAULT'
  encoding 'DEFAULT'
  tablespace 'DEFAULT'
  connection_limit '-1'
  owner 'postgres'
  action :create
end

# Externalize conection info in a ruby hash
mysql_connection_info = {
  :host     => 'localhost',
  :username => 'root',
  :password => node['mysql']['server_root_password']
}

sql_server_connection_info = {
  :host     => 'localhost',
  :port     => node['sql_server']['port'],
  :username => 'sa',
  :password => node['sql_server']['server_sa_password']
}

postgresql_connection_info = {
  :host     => '127.0.0.1',
  :port     => node['postgresql']['config']['port'],
  :username => 'postgres',
  :password => node['postgresql']['password']['postgres']
}



# Same create commands, connection info as an external hash
mysql_database 'foo' do
  connection mysql_connection_info
  action :create
end

sql_server_database 'foo' do
  connection sql_server_connection_info
  action     :create
end

postgresql_database 'foo' do
  connection postgresql_connection_info
  action     :create
end



# Create database, set provider in resource parameter
database 'bar' do
  connection mysql_connection_info
  provider   Chef::Provider::Database::Mysql
  action     :create
end

database 'bar' do
  connection sql_server_connection_info
  provider   Chef::Provider::Database::SqlServer
  action     :create
end

database 'bar' do
  connection postgresql_connection_info
  provider   Chef::Provider::Database::Postgresql
  action     :create
end



# Drop a database
mysql_database 'baz' do
  connection mysql_connection_info
  action    :drop
end



# Query a database
mysql_database 'flush the privileges' do
  connection mysql_connection_info
  sql        'flush privileges'
  action     :query
end



# Query a database from a sql script on disk
mysql_database 'run script' do
  connection mysql_connection_info
  sql { ::File.open('/path/to/sql_script.sql').read }
  action :query
end



# Vacuum a postgres database
postgresql_database 'vacuum databases' do
  connection      postgresql_connection_info
  database_table 'template1'
  sql 'VACUUM FULL VERBOSE ANALYZE'
  action :query
end

database_user

Manage users and user privileges in a RDBMS. Use the proper shortcut resource depending on your RDBMS: mysql_database_user, postgresql_database_user, or sql_server_database_user.

Actions

• :create: create a user
• :drop: drop a user
• :grant: manipulate user privileges on database objects

Attribute Parameters

• username: name attribute. Name of the database user
• password: password for the user account
• database_name: Name of the database to interact with
• connection: hash of connection info. valid keys include :host, :port, :username, :password
• privileges: array of database privileges to grant user. used by the :grant action. default is :all
• grant_option: appends 'WITH GRANT OPTION' to grant statement. used by MySQL provider only. default is 'false'
• host: host where user connections are allowed from. used by MySQL provider only. default is 'localhost'
• table: table to grant privileges on. used by :grant action and MySQL provider only. default is '*' (all tables)

Providers

• Chef::Provider::Database::MysqlUser: shortcut resource mysql_database_user
• Chef::Provider::Database::PostgresqlUser: shortcut resource postgresql_database_user
• Chef::Provider::Database::SqlServerUser: shortcut resourcesql_server_database_user

Examples

# create connection info as an external ruby hash
mysql_connection_info = {
  :host     => 'localhost',
  :username => 'root',
  :password => node['mysql']['server_root_password']
}

postgresql_connection_info = {
  :host     => 'localhost',
  :port     => node['postgresql']['config']['port'],
  :username => 'postgres',
  :password => node['postgresql']['password']['postgres']
}

sql_server_connection_info = {
  :host     => 'localhost',
  :port     => node['sql_server']['port'],
  :username => 'sa',
  :password => node['sql_server']['server_sa_password']
}



# Create a mysql user but grant no privileges
mysql_database_user 'disenfranchised' do
  connection mysql_connection_info
  password   'super_secret'
  action     :create
end



# Do the same but pass the provider to the database resource
database_user 'disenfranchised' do
  connection mysql_connection_info
  password   'super_secret'
  provider   Chef::Provider::Database::MysqlUser
  action     :create
end



# Create a postgresql user but grant no privileges
postgresql_database_user 'disenfranchised' do
  connection postgresql_connection_info
  password   'super_secret'
  action     :create
end



# Do the same but pass the provider to the database resource
database_user 'disenfranchised' do
  connection postgresql_connection_info
  password   'super_secret'
  provider   Chef::Provider::Database::PostgresqlUser
  action     :create
end



# Create a sql server user but grant no privileges
sql_server_database_user 'disenfranchised' do
  connection sql_server_connection_info
  password   'super_secret'
  action     :create
end



# Drop a mysql user
mysql_database_user 'foo_user' do
  connection mysql_connection_info
  action     :drop
end



# Bulk drop sql server users
%w(disenfranchised foo_user).each do |user|
  sql_server_database_user user do
    connection sql_server_connection_info
    action     :drop
  end
end



# Grant SELECT, UPDATE, and INSERT privileges to all tables in foo db from all hosts
mysql_database_user 'foo_user' do
  connection    mysql_connection_info
  password      'super_secret'
  database_name 'foo'
  host          '%'
  privileges    [:select,:update,:insert]
  action        :grant
end



# Grant all privileges on all databases/tables from localhost
mysql_database_user 'super_user' do
  connection mysql_connection_info
  password   'super_secret'
  action     :grant
end



# Grant all privileges on all tables in foo db
postgresql_database_user 'foo_user' do
  connection    postgresql_connection_info
  database_name 'foo'
  privileges    [:all]
  action        :grant
end

# grant select,update,insert privileges to all tables in foo db
sql_server_database_user 'foo_user' do
  connection    sql_server_connection_info
  password      'super_secret'
  database_name 'foo'
  privileges    [:select,:update,:insert]
  action        :grant
end

Recipes

ebs_volume

Note: This recipe does not currently work on RHEL platforms due to the xfs cookbook not supporting RHEL yet.

Loads the aws information from the data bag. Searches the applications data bag for the database master or slave role and checks that role is applied to the node. Loads the EBS information and the master information from data bags. Uses the aws cookbook LWRP, aws_ebs_volume to manage the volume.

On a master node:

• if we have an ebs volume already as stored in a data bag, attach it
• if we don't have the ebs information then create a new one and attach it
• store the volume information in a data bag via a ruby block

On a slave node:

• use the master volume information to generate a snapshot
• create the new volume from the snapshot and attach it

Also on a master node, generate some configuration for running a snapshot via chef-solo from cron.

On a new filesystem volume, create as XFS, then mount it in /mnt, and also bind-mount it to the mysql data directory (default /var/lib/mysql).

master

This recipe no longer loads AWS specific information, and the database position for replication is no longer stored in a databag because the client might not have permission to write to the databag item. This may be handled in a different way at a future date.

Searches the apps databag for applications, and for each one it will check that the specified database master role is set in both the databag and applied to the node's run list. Then, retrieves the passwords for root, repl and debian users and saves them to the node attributes. If the passwords are not found in the databag, it prints a message that they'll be generated by the mysql cookbook.

Then it adds the application databag database settings to a hash, to use later.

Then it will iterate over the databases and create them with the mysql_database resource while adding privileges for application specific database users using the mysql_database_user resource.

slave

TODO: Retrieve the master status from a data bag, then start replication using a ruby block. The replication status needs to be handled in some other way for now since the master recipe above doesn't actually set it in the databag anymore.

snapshot

Run via Chef Solo. Retrieves the db snapshot configuration from the specified JSON file. Uses the mysql_database resource to lock and unlock tables, and does a filesystem freeze and EBS snapshot.

Deprecated Recipes

The following recipe is considered deprecated. It is kept for reference purposes.

ebs_backup

Older style of doing mysql snapshot and replication using Adam Jacob's ec2_mysql script and library.

Data Bags

This cookbook uses the apps data bag item for the specified application; see the application cookbook's README.md. It also creates data bag items in a bag named 'aws' for storing volume information. In order to interact with EC2, it expects aws to have a main item:

{
  "id": "main",
  "ec2_private_key": "private key as a string",
  "ec2_cert": "certificate as a string",
  "aws_account_id": "",
  "aws_secret_access_key": "",
  "aws_access_key_id": ""
}

Note: with the Open Source Chef Server, the server using the database recipes must be an admin client or it will not be able to create data bag items. You can modify whether the client is admin by editing it with knife.

knife client edit <client_name>
{
  ...
  "admin": true
  ...
}

This is not required if the Chef Server is Opscode Hosted Chef, instead use the ACL feature to modify access for the node to be able to update the data bag.

Usage

Aside from the application data bag (see the README in the application cookbook), create a role for the database master. Use a role.rb in your chef-repo, or create the role directly with knife.

{
  "name": "my_app_database_master",
  "chef_type": "role",
  "json_class": "Chef::Role",
  "default_attributes": {},
  "description": "",
  "run_list": [
    "recipe[mysql::server]",
    "recipe[database::master]"
  ],
  "override_attributes": {}
}

Create a production environment. This is also used in the application cookbook.

{
  "name": "production",
  "description": "",
  "cookbook_versions": {},
  "json_class": "Chef::Environment",
  "chef_type": "environment",
  "default_attributes": {},
  "override_attributes": {}
}

The cookbook my_app_database is recommended to set up any application specific database resources such as configuration templates, trending monitors, etc. It is not required, but you would need to create it separately in site-cookbooks. Add it to the my_app_database_master role.

License & Authors

• Author:: Adam Jacob ([email protected])
• Author:: Joshua Timberman ([email protected])
• Author:: AJ Christensen ([email protected])
• Author:: Seth Chisamore ([email protected])
• Author:: Lamont Granquist ([email protected])

Copyright 2009-2013, Opscode, Inc.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.