Make IdsToBigints (mostly!) non-blocking (#5088)

* Make IdsToBigints (mostly!) non-blocking

This pulls in GitLab's MigrationHelpers, which include code to make
column changes in ways that Postgres can do without locking. In general,
this involves creating a new column, adding an index and any foreign
keys as appropriate, adding a trigger to keep it populated alongside
the old column, and then progressively copying data over to the new
column, before removing the old column and replacing it with the new
one.

A few changes to GitLab's MigrationHelpers were necessary:

* Some changes were made to remove dependencies on other GitLab code.
* We explicitly wait for index creation before forging ahead on column
  replacements.
* We use different temporary column names, to avoid running into index
  name length limits.
* We rename the generated indices back to what they "should" be after
  replacing columns.
* We rename the generated foreign keys to use the new column names when
  we had to create them. (This allows the migration to be rolled back
  without incident.)

# Big Scary Warning

There are two things here that may trip up large instances:

1. The change for tables' "id" columns is not concurrent. In
   particular, the stream_entries table may be big, and does not
   concurrently migrate its id column. (On the other hand, x_id type
   columns are all concurrent.)
2. This migration will take a long time to run, *but it should not
   lock tables during that time* (with the exception of the "id"
   columns as described above). That means this should probably be run
   in `screen` or some other session that can be run for a long time.
   Notably, the migration will take *longer* than it would without
   these changes, but the website will still be responsive during that
   time.

These changes were tested on a relatively large statuses table (256k
entries), and the service remained responsive during the migration.
Migrations both forward and backward were tested.

* Rubocop fixes

* MigrationHelpers: Support ID columns in some cases

This doesn't work in cases where the ID column is referred to as a
foreign key by another table.

* MigrationHelpers: support foreign keys for ID cols

Note that this does not yet support foreign keys on non-primary-key
columns, but Mastodon also doesn't yet have any that we've needed to
migrate.

This means we can perform fully "concurrent" migrations to change ID
column types, and the IdsToBigints migration can happen with effectively
no downtime. (A few operations require a transaction, such as renaming
columns or deleting them, but these transactions should not block for
noticeable amounts of time.)

The algorithm for generating foreign key names has changed with this,
and therefore all of those changed in schema.rb.

* Provide status, allow for interruptions

The MigrationHelpers now allow restarting the rename of a column if it
was interrupted, by removing the old "new column" and re-starting the
process.

Along with this, they now provide status updates on the changes which
are happening, as well as indications about when the changes can be
safely interrupted (when there are at least 10 seconds estimated to be
left before copying data is complete).

The IdsToBigints migration now also sorts the columns it migrates by
size, starting with the largest tables. This should provide
administrators a worst-case scenario estimate for the length of
migrations: each successive change will get faster, giving admins a
chance to abort early on if they need to run the migration later. The
idea is that this does not force them to try to time interruptions
between smaller migrations.

* Fix column sorting in IdsToBigints

Not a significant change, but it impacts the order of columns in the
database and db/schema.rb.

* Actually pause before IdsToBigints
signup-info-prompt
aschmitz 2017-10-02 14:28:59 -05:00 committed by Eugen Rochko
parent 4453c9a9f5
commit 97c02c3389
18 changed files with 1202 additions and 223 deletions

View File

@ -3,11 +3,11 @@
#
# Table name: account_domain_blocks
#
# id :integer not null, primary key
# account_id :integer
# domain :string
# created_at :datetime not null
# updated_at :datetime not null
# account_id :integer
# id :integer not null, primary key
#
class AccountDomainBlock < ApplicationRecord

View File

@ -3,11 +3,11 @@
#
# Table name: blocks
#
# id :integer not null, primary key
# account_id :integer not null
# target_account_id :integer not null
# created_at :datetime not null
# updated_at :datetime not null
# account_id :integer not null
# id :integer not null, primary key
# target_account_id :integer not null
#
class Block < ApplicationRecord

View File

@ -3,9 +3,9 @@
#
# Table name: conversation_mutes
#
# id :integer not null, primary key
# account_id :integer not null
# conversation_id :integer not null
# account_id :integer not null
# id :integer not null, primary key
#
class ConversationMute < ApplicationRecord

View File

@ -3,12 +3,12 @@
#
# Table name: domain_blocks
#
# id :integer not null, primary key
# domain :string default(""), not null
# created_at :datetime not null
# updated_at :datetime not null
# severity :integer default("silence")
# reject_media :boolean default(FALSE), not null
# id :integer not null, primary key
#
class DomainBlock < ApplicationRecord

View File

@ -3,11 +3,11 @@
#
# Table name: favourites
#
# id :integer not null, primary key
# account_id :integer not null
# status_id :integer not null
# created_at :datetime not null
# updated_at :datetime not null
# account_id :integer not null
# id :integer not null, primary key
# status_id :integer not null
#
class Favourite < ApplicationRecord

View File

@ -3,11 +3,11 @@
#
# Table name: follows
#
# id :integer not null, primary key
# account_id :integer not null
# target_account_id :integer not null
# created_at :datetime not null
# updated_at :datetime not null
# account_id :integer not null
# id :integer not null, primary key
# target_account_id :integer not null
#
class Follow < ApplicationRecord

View File

@ -3,11 +3,11 @@
#
# Table name: follow_requests
#
# id :integer not null, primary key
# account_id :integer not null
# target_account_id :integer not null
# created_at :datetime not null
# updated_at :datetime not null
# account_id :integer not null
# id :integer not null, primary key
# target_account_id :integer not null
#
class FollowRequest < ApplicationRecord

View File

@ -3,8 +3,6 @@
#
# Table name: imports
#
# id :integer not null, primary key
# account_id :integer not null
# type :integer not null
# approved :boolean default(FALSE), not null
# created_at :datetime not null
@ -13,6 +11,8 @@
# data_content_type :string
# data_file_size :integer
# data_updated_at :datetime
# account_id :integer not null
# id :integer not null, primary key
#
class Import < ApplicationRecord

View File

@ -3,11 +3,11 @@
#
# Table name: mentions
#
# id :integer not null, primary key
# account_id :integer
# status_id :integer
# created_at :datetime not null
# updated_at :datetime not null
# account_id :integer
# id :integer not null, primary key
#
class Mention < ApplicationRecord

View File

@ -3,11 +3,11 @@
#
# Table name: mutes
#
# id :integer not null, primary key
# account_id :integer not null
# target_account_id :integer not null
# created_at :datetime not null
# updated_at :datetime not null
# account_id :integer not null
# id :integer not null, primary key
# target_account_id :integer not null
#
class Mute < ApplicationRecord

View File

@ -3,15 +3,15 @@
#
# Table name: reports
#
# id :integer not null, primary key
# account_id :integer not null
# target_account_id :integer not null
# status_ids :integer default([]), not null, is an Array
# comment :text default(""), not null
# action_taken :boolean default(FALSE), not null
# created_at :datetime not null
# updated_at :datetime not null
# account_id :integer not null
# action_taken_by_account_id :integer
# id :integer not null, primary key
# target_account_id :integer not null
#
class Report < ApplicationRecord

View File

@ -3,13 +3,13 @@
#
# Table name: settings
#
# id :integer not null, primary key
# var :string not null
# value :text
# thing_type :string
# thing_id :integer
# created_at :datetime
# updated_at :datetime
# id :integer not null, primary key
# thing_id :integer
#
class Setting < RailsSettings::Base

View File

@ -1,16 +1,15 @@
# frozen_string_literal: true
# == Schema Information
#
# Table name: stream_entries
#
# id :integer not null, primary key
# account_id :integer
# activity_id :integer
# activity_type :string
# created_at :datetime not null
# updated_at :datetime not null
# hidden :boolean default(FALSE), not null
# account_id :integer
# id :integer not null, primary key
#
class StreamEntry < ApplicationRecord

View File

@ -3,16 +3,16 @@
#
# Table name: subscriptions
#
# id :integer not null, primary key
# callback_url :string default(""), not null
# secret :string
# expires_at :datetime
# confirmed :boolean default(FALSE), not null
# account_id :integer not null
# created_at :datetime not null
# updated_at :datetime not null
# last_successful_delivery_at :datetime
# domain :string
# account_id :integer not null
# id :integer not null, primary key
#
class Subscription < ApplicationRecord

View File

@ -3,11 +3,11 @@
#
# Table name: web_settings
#
# id :integer not null, primary key
# user_id :integer
# data :json
# created_at :datetime not null
# updated_at :datetime not null
# id :integer not null, primary key
# user_id :integer
#
class Web::Setting < ApplicationRecord

View File

@ -1,127 +1,119 @@
require Rails.root.join('lib', 'mastodon', 'migration_helpers')
class IdsToBigints < ActiveRecord::Migration[5.1]
include Mastodon::MigrationHelpers
disable_ddl_transaction!
INCLUDED_COLUMNS = [
[:account_domain_blocks, :account_id],
[:account_domain_blocks, :id],
[:accounts, :id],
[:blocks, :account_id],
[:blocks, :id],
[:blocks, :target_account_id],
[:conversation_mutes, :account_id],
[:conversation_mutes, :id],
[:domain_blocks, :id],
[:favourites, :account_id],
[:favourites, :id],
[:favourites, :status_id],
[:follow_requests, :account_id],
[:follow_requests, :id],
[:follow_requests, :target_account_id],
[:follows, :account_id],
[:follows, :id],
[:follows, :target_account_id],
[:imports, :account_id],
[:imports, :id],
[:media_attachments, :account_id],
[:media_attachments, :id],
[:mentions, :account_id],
[:mentions, :id],
[:mutes, :account_id],
[:mutes, :id],
[:mutes, :target_account_id],
[:notifications, :account_id],
[:notifications, :from_account_id],
[:notifications, :id],
[:oauth_access_grants, :application_id],
[:oauth_access_grants, :id],
[:oauth_access_grants, :resource_owner_id],
[:oauth_access_tokens, :application_id],
[:oauth_access_tokens, :id],
[:oauth_access_tokens, :resource_owner_id],
[:oauth_applications, :id],
[:oauth_applications, :owner_id],
[:reports, :account_id],
[:reports, :action_taken_by_account_id],
[:reports, :id],
[:reports, :target_account_id],
[:session_activations, :access_token_id],
[:session_activations, :user_id],
[:session_activations, :web_push_subscription_id],
[:settings, :id],
[:settings, :thing_id],
[:statuses, :account_id],
[:statuses, :application_id],
[:statuses, :in_reply_to_account_id],
[:stream_entries, :account_id],
[:stream_entries, :id],
[:subscriptions, :account_id],
[:subscriptions, :id],
[:tags, :id],
[:users, :account_id],
[:users, :id],
[:web_settings, :id],
[:web_settings, :user_id],
]
INCLUDED_COLUMNS << [:deprecated_preview_cards, :id] if table_exists?(:deprecated_preview_cards)
def migrate_columns(to_type)
# Print out a warning that this will probably take a while.
say ''
say 'WARNING: This migration may take a *long* time for large instances'
say 'It will *not* lock tables for any significant time, but it may run'
say 'for a very long time. We will pause for 10 seconds to allow you to'
say 'interrupt this migration if you are not ready.'
say ''
say 'This migration has some sections that can be safely interrupted'
say 'and restarted later, and will tell you when those are occurring.'
say ''
say 'For more information, see https://github.com/tootsuite/mastodon/pull/5088'
10.downto(1) do |i|
say "Continuing in #{i} second#{i == 1 ? '' : 's'}...", true
sleep 1
end
tables = INCLUDED_COLUMNS.map(&:first).uniq
table_sizes = {}
# Sort tables by their size
tables.each do |table|
table_sizes[table] = estimate_rows_in_table(table)
end
ordered_columns = INCLUDED_COLUMNS.sort_by do |col_parts|
[-table_sizes[col_parts.first], col_parts.last]
end
ordered_columns.each do |column_parts|
table, column = column_parts
# Skip this if we're resuming and already did this one.
next if column_for(table, column).sql_type == to_type.to_s
change_column_type_concurrently table, column, to_type
cleanup_concurrent_column_type_change table, column
end
end
def up
change_column :account_domain_blocks, :account_id, :bigint
change_column :account_domain_blocks, :id, :bigint
change_column :accounts, :id, :bigint
change_column :blocks, :account_id, :bigint
change_column :blocks, :id, :bigint
change_column :blocks, :target_account_id, :bigint
change_column :conversation_mutes, :account_id, :bigint
change_column :conversation_mutes, :id, :bigint
change_column :deprecated_preview_cards, :id, :bigint if table_exists?(:deprecated_preview_cards)
change_column :domain_blocks, :id, :bigint
change_column :favourites, :account_id, :bigint
change_column :favourites, :id, :bigint
change_column :favourites, :status_id, :bigint
change_column :follow_requests, :account_id, :bigint
change_column :follow_requests, :id, :bigint
change_column :follow_requests, :target_account_id, :bigint
change_column :follows, :account_id, :bigint
change_column :follows, :id, :bigint
change_column :follows, :target_account_id, :bigint
change_column :imports, :account_id, :bigint
change_column :imports, :id, :bigint
change_column :media_attachments, :account_id, :bigint
change_column :media_attachments, :id, :bigint
change_column :mentions, :account_id, :bigint
change_column :mentions, :id, :bigint
change_column :mutes, :account_id, :bigint
change_column :mutes, :id, :bigint
change_column :mutes, :target_account_id, :bigint
change_column :notifications, :account_id, :bigint
change_column :notifications, :from_account_id, :bigint
change_column :notifications, :id, :bigint
change_column :oauth_access_grants, :application_id, :bigint
change_column :oauth_access_grants, :id, :bigint
change_column :oauth_access_grants, :resource_owner_id, :bigint
change_column :oauth_access_tokens, :application_id, :bigint
change_column :oauth_access_tokens, :id, :bigint
change_column :oauth_access_tokens, :resource_owner_id, :bigint
change_column :oauth_applications, :id, :bigint
change_column :oauth_applications, :owner_id, :bigint
change_column :reports, :account_id, :bigint
change_column :reports, :action_taken_by_account_id, :bigint
change_column :reports, :id, :bigint
change_column :reports, :target_account_id, :bigint
change_column :session_activations, :access_token_id, :bigint
change_column :session_activations, :user_id, :bigint
change_column :session_activations, :web_push_subscription_id, :bigint
change_column :settings, :id, :bigint
change_column :settings, :thing_id, :bigint
change_column :statuses, :account_id, :bigint
change_column :statuses, :application_id, :bigint
change_column :statuses, :in_reply_to_account_id, :bigint
change_column :stream_entries, :account_id, :bigint
change_column :stream_entries, :id, :bigint
change_column :subscriptions, :account_id, :bigint
change_column :subscriptions, :id, :bigint
change_column :tags, :id, :bigint
change_column :users, :account_id, :bigint
change_column :users, :id, :bigint
change_column :web_settings, :id, :bigint
change_column :web_settings, :user_id, :bigint
migrate_columns(:bigint)
end
def down
change_column :account_domain_blocks, :account_id, :integer
change_column :account_domain_blocks, :id, :integer
change_column :accounts, :id, :integer
change_column :blocks, :account_id, :integer
change_column :blocks, :id, :integer
change_column :blocks, :target_account_id, :integer
change_column :conversation_mutes, :account_id, :integer
change_column :conversation_mutes, :id, :integer
change_column :deprecated_preview_cards, :id, :integer if table_exists?(:deprecated_preview_cards)
change_column :domain_blocks, :id, :integer
change_column :favourites, :account_id, :integer
change_column :favourites, :id, :integer
change_column :favourites, :status_id, :integer
change_column :follow_requests, :account_id, :integer
change_column :follow_requests, :id, :integer
change_column :follow_requests, :target_account_id, :integer
change_column :follows, :account_id, :integer
change_column :follows, :id, :integer
change_column :follows, :target_account_id, :integer
change_column :imports, :account_id, :integer
change_column :imports, :id, :integer
change_column :media_attachments, :account_id, :integer
change_column :media_attachments, :id, :integer
change_column :mentions, :account_id, :integer
change_column :mentions, :id, :integer
change_column :mutes, :account_id, :integer
change_column :mutes, :id, :integer
change_column :mutes, :target_account_id, :integer
change_column :notifications, :account_id, :integer
change_column :notifications, :from_account_id, :integer
change_column :notifications, :id, :integer
change_column :oauth_access_grants, :application_id, :integer
change_column :oauth_access_grants, :id, :integer
change_column :oauth_access_grants, :resource_owner_id, :integer
change_column :oauth_access_tokens, :application_id, :integer
change_column :oauth_access_tokens, :id, :integer
change_column :oauth_access_tokens, :resource_owner_id, :integer
change_column :oauth_applications, :id, :integer
change_column :oauth_applications, :owner_id, :integer
change_column :reports, :account_id, :integer
change_column :reports, :action_taken_by_account_id, :integer
change_column :reports, :id, :integer
change_column :reports, :target_account_id, :integer
change_column :session_activations, :access_token_id, :integer
change_column :session_activations, :user_id, :integer
change_column :session_activations, :web_push_subscription_id, :integer
change_column :settings, :id, :integer
change_column :settings, :thing_id, :integer
change_column :statuses, :account_id, :integer
change_column :statuses, :application_id, :integer
change_column :statuses, :in_reply_to_account_id, :integer
change_column :stream_entries, :account_id, :integer
change_column :stream_entries, :id, :integer
change_column :subscriptions, :account_id, :integer
change_column :subscriptions, :id, :integer
change_column :tags, :id, :integer
change_column :users, :account_id, :integer
change_column :users, :id, :integer
change_column :web_settings, :id, :integer
change_column :web_settings, :user_id, :integer
migrate_columns(:integer)
end
end

View File

@ -16,10 +16,10 @@ ActiveRecord::Schema.define(version: 20170927215609) do
enable_extension "plpgsql"
create_table "account_domain_blocks", force: :cascade do |t|
t.bigint "account_id"
t.string "domain"
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.bigint "account_id"
t.index ["account_id", "domain"], name: "index_account_domain_blocks_on_account_id_and_domain", unique: true
end
@ -69,16 +69,16 @@ ActiveRecord::Schema.define(version: 20170927215609) do
end
create_table "blocks", force: :cascade do |t|
t.bigint "account_id", null: false
t.bigint "target_account_id", null: false
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.bigint "account_id", null: false
t.bigint "target_account_id", null: false
t.index ["account_id", "target_account_id"], name: "index_blocks_on_account_id_and_target_account_id", unique: true
end
create_table "conversation_mutes", force: :cascade do |t|
t.bigint "account_id", null: false
t.bigint "conversation_id", null: false
t.bigint "account_id", null: false
t.index ["account_id", "conversation_id"], name: "index_conversation_mutes_on_account_id_and_conversation_id", unique: true
end
@ -111,33 +111,32 @@ ActiveRecord::Schema.define(version: 20170927215609) do
end
create_table "favourites", force: :cascade do |t|
t.bigint "account_id", null: false
t.bigint "status_id", null: false
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.bigint "account_id", null: false
t.bigint "status_id", null: false
t.index ["account_id", "id"], name: "index_favourites_on_account_id_and_id"
t.index ["account_id", "status_id"], name: "index_favourites_on_account_id_and_status_id", unique: true
t.index ["status_id"], name: "index_favourites_on_status_id"
end
create_table "follow_requests", force: :cascade do |t|
t.bigint "account_id", null: false
t.bigint "target_account_id", null: false
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.bigint "account_id", null: false
t.bigint "target_account_id", null: false
t.index ["account_id", "target_account_id"], name: "index_follow_requests_on_account_id_and_target_account_id", unique: true
end
create_table "follows", force: :cascade do |t|
t.bigint "account_id", null: false
t.bigint "target_account_id", null: false
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.bigint "account_id", null: false
t.bigint "target_account_id", null: false
t.index ["account_id", "target_account_id"], name: "index_follows_on_account_id_and_target_account_id", unique: true
end
create_table "imports", force: :cascade do |t|
t.bigint "account_id", null: false
t.integer "type", null: false
t.boolean "approved", default: false, null: false
t.datetime "created_at", null: false
@ -146,6 +145,7 @@ ActiveRecord::Schema.define(version: 20170927215609) do
t.string "data_content_type"
t.integer "data_file_size"
t.datetime "data_updated_at"
t.bigint "account_id", null: false
end
create_table "media_attachments", force: :cascade do |t|
@ -155,12 +155,12 @@ ActiveRecord::Schema.define(version: 20170927215609) do
t.integer "file_file_size"
t.datetime "file_updated_at"
t.string "remote_url", default: "", null: false
t.bigint "account_id"
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.string "shortcode"
t.integer "type", default: 0, null: false
t.json "file_meta"
t.bigint "account_id"
t.text "description"
t.index ["account_id"], name: "index_media_attachments_on_account_id"
t.index ["shortcode"], name: "index_media_attachments_on_shortcode", unique: true
@ -168,28 +168,28 @@ ActiveRecord::Schema.define(version: 20170927215609) do
end
create_table "mentions", force: :cascade do |t|
t.bigint "account_id"
t.bigint "status_id"
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.bigint "account_id"
t.index ["account_id", "status_id"], name: "index_mentions_on_account_id_and_status_id", unique: true
t.index ["status_id"], name: "index_mentions_on_status_id"
end
create_table "mutes", force: :cascade do |t|
t.bigint "account_id", null: false
t.bigint "target_account_id", null: false
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.bigint "account_id", null: false
t.bigint "target_account_id", null: false
t.index ["account_id", "target_account_id"], name: "index_mutes_on_account_id_and_target_account_id", unique: true
end
create_table "notifications", force: :cascade do |t|
t.bigint "account_id"
t.bigint "activity_id"
t.string "activity_type"
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.bigint "account_id"
t.bigint "from_account_id"
t.index ["account_id", "activity_id", "activity_type"], name: "account_activity", unique: true
t.index ["activity_id", "activity_type"], name: "index_notifications_on_activity_id_and_activity_type"
@ -197,26 +197,26 @@ ActiveRecord::Schema.define(version: 20170927215609) do
end
create_table "oauth_access_grants", force: :cascade do |t|
t.bigint "resource_owner_id", null: false
t.bigint "application_id", null: false
t.string "token", null: false
t.integer "expires_in", null: false
t.text "redirect_uri", null: false
t.datetime "created_at", null: false
t.datetime "revoked_at"
t.string "scopes"
t.bigint "application_id", null: false
t.bigint "resource_owner_id", null: false
t.index ["token"], name: "index_oauth_access_grants_on_token", unique: true
end
create_table "oauth_access_tokens", force: :cascade do |t|
t.bigint "resource_owner_id"
t.bigint "application_id"
t.string "token", null: false
t.string "refresh_token"
t.integer "expires_in"
t.datetime "revoked_at"
t.datetime "created_at", null: false
t.string "scopes"
t.bigint "application_id"
t.bigint "resource_owner_id"
t.index ["refresh_token"], name: "index_oauth_access_tokens_on_refresh_token", unique: true
t.index ["resource_owner_id"], name: "index_oauth_access_tokens_on_resource_owner_id"
t.index ["token"], name: "index_oauth_access_tokens_on_token", unique: true
@ -232,8 +232,8 @@ ActiveRecord::Schema.define(version: 20170927215609) do
t.datetime "updated_at"
t.boolean "superapp", default: false, null: false
t.string "website"
t.bigint "owner_id"
t.string "owner_type"
t.bigint "owner_id"
t.index ["owner_id", "owner_type"], name: "index_oauth_applications_on_owner_id_and_owner_type"
t.index ["uid"], name: "index_oauth_applications_on_uid", unique: true
end
@ -266,26 +266,26 @@ ActiveRecord::Schema.define(version: 20170927215609) do
end
create_table "reports", force: :cascade do |t|
t.bigint "account_id", null: false
t.bigint "target_account_id", null: false
t.bigint "status_ids", default: [], null: false, array: true
t.text "comment", default: "", null: false
t.boolean "action_taken", default: false, null: false
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.bigint "account_id", null: false
t.bigint "action_taken_by_account_id"
t.bigint "target_account_id", null: false
t.index ["account_id"], name: "index_reports_on_account_id"
t.index ["target_account_id"], name: "index_reports_on_target_account_id"
end
create_table "session_activations", force: :cascade do |t|
t.bigint "user_id", null: false
t.string "session_id", null: false
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.string "user_agent", default: "", null: false
t.inet "ip"
t.bigint "access_token_id"
t.bigint "user_id", null: false
t.bigint "web_push_subscription_id"
t.index ["session_id"], name: "index_session_activations_on_session_id", unique: true
t.index ["user_id"], name: "index_session_activations_on_user_id"
@ -295,9 +295,9 @@ ActiveRecord::Schema.define(version: 20170927215609) do
t.string "var", null: false
t.text "value"
t.string "thing_type"
t.bigint "thing_id"
t.datetime "created_at"
t.datetime "updated_at"
t.bigint "thing_id"
t.index ["thing_type", "thing_id", "var"], name: "index_settings_on_thing_type_and_thing_id_and_var", unique: true
end
@ -323,7 +323,6 @@ ActiveRecord::Schema.define(version: 20170927215609) do
create_table "statuses", force: :cascade do |t|
t.string "uri"
t.bigint "account_id", null: false
t.text "text", default: "", null: false
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
@ -332,8 +331,6 @@ ActiveRecord::Schema.define(version: 20170927215609) do
t.string "url"
t.boolean "sensitive", default: false, null: false
t.integer "visibility", default: 0, null: false
t.bigint "in_reply_to_account_id"
t.bigint "application_id"
t.text "spoiler_text", default: "", null: false
t.boolean "reply", default: false, null: false
t.integer "favourites_count", default: 0, null: false
@ -341,6 +338,9 @@ ActiveRecord::Schema.define(version: 20170927215609) do
t.string "language"
t.bigint "conversation_id"
t.boolean "local"
t.bigint "account_id", null: false
t.bigint "application_id"
t.bigint "in_reply_to_account_id"
t.index ["account_id", "id"], name: "index_statuses_on_account_id_id"
t.index ["conversation_id"], name: "index_statuses_on_conversation_id"
t.index ["in_reply_to_id"], name: "index_statuses_on_in_reply_to_id"
@ -356,12 +356,12 @@ ActiveRecord::Schema.define(version: 20170927215609) do
end
create_table "stream_entries", force: :cascade do |t|
t.bigint "account_id"
t.bigint "activity_id"
t.string "activity_type"
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.boolean "hidden", default: false, null: false
t.bigint "account_id"
t.index ["account_id"], name: "index_stream_entries_on_account_id"
t.index ["activity_id", "activity_type"], name: "index_stream_entries_on_activity_id_and_activity_type"
end
@ -371,11 +371,11 @@ ActiveRecord::Schema.define(version: 20170927215609) do
t.string "secret"
t.datetime "expires_at"
t.boolean "confirmed", default: false, null: false
t.bigint "account_id", null: false
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.datetime "last_successful_delivery_at"
t.string "domain"
t.bigint "account_id", null: false
t.index ["account_id", "callback_url"], name: "index_subscriptions_on_account_id_and_callback_url", unique: true
end
@ -389,7 +389,6 @@ ActiveRecord::Schema.define(version: 20170927215609) do
create_table "users", force: :cascade do |t|
t.string "email", default: "", null: false
t.bigint "account_id", null: false
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.string "encrypted_password", default: "", null: false
@ -415,6 +414,7 @@ ActiveRecord::Schema.define(version: 20170927215609) do
t.datetime "last_emailed_at"
t.string "otp_backup_codes", array: true
t.string "filtered_languages", default: [], null: false, array: true
t.bigint "account_id", null: false
t.index ["account_id"], name: "index_users_on_account_id"
t.index ["confirmation_token"], name: "index_users_on_confirmation_token", unique: true
t.index ["email"], name: "index_users_on_email", unique: true
@ -432,53 +432,53 @@ ActiveRecord::Schema.define(version: 20170927215609) do
end
create_table "web_settings", force: :cascade do |t|
t.bigint "user_id"
t.json "data"
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.bigint "user_id"
t.index ["user_id"], name: "index_web_settings_on_user_id", unique: true
end
add_foreign_key "account_domain_blocks", "accounts", on_delete: :cascade
add_foreign_key "blocks", "accounts", column: "target_account_id", on_delete: :cascade
add_foreign_key "blocks", "accounts", on_delete: :cascade
add_foreign_key "conversation_mutes", "accounts", on_delete: :cascade
add_foreign_key "account_domain_blocks", "accounts", name: "fk_206c6029bd", on_delete: :cascade
add_foreign_key "blocks", "accounts", column: "target_account_id", name: "fk_9571bfabc1", on_delete: :cascade
add_foreign_key "blocks", "accounts", name: "fk_4269e03e65", on_delete: :cascade
add_foreign_key "conversation_mutes", "accounts", name: "fk_225b4212bb", on_delete: :cascade
add_foreign_key "conversation_mutes", "conversations", on_delete: :cascade
add_foreign_key "favourites", "accounts", on_delete: :cascade
add_foreign_key "favourites", "statuses", on_delete: :cascade
add_foreign_key "follow_requests", "accounts", column: "target_account_id", on_delete: :cascade
add_foreign_key "follow_requests", "accounts", on_delete: :cascade
add_foreign_key "follows", "accounts", column: "target_account_id", on_delete: :cascade
add_foreign_key "follows", "accounts", on_delete: :cascade
add_foreign_key "imports", "accounts", on_delete: :cascade
add_foreign_key "media_attachments", "accounts", on_delete: :nullify
add_foreign_key "favourites", "accounts", name: "fk_5eb6c2b873", on_delete: :cascade
add_foreign_key "favourites", "statuses", name: "fk_b0e856845e", on_delete: :cascade
add_foreign_key "follow_requests", "accounts", column: "target_account_id", name: "fk_9291ec025d", on_delete: :cascade
add_foreign_key "follow_requests", "accounts", name: "fk_76d644b0e7", on_delete: :cascade
add_foreign_key "follows", "accounts", column: "target_account_id", name: "fk_745ca29eac", on_delete: :cascade
add_foreign_key "follows", "accounts", name: "fk_32ed1b5560", on_delete: :cascade
add_foreign_key "imports", "accounts", name: "fk_6db1b6e408", on_delete: :cascade
add_foreign_key "media_attachments", "accounts", name: "fk_96dd81e81b", on_delete: :nullify
add_foreign_key "media_attachments", "statuses", on_delete: :nullify
add_foreign_key "mentions", "accounts", on_delete: :cascade
add_foreign_key "mentions", "accounts", name: "fk_970d43f9d1", on_delete: :cascade
add_foreign_key "mentions", "statuses", on_delete: :cascade
add_foreign_key "mutes", "accounts", column: "target_account_id", on_delete: :cascade
add_foreign_key "mutes", "accounts", on_delete: :cascade
add_foreign_key "notifications", "accounts", column: "from_account_id", on_delete: :cascade
add_foreign_key "notifications", "accounts", on_delete: :cascade
add_foreign_key "oauth_access_grants", "oauth_applications", column: "application_id", on_delete: :cascade
add_foreign_key "oauth_access_grants", "users", column: "resource_owner_id", on_delete: :cascade
add_foreign_key "oauth_access_tokens", "oauth_applications", column: "application_id", on_delete: :cascade
add_foreign_key "oauth_access_tokens", "users", column: "resource_owner_id", on_delete: :cascade
add_foreign_key "oauth_applications", "users", column: "owner_id", on_delete: :cascade
add_foreign_key "reports", "accounts", column: "action_taken_by_account_id", on_delete: :nullify
add_foreign_key "reports", "accounts", column: "target_account_id", on_delete: :cascade
add_foreign_key "reports", "accounts", on_delete: :cascade
add_foreign_key "session_activations", "oauth_access_tokens", column: "access_token_id", on_delete: :cascade
add_foreign_key "session_activations", "users", on_delete: :cascade
add_foreign_key "status_pins", "accounts", on_delete: :cascade
add_foreign_key "mutes", "accounts", column: "target_account_id", name: "fk_eecff219ea", on_delete: :cascade
add_foreign_key "mutes", "accounts", name: "fk_b8d8daf315", on_delete: :cascade
add_foreign_key "notifications", "accounts", column: "from_account_id", name: "fk_fbd6b0bf9e", on_delete: :cascade
add_foreign_key "notifications", "accounts", name: "fk_c141c8ee55", on_delete: :cascade
add_foreign_key "oauth_access_grants", "oauth_applications", column: "application_id", name: "fk_34d54b0a33", on_delete: :cascade
add_foreign_key "oauth_access_grants", "users", column: "resource_owner_id", name: "fk_63b044929b", on_delete: :cascade
add_foreign_key "oauth_access_tokens", "oauth_applications", column: "application_id", name: "fk_f5fc4c1ee3", on_delete: :cascade
add_foreign_key "oauth_access_tokens", "users", column: "resource_owner_id", name: "fk_e84df68546", on_delete: :cascade
add_foreign_key "oauth_applications", "users", column: "owner_id", name: "fk_b0988c7c0a", on_delete: :cascade
add_foreign_key "reports", "accounts", column: "action_taken_by_account_id", name: "fk_bca45b75fd", on_delete: :nullify
add_foreign_key "reports", "accounts", column: "target_account_id", name: "fk_eb37af34f0", on_delete: :cascade
add_foreign_key "reports", "accounts", name: "fk_4b81f7522c", on_delete: :cascade
add_foreign_key "session_activations", "oauth_access_tokens", column: "access_token_id", name: "fk_957e5bda89", on_delete: :cascade
add_foreign_key "session_activations", "users", name: "fk_e5fda67334", on_delete: :cascade
add_foreign_key "status_pins", "accounts", name: "fk_d4cb435b62", on_delete: :cascade
add_foreign_key "status_pins", "statuses", on_delete: :cascade
add_foreign_key "statuses", "accounts", column: "in_reply_to_account_id", on_delete: :nullify
add_foreign_key "statuses", "accounts", on_delete: :cascade
add_foreign_key "statuses", "accounts", column: "in_reply_to_account_id", name: "fk_c7fa917661", on_delete: :nullify
add_foreign_key "statuses", "accounts", name: "fk_9bda1543f7", on_delete: :cascade
add_foreign_key "statuses", "statuses", column: "in_reply_to_id", on_delete: :nullify
add_foreign_key "statuses", "statuses", column: "reblog_of_id", on_delete: :cascade
add_foreign_key "statuses_tags", "statuses", on_delete: :cascade
add_foreign_key "statuses_tags", "tags", on_delete: :cascade
add_foreign_key "stream_entries", "accounts", on_delete: :cascade
add_foreign_key "subscriptions", "accounts", on_delete: :cascade
add_foreign_key "users", "accounts", on_delete: :cascade
add_foreign_key "web_settings", "users", on_delete: :cascade
add_foreign_key "statuses_tags", "tags", name: "fk_3081861e21", on_delete: :cascade
add_foreign_key "stream_entries", "accounts", name: "fk_5659b17554", on_delete: :cascade
add_foreign_key "subscriptions", "accounts", name: "fk_9847d1cbb5", on_delete: :cascade
add_foreign_key "users", "accounts", name: "fk_50500f500d", on_delete: :cascade
add_foreign_key "web_settings", "users", name: "fk_11910667b2", on_delete: :cascade
end

View File

@ -0,0 +1,988 @@
# frozen_string_literal: true
# This file is copied almost entirely from GitLab, which has done a large
# amount of work to ensure that migrations can happen with minimal downtime.
# Many thanks to those engineers.
# Changes have been made to remove dependencies on other GitLab files and to
# shorten temporary column names.
# Documentation on using these functions (and why one might do so):
# https://gitlab.com/gitlab-org/gitlab-ce/blob/master/doc/development/what_requires_downtime.md
# The file itself:
# https://gitlab.com/gitlab-org/gitlab-ce/blob/master/lib/gitlab/database/migration_helpers.rb
# It is licensed as follows:
# Copyright (c) 2011-2017 GitLab B.V.
# 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.
# This is bad form, but there are enough differences that it's impractical to do
# otherwise:
# rubocop:disable all
module Mastodon
module MigrationHelpers
# Stub for Database.postgresql? from GitLab
def self.postgresql?
ActiveRecord::Base.configurations[Rails.env]['adapter'].casecmp('postgresql').zero?
end
# Stub for Database.mysql? from GitLab
def self.mysql?
ActiveRecord::Base.configurations[Rails.env]['adapter'].casecmp('mysql2').zero?
end
# Model that can be used for querying permissions of a SQL user.
class Grant < ActiveRecord::Base
self.table_name =
if Mastodon::MigrationHelpers.postgresql?
'information_schema.role_table_grants'
else
'mysql.user'
end
def self.scope_to_current_user
if Mastodon::MigrationHelpers.postgresql?
where('grantee = user')
else
where("CONCAT(User, '@', Host) = current_user()")
end
end
# Returns true if the current user can create and execute triggers on the
# given table.
def self.create_and_execute_trigger?(table)
priv =
if Mastodon::MigrationHelpers.postgresql?
where(privilege_type: 'TRIGGER', table_name: table)
else
where(Trigger_priv: 'Y')
end
priv.scope_to_current_user.any?
end
end
BACKGROUND_MIGRATION_BATCH_SIZE = 1000 # Number of rows to process per job
BACKGROUND_MIGRATION_JOB_BUFFER_SIZE = 1000 # Number of jobs to bulk queue at a time
# Gets an estimated number of rows for a table
def estimate_rows_in_table(table_name)
exec_query('SELECT reltuples FROM pg_class WHERE relname = ' +
"'#{table_name}'").to_a.first['reltuples']
end
# Adds `created_at` and `updated_at` columns with timezone information.
#
# This method is an improved version of Rails' built-in method `add_timestamps`.
#
# Available options are:
# default - The default value for the column.
# null - When set to `true` the column will allow NULL values.
# The default is to not allow NULL values.
def add_timestamps_with_timezone(table_name, options = {})
options[:null] = false if options[:null].nil?
[:created_at, :updated_at].each do |column_name|
if options[:default] && transaction_open?
raise '`add_timestamps_with_timezone` with default value cannot be run inside a transaction. ' \
'You can disable transactions by calling `disable_ddl_transaction!` ' \
'in the body of your migration class'
end
# If default value is presented, use `add_column_with_default` method instead.
if options[:default]
add_column_with_default(
table_name,
column_name,
:datetime_with_timezone,
default: options[:default],
allow_null: options[:null]
)
else
add_column(table_name, column_name, :datetime_with_timezone, options)
end
end
end
# Creates a new index, concurrently when supported
#
# On PostgreSQL this method creates an index concurrently, on MySQL this
# creates a regular index.
#
# Example:
#
# add_concurrent_index :users, :some_column
#
# See Rails' `add_index` for more info on the available arguments.
def add_concurrent_index(table_name, column_name, options = {})
if transaction_open?
raise 'add_concurrent_index can not be run inside a transaction, ' \
'you can disable transactions by calling disable_ddl_transaction! ' \
'in the body of your migration class'
end
if MigrationHelpers.postgresql?
options = options.merge({ algorithm: :concurrently })
disable_statement_timeout
end
add_index(table_name, column_name, options)
end
# Removes an existed index, concurrently when supported
#
# On PostgreSQL this method removes an index concurrently.
#
# Example:
#
# remove_concurrent_index :users, :some_column
#
# See Rails' `remove_index` for more info on the available arguments.
def remove_concurrent_index(table_name, column_name, options = {})
if transaction_open?
raise 'remove_concurrent_index can not be run inside a transaction, ' \
'you can disable transactions by calling disable_ddl_transaction! ' \
'in the body of your migration class'
end
if supports_drop_index_concurrently?
options = options.merge({ algorithm: :concurrently })
disable_statement_timeout
end
remove_index(table_name, options.merge({ column: column_name }))
end
# Removes an existing index, concurrently when supported
#
# On PostgreSQL this method removes an index concurrently.
#
# Example:
#
# remove_concurrent_index :users, "index_X_by_Y"
#
# See Rails' `remove_index` for more info on the available arguments.
def remove_concurrent_index_by_name(table_name, index_name, options = {})
if transaction_open?
raise 'remove_concurrent_index_by_name can not be run inside a transaction, ' \
'you can disable transactions by calling disable_ddl_transaction! ' \
'in the body of your migration class'
end
if supports_drop_index_concurrently?
options = options.merge({ algorithm: :concurrently })
disable_statement_timeout
end
remove_index(table_name, options.merge({ name: index_name }))
end
# Only available on Postgresql >= 9.2
def supports_drop_index_concurrently?
return false unless MigrationHelpers.postgresql?
version = select_one("SELECT current_setting('server_version_num') AS v")['v'].to_i
version >= 90200
end
# Adds a foreign key with only minimal locking on the tables involved.
#
# This method only requires minimal locking when using PostgreSQL. When
# using MySQL this method will use Rails' default `add_foreign_key`.
#
# source - The source table containing the foreign key.
# target - The target table the key points to.
# column - The name of the column to create the foreign key on.
# on_delete - The action to perform when associated data is removed,
# defaults to "CASCADE".
def add_concurrent_foreign_key(source, target, column:, on_delete: :cascade, target_col: 'id')
# Transactions would result in ALTER TABLE locks being held for the
# duration of the transaction, defeating the purpose of this method.
if transaction_open?
raise 'add_concurrent_foreign_key can not be run inside a transaction'
end
# While MySQL does allow disabling of foreign keys it has no equivalent
# of PostgreSQL's "VALIDATE CONSTRAINT". As a result we'll just fall
# back to the normal foreign key procedure.
if MigrationHelpers.mysql?
return add_foreign_key(source, target,
column: column,
on_delete: on_delete)
else
on_delete = 'SET NULL' if on_delete == :nullify
end
disable_statement_timeout
key_name = concurrent_foreign_key_name(source, column, target_col)
# Using NOT VALID allows us to create a key without immediately
# validating it. This means we keep the ALTER TABLE lock only for a
# short period of time. The key _is_ enforced for any newly created
# data.
execute <<-EOF.strip_heredoc
ALTER TABLE #{source}
ADD CONSTRAINT #{key_name}
FOREIGN KEY (#{column})
REFERENCES #{target} (#{target_col})
#{on_delete ? "ON DELETE #{on_delete.upcase}" : ''}
NOT VALID;
EOF
# Validate the existing constraint. This can potentially take a very
# long time to complete, but fortunately does not lock the source table
# while running.
execute("ALTER TABLE #{source} VALIDATE CONSTRAINT #{key_name};")
end
# Returns the name for a concurrent foreign key.
#
# PostgreSQL constraint names have a limit of 63 bytes. The logic used
# here is based on Rails' foreign_key_name() method, which unfortunately
# is private so we can't rely on it directly.
def concurrent_foreign_key_name(table, column, target_col)
"fk_#{Digest::SHA256.hexdigest("#{table}_#{column}_#{target_col}_fk").first(10)}"
end
# Long-running migrations may take more than the timeout allowed by
# the database. Disable the session's statement timeout to ensure
# migrations don't get killed prematurely. (PostgreSQL only)
def disable_statement_timeout
execute('SET statement_timeout TO 0') if MigrationHelpers.postgresql?
end
# Updates the value of a column in batches.
#
# This method updates the table in batches of 5% of the total row count.
# This method will continue updating rows until no rows remain.
#
# When given a block this method will yield two values to the block:
#
# 1. An instance of `Arel::Table` for the table that is being updated.
# 2. The query to run as an Arel object.
#
# By supplying a block one can add extra conditions to the queries being
# executed. Note that the same block is used for _all_ queries.
#
# Example:
#
# update_column_in_batches(:projects, :foo, 10) do |table, query|
# query.where(table[:some_column].eq('hello'))
# end
#
# This would result in this method updating only rows where
# `projects.some_column` equals "hello".
#
# table - The name of the table.
# column - The name of the column to update.
# value - The value for the column.
#
# Rubocop's Metrics/AbcSize metric is disabled for this method as Rubocop
# determines this method to be too complex while there's no way to make it
# less "complex" without introducing extra methods (which actually will
# make things _more_ complex).
#
# rubocop: disable Metrics/AbcSize
def update_column_in_batches(table_name, column, value)
if transaction_open?
raise 'update_column_in_batches can not be run inside a transaction, ' \
'you can disable transactions by calling disable_ddl_transaction! ' \
'in the body of your migration class'
end
table = Arel::Table.new(table_name)
total = estimate_rows_in_table(table_name).to_i
if total == 0
count_arel = table.project(Arel.star.count.as('count'))
count_arel = yield table, count_arel if block_given?
total = exec_query(count_arel.to_sql).to_hash.first['count'].to_i
return if total == 0
end
# Update in batches of 5% until we run out of any rows to update.
batch_size = ((total / 100.0) * 5.0).ceil
max_size = 1000
# The upper limit is 1000 to ensure we don't lock too many rows. For
# example, for "merge_requests" even 1% of the table is around 35 000
# rows for GitLab.com.
batch_size = max_size if batch_size > max_size
start_arel = table.project(table[:id]).order(table[:id].asc).take(1)
start_arel = yield table, start_arel if block_given?
start_id = exec_query(start_arel.to_sql).to_hash.first['id'].to_i
say "Migrating #{table_name}.#{column} (~#{total.to_i} rows)"
started_time = Time.now
last_time = Time.now
migrated = 0
loop do
stop_row = nil
suppress_messages do
stop_arel = table.project(table[:id])
.where(table[:id].gteq(start_id))
.order(table[:id].asc)
.take(1)
.skip(batch_size)
stop_arel = yield table, stop_arel if block_given?
stop_row = exec_query(stop_arel.to_sql).to_hash.first
update_arel = Arel::UpdateManager.new
.table(table)
.set([[table[column], value]])
.where(table[:id].gteq(start_id))
if stop_row
stop_id = stop_row['id'].to_i
start_id = stop_id
update_arel = update_arel.where(table[:id].lt(stop_id))
end
update_arel = yield table, update_arel if block_given?
execute(update_arel.to_sql)
end
migrated += batch_size
if Time.now - last_time > 1
status = "Migrated #{migrated} rows"
percentage = 100.0 * migrated / total
status += " (~#{sprintf('%.2f', percentage)}%, "
remaining_time = (100.0 - percentage) * (Time.now - started_time) / percentage
status += "#{(remaining_time / 60).to_i}:"
status += sprintf('%02d', remaining_time.to_i % 60)
status += ' remaining, '
# Tell users not to interrupt if we're almost done.
if remaining_time > 10
status += 'safe to interrupt'
else
status += 'DO NOT interrupt'
end
status += ')'
say status, true
last_time = Time.now
end
# There are no more rows left to update.
break unless stop_row
end
end
# Adds a column with a default value without locking an entire table.
#
# This method runs the following steps:
#
# 1. Add the column with a default value of NULL.
# 2. Change the default value of the column to the specified value.
# 3. Update all existing rows in batches.
# 4. Set a `NOT NULL` constraint on the column if desired (the default).
#
# These steps ensure a column can be added to a large and commonly used
# table without locking the entire table for the duration of the table
# modification.
#
# table - The name of the table to update.
# column - The name of the column to add.
# type - The column type (e.g. `:integer`).
# default - The default value for the column.
# limit - Sets a column limit. For example, for :integer, the default is
# 4-bytes. Set `limit: 8` to allow 8-byte integers.
# allow_null - When set to `true` the column will allow NULL values, the
# default is to not allow NULL values.
#
# This method can also take a block which is passed directly to the
# `update_column_in_batches` method.
def add_column_with_default(table, column, type, default:, limit: nil, allow_null: false, &block)
if transaction_open?
raise 'add_column_with_default can not be run inside a transaction, ' \
'you can disable transactions by calling disable_ddl_transaction! ' \
'in the body of your migration class'
end
disable_statement_timeout
transaction do
if limit
add_column(table, column, type, default: nil, limit: limit)
else
add_column(table, column, type, default: nil)
end
# Changing the default before the update ensures any newly inserted
# rows already use the proper default value.
change_column_default(table, column, default)
end
begin
update_column_in_batches(table, column, default, &block)
change_column_null(table, column, false) unless allow_null
# We want to rescue _all_ exceptions here, even those that don't inherit
# from StandardError.
rescue Exception => error # rubocop: disable all
remove_column(table, column)
raise error
end
end
# Renames a column without requiring downtime.
#
# Concurrent renames work by using database triggers to ensure both the
# old and new column are in sync. However, this method will _not_ remove
# the triggers or the old column automatically; this needs to be done
# manually in a post-deployment migration. This can be done using the
# method `cleanup_concurrent_column_rename`.
#
# table - The name of the database table containing the column.
# old - The old column name.
# new - The new column name.
# type - The type of the new column. If no type is given the old column's
# type is used.
def rename_column_concurrently(table, old, new, type: nil)
if transaction_open?
raise 'rename_column_concurrently can not be run inside a transaction'
end
check_trigger_permissions!(table)
trigger_name = rename_trigger_name(table, old, new)
# If we were in the middle of update_column_in_batches, we should remove
# the old column and start over, as we have no idea where we were.
if column_for(table, new)
if MigrationHelpers.postgresql?
remove_rename_triggers_for_postgresql(table, trigger_name)
else
remove_rename_triggers_for_mysql(trigger_name)
end
remove_column(table, new)
end
old_col = column_for(table, old)
new_type = type || old_col.type
col_opts = {
precision: old_col.precision,
scale: old_col.scale,
}
# We may be trying to reset the limit on an integer column type, so let
# Rails handle that.
unless [:bigint, :integer].include?(new_type)
col_opts[:limit] = old_col.limit
end
add_column(table, new, new_type, col_opts)
# We set the default value _after_ adding the column so we don't end up
# updating any existing data with the default value. This isn't
# necessary since we copy over old values further down.
change_column_default(table, new, old_col.default) if old_col.default
quoted_table = quote_table_name(table)
quoted_old = quote_column_name(old)
quoted_new = quote_column_name(new)
if MigrationHelpers.postgresql?
install_rename_triggers_for_postgresql(trigger_name, quoted_table,
quoted_old, quoted_new)
else
install_rename_triggers_for_mysql(trigger_name, quoted_table,
quoted_old, quoted_new)
end
update_column_in_batches(table, new, Arel::Table.new(table)[old])
change_column_null(table, new, false) unless old_col.null
copy_indexes(table, old, new)
copy_foreign_keys(table, old, new)
end
# Changes the type of a column concurrently.
#
# table - The table containing the column.
# column - The name of the column to change.
# new_type - The new column type.
def change_column_type_concurrently(table, column, new_type)
temp_column = rename_column_name(column)
rename_column_concurrently(table, column, temp_column, type: new_type)
# Primary keys don't necessarily have an associated index.
if ActiveRecord::Base.get_primary_key(table) == column.to_s
old_pk_index_name = "index_#{table}_on_#{column}"
new_pk_index_name = "index_#{table}_on_#{column}_cm"
unless indexes_for(table, column).find{|i| i.name == old_pk_index_name}
add_concurrent_index(table, [temp_column], {
unique: true,
name: new_pk_index_name
})
end
end
end
# Performs cleanup of a concurrent type change.
#
# table - The table containing the column.
# column - The name of the column to change.
# new_type - The new column type.
def cleanup_concurrent_column_type_change(table, column)
temp_column = rename_column_name(column)
# Wait for the indices to be built
indexes_for(table, column).each do |index|
expected_name = index.name + '_cm'
puts "Waiting for index #{expected_name}"
sleep 1 until indexes_for(table, temp_column).find {|i| i.name == expected_name }
end
was_primary = (ActiveRecord::Base.get_primary_key(table) == column.to_s)
old_default_fn = column_for(table, column).default_function
old_fks = []
if was_primary
# Get any foreign keys pointing at this column we need to recreate, and
# remove the old ones.
# Based on code from:
# http://errorbank.blogspot.com/2011/03/list-all-foreign-keys-references-for.html
old_fks_res = execute <<-EOF.strip_heredoc
select m.relname as src_table,
(select a.attname
from pg_attribute a
where a.attrelid = m.oid
and a.attnum = o.conkey[1]
and a.attisdropped = false) as src_col,
o.conname as name,
o.confdeltype as on_delete
from pg_constraint o
left join pg_class f on f.oid = o.confrelid
left join pg_class c on c.oid = o.conrelid
left join pg_class m on m.oid = o.conrelid
where o.contype = 'f'
and o.conrelid in (
select oid from pg_class c where c.relkind = 'r')
and f.relname = '#{table}';
EOF
old_fks = old_fks_res.to_a
old_fks.each do |old_fk|
add_concurrent_foreign_key(
old_fk['src_table'],
table,
column: old_fk['src_col'],
target_col: temp_column,
on_delete: extract_foreign_key_action(old_fk['on_delete'])
)
remove_foreign_key(old_fk['src_table'], name: old_fk['name'])
end
end
# If there was a sequence owned by the old column, make it owned by the
# new column, as it will otherwise be deleted when we get rid of the
# old column.
if (seq_match = /^nextval\('([^']*)'(::text|::regclass)?\)/.match(old_default_fn))
seq_name = seq_match[1]
execute("ALTER SEQUENCE #{seq_name} OWNED BY #{table}.#{temp_column}")
end
transaction do
# This has to be performed in a transaction as otherwise we might have
# inconsistent data.
cleanup_concurrent_column_rename(table, column, temp_column)
rename_column(table, temp_column, column)
# If there was an old default function, we didn't copy it. Do that now
# in the transaction, so we don't miss anything.
change_column_default(table, column, -> { old_default_fn }) if old_default_fn
end
# Rename any indices back to what they should be.
indexes_for(table, column).each do |index|
next unless index.name.end_with?('_cm')
real_index_name = index.name.sub(/_cm$/, '')
rename_index(table, index.name, real_index_name)
end
# Rename any foreign keys back to names based on the real column.
foreign_keys_for(table, column).each do |fk|
old_fk_name = concurrent_foreign_key_name(fk.from_table, temp_column, 'id')
new_fk_name = concurrent_foreign_key_name(fk.from_table, column, 'id')
execute("ALTER TABLE #{fk.from_table} RENAME CONSTRAINT " +
"#{old_fk_name} TO #{new_fk_name}")
end
# Rename any foreign keys from other tables to names based on the real
# column.
old_fks.each do |old_fk|
old_fk_name = concurrent_foreign_key_name(old_fk['src_table'],
old_fk['src_col'], temp_column)
new_fk_name = concurrent_foreign_key_name(old_fk['src_table'],
old_fk['src_col'], column)
execute("ALTER TABLE #{old_fk['src_table']} RENAME CONSTRAINT " +
"#{old_fk_name} TO #{new_fk_name}")
end
# If the old column was a primary key, mark the new one as a primary key.
if was_primary
execute("ALTER TABLE #{table} ADD PRIMARY KEY USING INDEX " +
"index_#{table}_on_#{column}")
end
end
# Cleans up a concurrent column name.
#
# This method takes care of removing previously installed triggers as well
# as removing the old column.
#
# table - The name of the database table.
# old - The name of the old column.
# new - The name of the new column.
def cleanup_concurrent_column_rename(table, old, new)
trigger_name = rename_trigger_name(table, old, new)
check_trigger_permissions!(table)
if MigrationHelpers.postgresql?
remove_rename_triggers_for_postgresql(table, trigger_name)
else
remove_rename_triggers_for_mysql(trigger_name)
end
remove_column(table, old)
end
# Performs a concurrent column rename when using PostgreSQL.
def install_rename_triggers_for_postgresql(trigger, table, old, new)
execute <<-EOF.strip_heredoc
CREATE OR REPLACE FUNCTION #{trigger}()
RETURNS trigger AS
$BODY$
BEGIN
NEW.#{new} := NEW.#{old};
RETURN NEW;
END;
$BODY$
LANGUAGE 'plpgsql'
VOLATILE
EOF
execute <<-EOF.strip_heredoc
CREATE TRIGGER #{trigger}
BEFORE INSERT OR UPDATE
ON #{table}
FOR EACH ROW
EXECUTE PROCEDURE #{trigger}()
EOF
end
# Installs the triggers necessary to perform a concurrent column rename on
# MySQL.
def install_rename_triggers_for_mysql(trigger, table, old, new)
execute <<-EOF.strip_heredoc
CREATE TRIGGER #{trigger}_insert
BEFORE INSERT
ON #{table}
FOR EACH ROW
SET NEW.#{new} = NEW.#{old}
EOF
execute <<-EOF.strip_heredoc
CREATE TRIGGER #{trigger}_update
BEFORE UPDATE
ON #{table}
FOR EACH ROW
SET NEW.#{new} = NEW.#{old}
EOF
end
# Removes the triggers used for renaming a PostgreSQL column concurrently.
def remove_rename_triggers_for_postgresql(table, trigger)
execute("DROP TRIGGER IF EXISTS #{trigger} ON #{table}")
execute("DROP FUNCTION IF EXISTS #{trigger}()")
end
# Removes the triggers used for renaming a MySQL column concurrently.
def remove_rename_triggers_for_mysql(trigger)
execute("DROP TRIGGER IF EXISTS #{trigger}_insert")
execute("DROP TRIGGER IF EXISTS #{trigger}_update")
end
# Returns the (base) name to use for triggers when renaming columns.
def rename_trigger_name(table, old, new)
'trigger_' + Digest::SHA256.hexdigest("#{table}_#{old}_#{new}").first(12)
end
# Returns the name to use for temporary rename columns.
def rename_column_name(base)
base.to_s + '_cm'
end
# Returns an Array containing the indexes for the given column
def indexes_for(table, column)
column = column.to_s
indexes(table).select { |index| index.columns.include?(column) }
end
# Returns an Array containing the foreign keys for the given column.
def foreign_keys_for(table, column)
column = column.to_s
foreign_keys(table).select { |fk| fk.column == column }
end
# Copies all indexes for the old column to a new column.
#
# table - The table containing the columns and indexes.
# old - The old column.
# new - The new column.
def copy_indexes(table, old, new)
old = old.to_s
new = new.to_s
indexes_for(table, old).each do |index|
new_columns = index.columns.map do |column|
column == old ? new : column
end
# This is necessary as we can't properly rename indexes such as
# "ci_taggings_idx".
name = index.name + '_cm'
# If the order contained the old column, map it to the new one.
order = index.orders
if order.key?(old)
order[new] = order.delete(old)
end
options = {
unique: index.unique,
name: name,
length: index.lengths,
order: order
}
# These options are not supported by MySQL, so we only add them if
# they were previously set.
options[:using] = index.using if index.using
options[:where] = index.where if index.where
add_concurrent_index(table, new_columns, options)
end
end
# Copies all foreign keys for the old column to the new column.
#
# table - The table containing the columns and indexes.
# old - The old column.
# new - The new column.
def copy_foreign_keys(table, old, new)
foreign_keys_for(table, old).each do |fk|
add_concurrent_foreign_key(fk.from_table,
fk.to_table,
column: new,
on_delete: fk.on_delete)
end
end
# Returns the column for the given table and column name.
def column_for(table, name)
name = name.to_s
columns(table).find { |column| column.name == name }
end
# This will replace the first occurance of a string in a column with
# the replacement
# On postgresql we can use `regexp_replace` for that.
# On mysql we find the location of the pattern, and overwrite it
# with the replacement
def replace_sql(column, pattern, replacement)
quoted_pattern = Arel::Nodes::Quoted.new(pattern.to_s)
quoted_replacement = Arel::Nodes::Quoted.new(replacement.to_s)
if MigrationHelpers.mysql?
locate = Arel::Nodes::NamedFunction
.new('locate', [quoted_pattern, column])
insert_in_place = Arel::Nodes::NamedFunction
.new('insert', [column, locate, pattern.size, quoted_replacement])
Arel::Nodes::SqlLiteral.new(insert_in_place.to_sql)
else
replace = Arel::Nodes::NamedFunction
.new("regexp_replace", [column, quoted_pattern, quoted_replacement])
Arel::Nodes::SqlLiteral.new(replace.to_sql)
end
end
def remove_foreign_key_without_error(*args)
remove_foreign_key(*args)
rescue ArgumentError
end
def sidekiq_queue_migrate(queue_from, to:)
while sidekiq_queue_length(queue_from) > 0
Sidekiq.redis do |conn|
conn.rpoplpush "queue:#{queue_from}", "queue:#{to}"
end
end
end
def sidekiq_queue_length(queue_name)
Sidekiq.redis do |conn|
conn.llen("queue:#{queue_name}")
end
end
def check_trigger_permissions!(table)
unless Grant.create_and_execute_trigger?(table)
dbname = ActiveRecord::Base.configurations[Rails.env]['database']
user = ActiveRecord::Base.configurations[Rails.env]['username'] || ENV['USER']
raise <<-EOF
Your database user is not allowed to create, drop, or execute triggers on the
table #{table}.
If you are using PostgreSQL you can solve this by logging in to the GitLab
database (#{dbname}) using a super user and running:
ALTER #{user} WITH SUPERUSER
For MySQL you instead need to run:
GRANT ALL PRIVILEGES ON *.* TO #{user}@'%'
Both queries will grant the user super user permissions, ensuring you don't run
into similar problems in the future (e.g. when new tables are created).
EOF
end
end
# Bulk queues background migration jobs for an entire table, batched by ID range.
# "Bulk" meaning many jobs will be pushed at a time for efficiency.
# If you need a delay interval per job, then use `queue_background_migration_jobs_by_range_at_intervals`.
#
# model_class - The table being iterated over
# job_class_name - The background migration job class as a string
# batch_size - The maximum number of rows per job
#
# Example:
#
# class Route < ActiveRecord::Base
# include EachBatch
# self.table_name = 'routes'
# end
#
# bulk_queue_background_migration_jobs_by_range(Route, 'ProcessRoutes')
#
# Where the model_class includes EachBatch, and the background migration exists:
#
# class Gitlab::BackgroundMigration::ProcessRoutes
# def perform(start_id, end_id)
# # do something
# end
# end
def bulk_queue_background_migration_jobs_by_range(model_class, job_class_name, batch_size: BACKGROUND_MIGRATION_BATCH_SIZE)
raise "#{model_class} does not have an ID to use for batch ranges" unless model_class.column_names.include?('id')
jobs = []
model_class.each_batch(of: batch_size) do |relation|
start_id, end_id = relation.pluck('MIN(id), MAX(id)').first
if jobs.length >= BACKGROUND_MIGRATION_JOB_BUFFER_SIZE
# Note: This code path generally only helps with many millions of rows
# We push multiple jobs at a time to reduce the time spent in
# Sidekiq/Redis operations. We're using this buffer based approach so we
# don't need to run additional queries for every range.
BackgroundMigrationWorker.perform_bulk(jobs)
jobs.clear
end
jobs << [job_class_name, [start_id, end_id]]
end
BackgroundMigrationWorker.perform_bulk(jobs) unless jobs.empty?
end
# Queues background migration jobs for an entire table, batched by ID range.
# Each job is scheduled with a `delay_interval` in between.
# If you use a small interval, then some jobs may run at the same time.
#
# model_class - The table being iterated over
# job_class_name - The background migration job class as a string
# delay_interval - The duration between each job's scheduled time (must respond to `to_f`)
# batch_size - The maximum number of rows per job
#
# Example:
#
# class Route < ActiveRecord::Base
# include EachBatch
# self.table_name = 'routes'
# end
#
# queue_background_migration_jobs_by_range_at_intervals(Route, 'ProcessRoutes', 1.minute)
#
# Where the model_class includes EachBatch, and the background migration exists:
#
# class Gitlab::BackgroundMigration::ProcessRoutes
# def perform(start_id, end_id)
# # do something
# end
# end
def queue_background_migration_jobs_by_range_at_intervals(model_class, job_class_name, delay_interval, batch_size: BACKGROUND_MIGRATION_BATCH_SIZE)
raise "#{model_class} does not have an ID to use for batch ranges" unless model_class.column_names.include?('id')
model_class.each_batch(of: batch_size) do |relation, index|
start_id, end_id = relation.pluck('MIN(id), MAX(id)').first
# `BackgroundMigrationWorker.bulk_perform_in` schedules all jobs for
# the same time, which is not helpful in most cases where we wish to
# spread the work over time.
BackgroundMigrationWorker.perform_in(delay_interval * index, job_class_name, [start_id, end_id])
end
end
end
end
# rubocop:enable all