diff options
Diffstat (limited to 'lib/gitlab/background_migration/copy_column_using_background_migration_job.rb')
| -rw-r--r-- | lib/gitlab/background_migration/copy_column_using_background_migration_job.rb | 49 |
1 files changed, 12 insertions, 37 deletions
diff --git a/lib/gitlab/background_migration/copy_column_using_background_migration_job.rb b/lib/gitlab/background_migration/copy_column_using_background_migration_job.rb index 137b4d4bc4e..826845935b8 100644 --- a/lib/gitlab/background_migration/copy_column_using_background_migration_job.rb +++ b/lib/gitlab/background_migration/copy_column_using_background_migration_job.rb @@ -13,50 +13,25 @@ module Gitlab # - We skip the NULL checks as they may result in not using an index scan # - The table that is migrated does _not_ need `id` as the primary key # We use the provided primary_key column to perform the update. - class CopyColumnUsingBackgroundMigrationJob < BaseJob - include Gitlab::Database::DynamicModelHelpers + class CopyColumnUsingBackgroundMigrationJob < BatchedMigrationJob + def perform(copy_from, copy_to) + assignment_clauses = build_assignment_clauses(copy_from, copy_to) - # start_id - The start ID of the range of rows to update. - # end_id - The end ID of the range of rows to update. - # batch_table - The name of the table that contains the columns. - # batch_column - The name of the column we use to batch over the table. - # sub_batch_size - We don't want updates to take more than ~100ms - # This allows us to run multiple smaller batches during - # the minimum 2.minute interval that we can schedule jobs - # pause_ms - The number of milliseconds to sleep between each subbatch execution. - # copy_from - List of columns containing the data to copy. - # copy_to - List of columns to copy the data to. Order must match the order in `copy_from`. - def perform(start_id, end_id, batch_table, batch_column, sub_batch_size, pause_ms, copy_from, copy_to) - copy_from = Array.wrap(copy_from) - copy_to = Array.wrap(copy_to) - - raise ArgumentError, 'number of source and destination columns must match' unless copy_from.count == copy_to.count - - assignment_clauses = column_assignment_clauses(copy_from, copy_to) - - parent_batch_relation = relation_scoped_to_range(batch_table, batch_column, start_id, end_id) - - parent_batch_relation.each_batch(column: batch_column, of: sub_batch_size) do |sub_batch| - batch_metrics.time_operation(:update_all) do - sub_batch.update_all(assignment_clauses) - end - - pause_ms = 0 if pause_ms < 0 - sleep(pause_ms * 0.001) + each_sub_batch(operation_name: :update_all) do |relation| + relation.update_all(assignment_clauses) end end - def batch_metrics - @batch_metrics ||= Gitlab::Database::BackgroundMigration::BatchMetrics.new - end - private - def relation_scoped_to_range(source_table, source_key_column, start_id, stop_id) - define_batchable_model(source_table, connection: connection).where(source_key_column => start_id..stop_id) - end + def build_assignment_clauses(copy_from, copy_to) + copy_from = Array.wrap(copy_from) + copy_to = Array.wrap(copy_to) + + unless copy_from.count == copy_to.count + raise ArgumentError, 'number of source and destination columns must match' + end - def column_assignment_clauses(copy_from, copy_to) assignments = copy_from.zip(copy_to).map do |from_column, to_column| from_column = connection.quote_column_name(from_column) to_column = connection.quote_column_name(to_column) |