FilesExpand file tree

 issue4592

/

test_checkpoint_sync.py

Copy path

BlameMore file actions

BlameMore file actions

Latest commit

 

History

History

History

320 lines (320 loc) · 10.8 KB

 issue4592

/

test_checkpoint_sync.py

Top

File metadata and controls

• Code
• Blame

320 lines (320 loc) · 10.8 KB

Raw

Copy raw file

Download raw file

Open symbols panel

Edit and raw actions

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

# from tests.shared_test_pipeline import SharedTestPipeline

# from tests import enterprise_only

# from feldera.runtime_config import RuntimeConfig, Storage

# from feldera.enums import PipelineStatus, FaultToleranceModel

# from typing import Optional

# import os

# import sys

# import time

# from uuid import uuid4

# import random

# import pytest

#

#

# DEFAULT_ENDPOINT = os.environ.get(

# "DEFAULT_MINIO_ENDPOINT", "http://minio.extra.svc.cluster.local:9000"

# )

# DEFAULT_BUCKET = "default"

# ACCESS_KEY = "minio"

# SECRET_KEY = "miniopasswd"

#

#

# def storage_cfg(

# pipeline_name: str,

# endpoint: Optional[str] = None,

# start_from_checkpoint: Optional[str] = None,

# strict: bool = False,

# auth_err: bool = False,

# standby: bool = False,

# pull_interval: int = 2,

# ) -> dict:

# return {

# "backend": {

# "name": "file",

# "config": {

# "sync": {

# "bucket": f"{DEFAULT_BUCKET}/{pipeline_name}",

# "access_key": ACCESS_KEY,

# "secret_key": SECRET_KEY if not auth_err else SECRET_KEY + "extra",

# "provider": "Minio",

# "endpoint": endpoint or DEFAULT_ENDPOINT,

# "start_from_checkpoint": start_from_checkpoint,

# "fail_if_no_checkpoint": strict,

# "standby": standby,

# "pull_interval": pull_interval,

# }

# },

# }

# }

#

#

# class TestCheckpointSync(SharedTestPipeline):

# @enterprise_only

# def test_checkpoint_sync(

# self,

# from_uuid: bool = False,

# random_uuid: bool = False,

# clear_storage: bool = True,

# auth_err: bool = False,

# strict: bool = False,

# expect_empty: bool = False,

# standby: bool = False,

# ):

# """

# CREATE TABLE t0 (c0 INT, c1 VARCHAR);

# CREATE MATERIALIZED VIEW v0 AS SELECT * FROM t0;

# """

#

# storage_config = storage_cfg(self.pipeline.name)

# ft = FaultToleranceModel.AtLeastOnce

#

# self.pipeline.set_runtime_config(

# RuntimeConfig(

# fault_tolerance_model=ft, storage=Storage(config=storage_config)

# )

# )

# self.pipeline.start()

#

# random.seed(time.time())

# total = random.randint(10, 20)

# data = [{"c0": i, "c1": str(i)} for i in range(1, total)]

# self.pipeline.input_json("t0", data)

# self.pipeline.execute("INSERT INTO t0 VALUES (21, 'exists')")

#

# start = time.time()

# timeout = 5

#

# while True:

# processed = self.pipeline.stats().global_metrics.total_processed_records

# if processed == total:

# break

#

# if time.time() - start > timeout:

# raise TimeoutError(

# f"timed out while waiting for pipeline to process {total} records"

# )

#

# time.sleep(0.1)

#

# got_before = list(self.pipeline.query("SELECT * FROM v0"))

# print(f"{self.pipeline.name}: records: {total}, {got_before}", file=sys.stderr)

#

# if len(got_before) != processed:

# raise RuntimeError(

# f"adhoc query returned {len(got_before)} but {processed} records were processed: {got_before}"

# )

#

# self.pipeline.checkpoint(wait=True)

# uuid = self.pipeline.sync_checkpoint(wait=True)

#

# self.pipeline.stop(force=True)

#

# if clear_storage:

# self.pipeline.clear_storage()

#

# if random_uuid:

# uuid = uuid4()

#

# # Restart pipeline from checkpoint

# storage_config = storage_cfg(

# pipeline_name=self.pipeline.name,

# start_from_checkpoint=uuid if from_uuid else "latest",

# auth_err=auth_err,

# strict=strict,

# standby=standby,

# )

# self.pipeline.set_runtime_config(

# RuntimeConfig(

# fault_tolerance_model=ft, storage=Storage(config=storage_config)

# )

# )

#

# if not standby:

# self.pipeline.start()

# else:

# self.pipeline.start(wait=False)

#

# # wait for the pipeline to initialize

# start = time.monotonic()

# # wait for a maximum of 120 seconds for the pipeline to provison

# end = start + 120

#

# # wait for the pipeline to finish provisoning

# for log in self.pipeline.logs():

# if "checkpoint pulled successfully" in log:

# break

#

# if time.monotonic() > end:

# raise TimeoutError(

# f"{self.pipeline.name} timedout waiting to pull checkpoint"

# )

#

# if standby:

# # wait for 8 seconds, this should be more than enough time

# time.sleep(8)

# assert self.pipeline.status() == PipelineStatus.INITIALIZING

#

# self.pipeline.activate(timeout_s=10)

#

# got_after = list(self.pipeline.query("SELECT * FROM v0"))

#

# print(

# f"{self.pipeline.name}: after: {len(got_after)}, {got_after}",

# file=sys.stderr,

# )

#

# if expect_empty:

# got_before = []

#

# self.assertCountEqual(got_before, got_after)

#

# self.pipeline.stop(force=True)

#

# if clear_storage:

# self.pipeline.clear_storage()

#

# @enterprise_only

# def test_from_uuid(self):

# self.test_checkpoint_sync(from_uuid=True)

#

# @enterprise_only

# def test_without_clearing_storage(self):

# self.test_checkpoint_sync(clear_storage=False)

#

# @enterprise_only

# def test_autherr_fail(self):

# with self.assertRaisesRegex(RuntimeError, "SignatureDoesNotMatch"):

# self.test_checkpoint_sync(auth_err=True, strict=True)

#

# @enterprise_only

# def test_autherr(self):

# self.test_checkpoint_sync(auth_err=True, strict=False, expect_empty=True)

#

# @enterprise_only

# def test_nonexistent_checkpoint_fail(self):

# with self.assertRaisesRegex(RuntimeError, "were not found in source"):

# self.test_checkpoint_sync(random_uuid=True, from_uuid=True, strict=True)

#

# @enterprise_only

# def test_nonexistent_checkpoint(self):

# self.test_checkpoint_sync(random_uuid=True, from_uuid=True, expect_empty=True)

#

# @enterprise_only

# def test_standby_activation(self):

# self.test_checkpoint_sync(standby=True)

#

# @enterprise_only

# def test_standby_activation_from_uuid(self):

# self.test_checkpoint_sync(standby=True, from_uuid=True)

#

# @enterprise_only

# def test_standby_fallback(self, from_uuid: bool = False):

# # Step 1: Start main pipeline

# storage_config = storage_cfg(self.pipeline.name)

# ft = FaultToleranceModel.AtLeastOnce

# self.pipeline.set_runtime_config(

# RuntimeConfig(

# fault_tolerance_model=ft, storage=Storage(config=storage_config)

# )

# )

# self.pipeline.start()

#

# # Insert initial data

# random.seed(time.time())

# total_initial = random.randint(10, 20)

# data_initial = [{"c0": i, "c1": str(i)} for i in range(1, total_initial)]

# self.pipeline.input_json("t0", data_initial)

# self.pipeline.execute("INSERT INTO t0 VALUES (21, 'exists')")

# self.pipeline.wait_for_completion()

#

# got_before = list(self.pipeline.query("select * from v0"))

#

# # Step 2: Create checkpoint and sync

# self.pipeline.checkpoint(wait=True)

# uuid = self.pipeline.sync_checkpoint(wait=True)

#

# # Step 3: Start standby pipeline

# standby = self.new_pipeline_with_suffix("standby")

# pull_interval = 1

# standby.set_runtime_config(

# RuntimeConfig(

# fault_tolerance_model=ft,

# storage=Storage(

# config=storage_cfg(

# self.pipeline.name,

# start_from_checkpoint=uuid if from_uuid else "latest",

# standby=True,

# pull_interval=pull_interval,

# )

# ),

# )

# )

# standby.start(wait=False)

#

# # Wait until standby pulls the first checkpoint

# start = time.monotonic()

# end = start + 120

# for log in standby.logs():

# if "checkpoint pulled successfully" in log:

# break

# if time.monotonic() > end:

# raise TimeoutError(

# "Timed out waiting for standby pipeline to pull checkpoint"

# )

#

# # Step 4: Add more data and make 3-10 checkpoints

# extra_ckpts = random.randint(3, 10)

# total_additional = 0

#

# for i in range(extra_ckpts):

# new_val = 100 + i

# new_data = [{"c0": new_val, "c1": f"extra_{new_val}"}]

# self.pipeline.input_json("t0", new_data)

# self.pipeline.wait_for_completion()

# total_additional += 1

# self.pipeline.checkpoint(wait=True)

# self.pipeline.sync_checkpoint(wait=True)

# time.sleep(0.2)

#

# got_expected = (

# got_before if from_uuid else list(self.pipeline.query("SELECT * FROM v0"))

# )

# print(

# f"{self.pipeline.name}: final records before shutdown: {got_expected}",

# file=sys.stderr,

# )

#

# # Step 5: Stop main and activate standby

# self.pipeline.stop(force=True)

#

# assert standby.status() == PipelineStatus.INITIALIZING

# standby.activate(timeout_s=(pull_interval * extra_ckpts) + 60)

#

# for log in standby.logs():

# if "activated" in log:

# break

# if time.monotonic() > end:

# raise TimeoutError("Timed out waiting for standby pipeline to activate")

#

# # Step 6: Validate standby has all expected records

# got_after = list(standby.query("SELECT * FROM v0"))

# print(

# f"{standby.name}: final records after activation: {got_after}",

# file=sys.stderr,

# )

# self.assertCountEqual(got_expected, got_after)

#

# # Cleanup

# standby.stop(force=True)

#

# standby.start()

# got_final = list(standby.query("SELECT * FROM v0"))

# standby.stop(force=True)

#

# self.assertCountEqual(got_after, got_final)

#

# self.pipeline.clear_storage()

#

# @enterprise_only

# def test_standby_fallback_from_uuid(self):

# self.test_standby_fallback(from_uuid=True)