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
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
|
# frozen_string_literal: true
RSpec.shared_examples 'it runs batched background migration jobs' do |tracking_database|
include ExclusiveLeaseHelpers
describe 'defining the job attributes' do
it 'defines the data_consistency as always' do
expect(described_class.get_data_consistency).to eq(:always)
end
it 'defines the feature_category as database' do
expect(described_class.get_feature_category).to eq(:database)
end
it 'defines the idempotency as true' do
expect(described_class.idempotent?).to be_truthy
end
end
describe '.tracking_database' do
it 'does not raise an error' do
expect { described_class.tracking_database }.not_to raise_error
end
it 'overrides the method to return the tracking database' do
expect(described_class.tracking_database).to eq(tracking_database)
end
end
describe '.lease_key' do
let(:lease_key) { described_class.name.demodulize.underscore }
it 'does not raise an error' do
expect { described_class.lease_key }.not_to raise_error
end
it 'returns the lease key' do
expect(described_class.lease_key).to eq(lease_key)
end
end
describe '.enabled?' do
it 'returns true when execute_batched_migrations_on_schedule feature flag is enabled' do
stub_feature_flags(execute_batched_migrations_on_schedule: true)
expect(described_class.enabled?).to be_truthy
end
it 'returns false when execute_batched_migrations_on_schedule feature flag is disabled' do
stub_feature_flags(execute_batched_migrations_on_schedule: false)
expect(described_class.enabled?).to be_falsey
end
end
describe '#perform' do
subject(:worker) { described_class.new }
context 'when the base model does not exist' do
before do
if Gitlab::Database.has_config?(tracking_database)
skip "because the base model for #{tracking_database} exists"
end
end
it 'does nothing' do
expect(worker).not_to receive(:active_migration)
expect(worker).not_to receive(:run_active_migration)
expect { worker.perform }.not_to raise_error
end
it 'logs a message indicating execution is skipped' do
expect(Sidekiq.logger).to receive(:info) do |payload|
expect(payload[:class]).to eq(described_class.name)
expect(payload[:database]).to eq(tracking_database)
expect(payload[:message]).to match(/skipping migration execution/)
end
expect { worker.perform }.not_to raise_error
end
end
context 'when the base model does exist' do
before do
unless Gitlab::Database.has_config?(tracking_database)
skip "because the base model for #{tracking_database} does not exist"
end
end
context 'when the feature flag is disabled' do
before do
stub_feature_flags(execute_batched_migrations_on_schedule: false)
end
it 'does nothing' do
expect(worker).not_to receive(:active_migration)
expect(worker).not_to receive(:run_active_migration)
worker.perform
end
end
context 'when the feature flag is enabled' do
let(:base_model) { Gitlab::Database.database_base_models[tracking_database] }
before do
stub_feature_flags(execute_batched_migrations_on_schedule: true)
allow(Gitlab::Database::BackgroundMigration::BatchedMigration).to receive(:active_migration)
.with(connection: base_model.connection)
.and_return(nil)
end
context 'when database config is shared' do
it 'does nothing' do
expect(Gitlab::Database).to receive(:db_config_share_with)
.with(base_model.connection_db_config).and_return('main')
expect(worker).not_to receive(:active_migration)
expect(worker).not_to receive(:run_active_migration)
worker.perform
end
end
context 'when no active migrations exist' do
it 'does nothing' do
expect(worker).not_to receive(:run_active_migration)
worker.perform
end
end
context 'when active migrations exist' do
let(:job_interval) { 5.minutes }
let(:lease_timeout) { 15.minutes }
let(:lease_key) { described_class.name.demodulize.underscore }
let(:migration) { build(:batched_background_migration, :active, interval: job_interval) }
let(:interval_variance) { described_class::INTERVAL_VARIANCE }
before do
allow(Gitlab::Database::BackgroundMigration::BatchedMigration).to receive(:active_migration)
.with(connection: base_model.connection)
.and_return(migration)
allow(migration).to receive(:interval_elapsed?).with(variance: interval_variance).and_return(true)
allow(migration).to receive(:reload)
end
context 'when the reloaded migration is no longer active' do
it 'does not run the migration' do
expect_to_obtain_exclusive_lease(lease_key, timeout: lease_timeout)
expect(migration).to receive(:reload)
expect(migration).to receive(:active?).and_return(false)
expect(worker).not_to receive(:run_active_migration)
worker.perform
end
end
context 'when the interval has not elapsed' do
it 'does not run the migration' do
expect_to_obtain_exclusive_lease(lease_key, timeout: lease_timeout)
expect(migration).to receive(:interval_elapsed?).with(variance: interval_variance).and_return(false)
expect(worker).not_to receive(:run_active_migration)
worker.perform
end
end
context 'when the reloaded migration is still active and the interval has elapsed' do
it 'runs the migration' do
expect_to_obtain_exclusive_lease(lease_key, timeout: lease_timeout)
expect_next_instance_of(Gitlab::Database::BackgroundMigration::BatchedMigrationRunner) do |instance|
expect(instance).to receive(:run_migration_job).with(migration)
end
expect(worker).to receive(:run_active_migration).and_call_original
worker.perform
end
end
context 'when the calculated timeout is less than the minimum allowed' do
let(:minimum_timeout) { described_class::MINIMUM_LEASE_TIMEOUT }
let(:job_interval) { 2.minutes }
it 'sets the lease timeout to the minimum value' do
expect_to_obtain_exclusive_lease(lease_key, timeout: minimum_timeout)
expect_next_instance_of(Gitlab::Database::BackgroundMigration::BatchedMigrationRunner) do |instance|
expect(instance).to receive(:run_migration_job).with(migration)
end
expect(worker).to receive(:run_active_migration).and_call_original
worker.perform
end
end
it 'always cleans up the exclusive lease' do
lease = stub_exclusive_lease_taken(lease_key, timeout: lease_timeout)
expect(lease).to receive(:try_obtain).and_return(true)
expect(worker).to receive(:run_active_migration).and_raise(RuntimeError, 'I broke')
expect(lease).to receive(:cancel)
expect { worker.perform }.to raise_error(RuntimeError, 'I broke')
end
it 'receives the correct connection' do
base_model = Gitlab::Database.database_base_models[tracking_database]
expect(Gitlab::Database::SharedModel).to receive(:using_connection).with(base_model.connection).and_yield
worker.perform
end
end
end
end
end
describe 'executing an entire migration', :freeze_time, if: Gitlab::Database.has_config?(tracking_database) do
include Gitlab::Database::DynamicModelHelpers
include Database::DatabaseHelpers
let(:migration_class) do
Class.new(Gitlab::BackgroundMigration::BatchedMigrationJob) do
def perform(matching_status)
each_sub_batch(
operation_name: :update_all,
batching_scope: -> (relation) { relation.where(status: matching_status) }
) do |sub_batch|
sub_batch.update_all(some_column: 0)
end
end
end
end
let(:gitlab_schema) { "gitlab_#{tracking_database}" }
let!(:migration) do
create(
:batched_background_migration,
:active,
table_name: table_name,
column_name: :id,
max_value: migration_records,
batch_size: batch_size,
sub_batch_size: sub_batch_size,
job_class_name: 'ExampleDataMigration',
job_arguments: [1],
gitlab_schema: gitlab_schema
)
end
let(:base_model) { Gitlab::Database.database_base_models[tracking_database] }
let(:table_name) { 'example_data' }
let(:batch_size) { 5 }
let(:sub_batch_size) { 2 }
let(:number_of_batches) { 10 }
let(:migration_records) { batch_size * number_of_batches }
let(:connection) { Gitlab::Database.database_base_models[tracking_database].connection }
let(:example_data) { define_batchable_model(table_name, connection: connection) }
around do |example|
Gitlab::Database::SharedModel.using_connection(connection) do
example.run
end
end
before do
# Create example table populated with test data to migrate.
#
# Test data should have two records that won't be updated:
# - one record beyond the migration's range
# - one record that doesn't match the migration job's batch condition
connection.execute(<<~SQL)
CREATE TABLE #{table_name} (
id integer primary key,
some_column integer,
status smallint);
INSERT INTO #{table_name} (id, some_column, status)
SELECT generate_series, generate_series, 1
FROM generate_series(1, #{migration_records + 1});
UPDATE #{table_name}
SET status = 0
WHERE some_column = #{migration_records - 5};
SQL
stub_feature_flags(execute_batched_migrations_on_schedule: true)
stub_const('Gitlab::BackgroundMigration::ExampleDataMigration', migration_class)
end
subject(:full_migration_run) do
# process all batches, then do an extra execution to mark the job as finished
(number_of_batches + 1).times do
described_class.new.perform
travel_to((migration.interval + described_class::INTERVAL_VARIANCE).seconds.from_now)
end
end
it 'marks the migration record as finished' do
expect { full_migration_run }.to change { migration.reload.status }.from(1).to(3) # active -> finished
end
it 'creates job records for each processed batch', :aggregate_failures do
expect { full_migration_run }.to change { migration.reload.batched_jobs.count }.from(0)
final_min_value = migration.batched_jobs.reduce(1) do |next_min_value, batched_job|
expect(batched_job.min_value).to eq(next_min_value)
batched_job.max_value + 1
end
final_max_value = final_min_value - 1
expect(final_max_value).to eq(migration_records)
end
it 'marks all job records as succeeded', :aggregate_failures do
expect { full_migration_run }.to change { migration.reload.batched_jobs.count }.from(0)
expect(migration.batched_jobs).to all(be_succeeded)
end
it 'updates matching records in the range', :aggregate_failures do
expect { full_migration_run }
.to change { example_data.where('status = 1 AND some_column <> 0').count }
.from(migration_records).to(1)
record_outside_range = example_data.last
expect(record_outside_range.status).to eq(1)
expect(record_outside_range.some_column).not_to eq(0)
end
it 'does not update non-matching records in the range' do
expect { full_migration_run }.not_to change { example_data.where('status <> 1 AND some_column <> 0').count }
end
context 'health status' do
subject(:migration_run) { described_class.new.perform }
it 'puts migration on hold when there is autovaccum activity on related tables' do
swapout_view_for_table(:postgres_autovacuum_activity, connection: connection)
create(
:postgres_autovacuum_activity,
table: migration.table_name,
table_identifier: "public.#{migration.table_name}"
)
expect { migration_run }.to change { migration.reload.on_hold? }.from(false).to(true)
end
end
end
end
|