Welcome to mirror list, hosted at ThFree Co, Russian Federation.

pipeline_worker.rb « bulk_imports « workers « app - gitlab.com/gitlab-org/gitlab-foss.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
blob: ca006f81813f1015688533a8402511a059bed40b (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
# frozen_string_literal: true

module BulkImports
  class PipelineWorker
    include ApplicationWorker
    include ExclusiveLeaseGuard
    include Gitlab::Utils::StrongMemoize

    FILE_EXTRACTION_PIPELINE_PERFORM_DELAY = 10.seconds

    LimitedBatches = Struct.new(:numbers, :final?, keyword_init: true).freeze

    DEFER_ON_HEALTH_DELAY = 5.minutes

    data_consistency :always
    feature_category :importers
    sidekiq_options dead: false, retry: 6
    worker_has_external_dependencies!
    deduplicate :until_executing
    worker_resource_boundary :memory
    idempotent!

    version 2

    sidekiq_retries_exhausted do |msg, exception|
      new.perform_failure(msg['args'][0], msg['args'][2], exception)
    end

    defer_on_database_health_signal(:gitlab_main, [], DEFER_ON_HEALTH_DELAY) do |job_args, schema, tables|
      pipeline_tracker = ::BulkImports::Tracker.find(job_args.first)
      pipeline_schema = ::BulkImports::PipelineSchemaInfo.new(
        pipeline_tracker.pipeline_class,
        pipeline_tracker.entity.portable_class
      )

      if pipeline_schema.db_schema && pipeline_schema.db_table
        schema = pipeline_schema.db_schema
        tables = [pipeline_schema.db_table]
      end

      [schema, tables]
    end

    def self.defer_on_database_health_signal?
      Feature.enabled?(:bulk_import_deferred_workers)
    end

    # Keep _stage parameter for backwards compatibility.
    def perform(pipeline_tracker_id, _stage, entity_id)
      @entity = ::BulkImports::Entity.find(entity_id)
      @pipeline_tracker = ::BulkImports::Tracker.find(pipeline_tracker_id)

      log_extra_metadata_on_done(:pipeline_class, @pipeline_tracker.pipeline_name)

      try_obtain_lease do
        if pipeline_tracker.enqueued? || pipeline_tracker.started?
          logger.info(log_attributes(message: 'Pipeline starting'))
          run
        end
      end
    end

    def perform_failure(pipeline_tracker_id, entity_id, exception)
      @entity = ::BulkImports::Entity.find(entity_id)
      @pipeline_tracker = ::BulkImports::Tracker.find(pipeline_tracker_id)

      fail_pipeline(exception)
    end

    private

    attr_reader :pipeline_tracker, :entity

    def run
      return skip_tracker if entity.failed?

      raise(Pipeline::FailedError, "Export from source instance failed: #{export_status.error}") if export_failed?
      raise(Pipeline::ExpiredError, 'Empty export status on source instance') if empty_export_timeout?

      return re_enqueue if export_empty? || export_started?

      if file_extraction_pipeline? && export_status.batched?
        log_extra_metadata_on_done(:batched, true)

        pipeline_tracker.update!(status_event: 'start', jid: jid, batched: true)

        return pipeline_tracker.finish! if export_status.batches_count < 1

        enqueue_limited_batches
        re_enqueue unless all_batches_enqueued?
      else
        log_extra_metadata_on_done(:batched, false)

        pipeline_tracker.update!(status_event: 'start', jid: jid)
        pipeline_tracker.pipeline_class.new(context).run
        pipeline_tracker.finish!
      end
    rescue BulkImports::RetryPipelineError => e
      retry_tracker(e)
    end

    def fail_pipeline(exception)
      pipeline_tracker.update!(status_event: 'fail_op', jid: jid)

      entity.fail_op! if pipeline_tracker.abort_on_failure?

      log_exception(exception, log_attributes(message: 'Pipeline failed'))

      Gitlab::ErrorTracking.track_exception(exception, log_attributes)

      BulkImports::Failure.create(
        bulk_import_entity_id: entity.id,
        pipeline_class: pipeline_tracker.pipeline_name,
        pipeline_step: 'pipeline_worker_run',
        exception_class: exception.class.to_s,
        exception_message: exception.message,
        correlation_id_value: Labkit::Correlation::CorrelationId.current_or_new_id
      )
    end

    def logger
      @logger ||= Logger.build.with_tracker(pipeline_tracker)
    end

    def re_enqueue(delay = FILE_EXTRACTION_PIPELINE_PERFORM_DELAY)
      log_extra_metadata_on_done(:re_enqueue, true)

      with_context(bulk_import_entity_id: entity.id) do
        self.class.perform_in(
          delay,
          pipeline_tracker.id,
          pipeline_tracker.stage,
          entity.id
        )
      end
    end

    def context
      @context ||= ::BulkImports::Pipeline::Context.new(pipeline_tracker)
    end

    def export_status
      @export_status ||= ExportStatus.new(pipeline_tracker, pipeline_tracker.pipeline_class.relation)
    end

    def file_extraction_pipeline?
      pipeline_tracker.file_extraction_pipeline?
    end

    def empty_export_timeout?
      export_empty? && time_since_tracker_created > Pipeline::EMPTY_EXPORT_STATUS_TIMEOUT
    end

    def export_failed?
      return false unless file_extraction_pipeline?

      export_status.failed?
    end

    def export_started?
      return false unless file_extraction_pipeline?

      export_status.started?
    end

    def export_empty?
      return false unless file_extraction_pipeline?

      export_status.empty?
    end

    def retry_tracker(exception)
      log_exception(exception, log_attributes(message: "Retrying pipeline"))

      pipeline_tracker.update!(status_event: 'retry', jid: jid)

      re_enqueue(exception.retry_delay)
    end

    def skip_tracker
      logger.info(log_attributes(message: 'Skipping pipeline due to failed entity'))

      pipeline_tracker.update!(status_event: 'skip', jid: jid)
    end

    def log_attributes(extra = {})
      logger.default_attributes.merge(extra)
    end

    def log_exception(exception, payload)
      Gitlab::ExceptionLogFormatter.format!(exception, payload)

      logger.error(structured_payload(payload))
    end

    def time_since_tracker_created
      Time.zone.now - (pipeline_tracker.created_at || entity.created_at)
    end

    def enqueue_limited_batches
      next_batch.numbers.each do |batch_number|
        batch = pipeline_tracker.batches.create!(batch_number: batch_number)

        with_context(bulk_import_entity_id: entity.id) do
          ::BulkImports::PipelineBatchWorker.perform_async(batch.id)
        end
      end

      log_extra_metadata_on_done(:tracker_batch_numbers_enqueued, next_batch.numbers)
      log_extra_metadata_on_done(:tracker_final_batch_was_enqueued, next_batch.final?)
    end

    def all_batches_enqueued?
      next_batch.final?
    end

    def next_batch
      all_batch_numbers = (1..export_status.batches_count).to_a

      created_batch_numbers = pipeline_tracker.batches.pluck_batch_numbers

      remaining_batch_numbers = all_batch_numbers - created_batch_numbers

      limit = next_batch_count

      LimitedBatches.new(
        numbers: remaining_batch_numbers.first(limit),
        final?: remaining_batch_numbers.count <= limit
      )
    end
    strong_memoize_attr :next_batch

    # Calculate the number of batches, up to `batch_limit`, to process in the
    # next round.
    def next_batch_count
      limit = batch_limit - pipeline_tracker.batches.in_progress.limit(batch_limit).count
      [limit, 0].max
    end

    def batch_limit
      ::Gitlab::CurrentSettings.bulk_import_concurrent_pipeline_batch_limit
    end

    def lease_timeout
      30
    end

    def lease_key
      "gitlab:bulk_imports:pipeline_worker:#{pipeline_tracker.id}"
    end
  end
end