AfterMigrate

Automatically run database maintenance after Rails migrations

after_migrate detects tables touched during rails db:migrate (or related tasks) and runs the appropriate optimizer commands:

PostgreSQL → ANALYZE (affected tables or all) + optional VACUUM
SQLite → PRAGMA optimize (or VACUUM + ANALYZE)
MySQL → ANALYZE TABLE

Stale statistics and fragmentation after schema changes silently hurt query performance.
after_migrate fixes it - automatically and precisely.

Because every migration deserves a cleanup.

✨ Features

Smart detection of affected tables (CREATE/ALTER/INSERT/UPDATE/DELETE/etc.)
Zero false positives - ignores views, columns, system tables, and complex joins
Configurable via environment variables or initializer block
Supports db:migrate, db:rollback, db:migrate:redo
No monkey-patching of ActiveRecord core classes
Works in development, test, CI, and production
Rails 7.0+ / Ruby 3.2+ only

📦 Installation

Add to your Gemfile:

gem 'after_migrate'

Then run:

bundle install

🚀 Usage

The gem activates automatically. No code required for default behavior.

Default behavior (recommended)

Out of the box, it runs ANALYZE on only the tables touched during the migration (PostgreSQL default).

Configuration

Create config/initializers/after_migrate.rb:

enabled is auto-detected -- do not set it here.

AFTER_MIGRATE_ENABLED result

true always on

false always off (explicit opt-out wins)

unset on only while running db:migrate, db:rollback, apartment:migrate, db:schema:load, after_migrate:*

It must be resolved from the environment, because Rails runs railtie initializers before config/initializers/*.rb -- by the time this block executes, the gem has already decided whether to subscribe. Assigning config.enabled here still affects AfterMigrate.run!, but cannot switch collection on or off, and setting it to a fixed value will override auto-detection for run!. Leave it alone unless you mean it.

`AFTER_MIGRATE_ENABLED`	result
`true`	always on
`false`	always off (explicit opt-out wins)
unset	*on only while running `db:migrate`, `db:rollback`, `apartment:migrate`, `db:schema:load`, `after_migrate:`**

AfterMigrate.configure do |config|

  # Log what’s happening
  config.verbose = true

  # Run VACUUM on affected tables (PostgreSQL only)
  config.vacuum = false

  # Choose ANALYZE strategy
  # "only_affected_tables" - default, precise
  # "all_tables"           - full database analyze
  # "none"                 - skip ANALYZE entirely
  config.analyze = "only_affected_tables"

  # Enhance rake tasks (runs maintenance after db:migrate etc.)
  # Set to false in test env if needed
  config.rake_tasks_enhanced = true
end

Store backends

The collected table names are kept in a store. Choose one with config.store:

:memory (default) — a process-local Concurrent::Map. Fine for a single-process migration.
:file — JSON file, shared across rake invocations in the same run.
:redis — shared across processes; use this for parallel / multi-tenant migrations.

config.store = :redis

[!IMPORTANT] Redis: always pass a pooled or memoized client — never a proc that builds a new connection per call.

config.redis is read on every collected SQL statement. If it returns a fresh Redis.new each time (e.g. config.redis = -> { Redis.new(...) }), every statement opens a new, unpooled connection. Under parallel or multi-tenant migrations this exhausts the Redis server connection limit (Connection refused / max number of clients reached).

Recommended — a thread-safe ConnectionPool (the store uses pool.with):
config.store = :redis
config.redis = ConnectionPool.new(size: 2, timeout: 5) { Redis.new(url: ENV["REDIS_URL"]) }
A memoized single client also works for sequential migrations:
config.redis = -> { @after_migrate_redis ||= Redis.new(url: ENV["REDIS_URL"]) }

🚢 Releasing

Use the guarded helper to make sure git is pushed before RubyGems publish:

bin/release

What it does:

requires a clean git worktree
ensures you are on the default branch (origin/HEAD fallback)
checks branch sync status vs origin
runs bundle exec rspec and bundle exec rubocop
pushes the branch first
runs bundle exec rake release (build/tag/push/publish)

Optional:

bin/release --skip-checks

🤝 Contributing

Bug reports, feature requests, and pull requests are very welcome!

https://github.com/moskvin/after_migrate

📝 License

This project is available under the MIT License.