diff options
author | Douwe Maan <douwe@gitlab.com> | 2017-09-18 11:24:52 +0300 |
---|---|---|
committer | Douwe Maan <douwe@gitlab.com> | 2017-09-18 11:24:52 +0300 |
commit | 36ad91d75a91bacb9b1b1c9997e734166cf4e201 (patch) | |
tree | 03e52e684612d3aa6f83363372940e3ea769b2ed /lib/gitlab/database | |
parent | b7cd5897098ec99b5eca7a3ebcc1c70d7501cd23 (diff) | |
parent | 7b7fb754118c7e86f94c4e5efaea632929d293da (diff) |
Merge branch 'mk-delete-conflicting-redirects-mysql' into 'master'
Clean up redirect routes that conflict with regular routes
Closes #36229
See merge request gitlab-org/gitlab-ce!13783
Diffstat (limited to 'lib/gitlab/database')
-rw-r--r-- | lib/gitlab/database/migration_helpers.rb | 88 |
1 files changed, 88 insertions, 0 deletions
diff --git a/lib/gitlab/database/migration_helpers.rb b/lib/gitlab/database/migration_helpers.rb index fb14798efe6..2c35da8f1aa 100644 --- a/lib/gitlab/database/migration_helpers.rb +++ b/lib/gitlab/database/migration_helpers.rb @@ -1,6 +1,9 @@ module Gitlab module Database module MigrationHelpers + 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 + # Adds `created_at` and `updated_at` columns with timezone information. # # This method is an improved version of Rails' built-in method `add_timestamps`. @@ -653,6 +656,91 @@ 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 end |