Drush Reload Parallel

Reloads Drupal databases using drush.

This module will dump all the tables from a source database in parallel and simultaneously begin importing tables into a destination database. Once the dump and import is complete drush updb will be run against the destination site. This is particularly useful for very large databases that would otherwise take a long time to dump then import.

Installation

$ npm install -g drush-reloadp

Usage

$ drush-reloadp -s @source.alias -d @dest.alias

Advanced usage

$ drush-reloadp -s <source.alias> -d <dest.alias> [-t <table1>[,<table2>...]] [-vr] [--version]

Options:
  -s, --source       The source drush alias to dump the database from.                  [required]
  -d, --dest         The destination drush alias to import the database to.             [required]
  -v, --verbose      Print more information about what's happening during the process.
  -t, --skip-tables  Comma delimited list of tables to skip imports of.
  -r, --skip-drop    Skip dropping tables from the destination database.
  --version          Return the version of drush-reloadp.

Caveats

This method sacrifices consistency for speed. If the source database is receiving new data it's possible to end up with a missing relations for data that was created on tables that were already imported.

Some strategies to avoid this:

1. Run this code against a low traffic site (i.e. a staging environment).
2. Lock the source site so new content cannot be created.
3. Do nothing – in practice this hasn't been a problem for the sites I've used it on.

License

MIT.

Credits

This module was inspired by mysql-parallel.