chuckya/lib/tasks/db.rake
ThibG 1242e57c27
Deal with collation-related index corruption (#14860)
* Add tootctl maintenance fix-duplicates

This tool goes through the database to detect and fix duplicates.
This operation is very slow and may cause data loss (of data that would be
inaccessible without intervention because of the existing index corruptions).
It tries its best to make sensible decisions, and asks the user in some cases.

* Add warning message in db:migrate hook

* Clear Rails cache after being done with database deduplication

Avoids followers hash cache being incorrect, among other things
2020-11-19 17:37:49 +01:00

86 lines
3 KiB
Ruby

# frozen_string_literal: true
require_relative '../mastodon/snowflake'
def each_schema_load_environment
# If we're in development, also run this for the test environment.
# This is a somewhat hacky way to do this, so here's why:
# 1. We have to define this before we load the schema, or we won't
# have a timestamp_id function when we get to it in the schema.
# 2. db:setup calls db:schema:load_if_ruby, which calls
# db:schema:load, which we define above as having a prerequisite
# of this task.
# 3. db:schema:load ends up running
# ActiveRecord::Tasks::DatabaseTasks.load_schema_current, which
# calls a private method `each_current_configuration`, which
# explicitly also does the loading for the `test` environment
# if the current environment is `development`, so we end up
# needing to do the same, and we can't even use the same method
# to do it.
if Rails.env.development?
test_conf = ActiveRecord::Base.configurations['test']
if test_conf['database']&.present?
ActiveRecord::Base.establish_connection(:test)
yield
ActiveRecord::Base.establish_connection(Rails.env.to_sym)
end
end
yield
end
namespace :db do
namespace :migrate do
desc 'Setup the db or migrate depending on state of db'
task setup: :environment do
begin
if ActiveRecord::Migrator.current_version.zero?
Rake::Task['db:migrate'].invoke
Rake::Task['db:seed'].invoke
end
rescue ActiveRecord::NoDatabaseError
Rake::Task['db:setup'].invoke
else
Rake::Task['db:migrate'].invoke
end
end
end
task :post_migration_hook do
at_exit do
unless %w(C POSIX).include?(ActiveRecord::Base.connection.execute('SELECT datcollate FROM pg_database WHERE datname = current_database();').first['datcollate'])
Rails.logger.warn 'WARNING: Your database is using an unsafe collation setting, which might result in index corruption.'
Rails.logger.warn 'WARNING: See https://docs.joinmastodon.org/admin/troubleshooting/index-corruption/#am-i-affected'
end
end
end
Rake::Task['db:migrate'].enhance(['db:post_migration_hook'])
# Before we load the schema, define the timestamp_id function.
# Idiomatically, we might do this in a migration, but then it
# wouldn't end up in schema.rb, so we'd need to figure out a way to
# get it in before doing db:setup as well. This is simpler, and
# ensures it's always in place.
Rake::Task['db:schema:load'].enhance ['db:define_timestamp_id']
# After we load the schema, make sure we have sequences for each
# table using timestamp IDs.
Rake::Task['db:schema:load'].enhance do
Rake::Task['db:ensure_id_sequences_exist'].invoke
end
task :define_timestamp_id do
each_schema_load_environment do
Mastodon::Snowflake.define_timestamp_id
end
end
task :ensure_id_sequences_exist do
each_schema_load_environment do
Mastodon::Snowflake.ensure_id_sequences_exist
end
end
end