- Options Details
- Sqitch Commands
- See Also
sqitch - Sane database change management
sqitch [<options>] <command> [<command-options>] [<args>]
- No opinions
Sqitch is not integrated with any framework, ORM, or platform. Rather, it is a standalone change management system with no opinions about your database engine, application framework, or your development environment.
- Native scripting
- Dependency resolution
Database changes may declare dependencies on other changes -- even on changes from other Sqitch projects. This ensures proper order of execution, even when you've committed changes to your VCS out-of-order.
- No numbering
Change deployment is managed by maintaining a plan file. As such, there is no need to number your changes, although you can if you want. Sqitch doesn't much care how you name your changes.
- Iterative Development
Up until you tag and release your project, you can modify your change deployment scripts as often as you like. They're not locked in just because they've been committed to your VCS. This allows you to take an iterative approach to developing your database schema. Or, better, you can do test-driven database development.
Ready to get started? Here's where:
- Sqitch Tutorials
Detailed tutorials demonstrating the creation, development, and maintenance of a database with Sqitch.
- PDX.pm Presentation
Slides from "Sane Database Management with Sqitch", presented to the Portland Perl Mongers in January, 2013.
- PDXPUG Presentation
Movie of "Sane Database Management with Sqitch", presented to the Portland PostgreSQL Users Group in September, 2012.
- Agile Database Development
A named unit of change. A change name must be used in the file names of its deploy and a revert scripts. It may also be used in a verify script file name.
A known deployment state, pointing to a single change, typically corresponding to a release. Think of it is a version number or VCS revision. A given point in the plan may have any number of tags.
The current state of the database. This is represented by the most recently-deployed change. If the state of the database is the same as the most recent change, then it is considered "up-to-date".
A list of one or more changes and their dependencies that define the order of deployment execution. The plan is stored in a "plan file," usually named sqitch.plan. Sqitch reads the plan file to determine what changes to execute to change the database from one state to another.
A named database to which to deploy changes. Always has an associated connection URI, and may also have an associated command-line client and registry name.
The name of the database object where Sqitch's state and history data is stored. Typically a schema name (as in PostgreSQL and Oracle) or a database name (as in SQLite and MySQL).
The act of adding a change to the plan. Sqitch will generate scripts for the change, which you then may modify with the necessary code (typically DDLs) to actually deploy, revert, and verify the change.
The act of deploying changes to a database. Sqitch reads the plan, checks the current state of the database, and applies all the changes necessary to either bring the database up-to-date or to a requested state (a change name or tag).
The act of reverting database changes to reach an earlier deployment state. Sqitch reads the list of deployed changes from the database and reverts them in the reverse of the order in which they were applied. All changes may be reverted, or changes may be reverted to a requested state (a change name or tag).
User who commits or reverts changes to a database.
User who adds a change to the plan.
-f --plan-file FILE Path to a deployment plan file. --engine ENGINE Database engine. --registry REGISTRY Registry schema or database. --db-client PATH Path to engine command-line client. -d --db-name NAME Database name. -u --db-user USER Database user name. -h --db-host HOST Database server host name. -p --db-port PORT Database server port number. --top-dir DIR Path to directory with plan and scripts. --etc-path Print path to etc directory and exit. --quiet Quiet mode with non-error output suppressed. -v --verbose Increment verbosity. --version Print version number and exit. --help Show a list of commands and exit. --man Print introductory documentation and exit.
sqitch --plan-file plan.txt sqitch -f main.plan
Path to the deployment plan file. Defaults to $top_dir/sqitch.plan. See "Plan File" for a description of its structure.
sqitch --engine pg
The database engine to use. Supported engines include:
sqitch --registry registry
The name of the Sqitch registry schema or database. Sqitch will store its own data here.
sqitch --client /usr/local/pgsql/bin/psql
Path to the command-line client for the database engine. Defaults to a client in the current path named appropriately for the specified engine.
sqitch --db-name widgets sqitch -d bricolage
Name of the database. In general, targets and URIs are preferred, but this option can be used to override the database name in a target.
sqitch --db-username root sqitch --db-user postgres sqitch -u Mom
User name to use when connecting to the database. Does not apply to all engines. In general, targets and URIs are preferred, but this option can be used to override the user name in a target.
sqitch --db-host db.example.com sqitch -h appdb.example.net
Host name to use when connecting to the database. Does not apply to all engines. In general, targets and URIs are preferred, but this option can be used to override the host name in a target.
sqitch --db-port 7654 sqitch -p 5431
Port number to connect to. Does not apply to all engines.In general, targets and URIs are preferred, but this option can be used to override the port in a target.
sqitch --top-dir migrations/
Path to directory containing deploy, revert, and verify scripts. It should contain subdirectories named
revert, and (optionally)
verify. Defaults to
Print out the path to the Sqitch etc directory and exit. This is the directory where the system-wide configuration file lives, as well as change script templates.
Suppress normal output messages. Error messages will still be emitted to
STDERR. Overrides any value specified by
sqitch --verbose sqitch -vvv
Pass multiple times to specify a value between 0 and 3 to determine how verbose Sqitch should be. Unless
--quietis specified, the default is 1, meaning that Sqitch will output basic status messages as it does its thing. Values of 2 and 3 each cause greater verbosity. Ignored if
Outputs a brief description all known Sqitch commands and exits.
Outputs this documentation and exits.
Outputs the program name and version and exits.
Create the plan file and directories for deploy, revert, and verify scripts if they do not already exist. This command is useful for starting a new Sqitch project.
Output information about the current deployment state of a database, including the name of the last deployed change, as well as any tags applied to it. If any changes in the plan have not been deployed, they will be listed separately.
Search and Output the complete change history of a database. Provides information about when changes were deployed, reverted, or failed, as well as who planned and committed the changes, and when.
Add a new change.
List tags or tag the latest change.
Rework an existing change.
Manage target databases.
Deploy changes to a database
Revert changes from a database.
Verify changes deployed to a database.
Get and set project, user, or system Sqitch options.
Bundle a Sqitch project for distribution. This command copies the Sqitch configuration, plan, and deploy, revert, and verify scripts to a directory, so that it can be packaged up for distribution, such as in an RPM or tarball.
Show help for a specific command or, if no command is specified, show the same documentation as
Sqitch configuration can be set up on a project, user, or system-wide basis. The format of the configuration file, named sqitch.conf, is the same as for git.
Here's an example of a configuration file that might be useful checked into a VCS for a project that deploys to PostgreSQL and stores its deployment scripts with the extension ddl under the
migrations directory. It also wants bundle to be created in the _build/sql directory, and to deploy starting with the "gamma" tag:
[core] engine = pg top_dir = migrations extension = ddl [engine "pg"] target = widgetopolis [revert] to = gamma [bundle] from = gamma tags_only = yes dest_dir = _build/sql [target "widgetopolis"] uri = db:pg:widgetopolis
And here's an example of useful configuration in ~/.sqitch/sqitch.conf, to point to system-specific engine information:
[user] name = Marge N. O’Vera email = email@example.com [engine "pg"] client = /usr/local/pgsql/bin/psql [engine "mysql"] client = /usr/local/mysql/bin/mysql [engine "oracle"] client = /usr/local/instantclient_11_2/sqlplus [engine "sqlite"] client = /usr/local/bin/sqlite3
Various commands read from the configuration file and adjust their operation accordingly. See sqitch-config for a list.
The original design for Sqitch was sketched out in a number of blog posts:
Other tools that do database change management include:
- Rails migrations
Numbered migrations for Ruby on Rails.
Numbered changes in pure SQL, integrated with Perl's Module::Build build system. Does not support reversion.
Numbered migrations in pure SQL.
PostgreSQL-specific dependency-tracking solution by depesz.
David E. Wheeler <firstname.lastname@example.org>
Copyright (c) 2012-2018 iovation Inc.
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.