segment_loader_test.go 14.3 KB
Newer Older
1 2 3 4 5 6
// Licensed to the LF AI & Data foundation under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
7 8
// with the License. You may obtain a copy of the License at
//
9
//     http://www.apache.org/licenses/LICENSE-2.0
10
//
11 12 13 14 15
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
16 17 18 19 20

package querynode

import (
	"context"
B
bigsheeper 已提交
21
	"math/rand"
22
	"runtime"
23 24
	"testing"

B
bigsheeper 已提交
25 26 27
	"github.com/stretchr/testify/assert"

	"github.com/milvus-io/milvus/internal/proto/commonpb"
28
	"github.com/milvus-io/milvus/internal/proto/querypb"
29
	"github.com/milvus-io/milvus/internal/proto/schemapb"
30
	"github.com/milvus-io/milvus/internal/util/funcutil"
31 32
)

B
bigsheeper 已提交
33 34 35 36
func TestSegmentLoader_loadSegment(t *testing.T) {
	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()

37
	schema := genSimpleInsertDataSchema()
B
bigsheeper 已提交
38 39 40 41

	fieldBinlog, err := saveSimpleBinLog(ctx)
	assert.NoError(t, err)

42
	t.Run("test load segment", func(t *testing.T) {
43
		node, err := genSimpleQueryNode(ctx)
44 45
		assert.NoError(t, err)

46
		err = node.historical.replica.removeSegment(defaultSegmentID)
47
		assert.NoError(t, err)
48 49

		loader := node.loader
50 51 52 53 54 55 56
		assert.NotNil(t, loader)

		req := &querypb.LoadSegmentsRequest{
			Base: &commonpb.MsgBase{
				MsgType: commonpb.MsgType_WatchQueryChannels,
				MsgID:   rand.Int63(),
			},
X
xige-16 已提交
57 58
			DstNodeID: 0,
			Schema:    schema,
59 60 61 62 63 64 65 66 67 68
			Infos: []*querypb.SegmentLoadInfo{
				{
					SegmentID:    defaultSegmentID,
					PartitionID:  defaultPartitionID,
					CollectionID: defaultCollectionID,
					BinlogPaths:  fieldBinlog,
				},
			},
		}

69
		err = loader.loadSegment(req, segmentTypeSealed)
70 71
		assert.NoError(t, err)
	})
72

73 74
	t.Run("test set segment error due to without partition", func(t *testing.T) {
		node, err := genSimpleQueryNode(ctx)
75 76
		assert.NoError(t, err)

77
		err = node.historical.replica.removePartition(defaultPartitionID)
78
		assert.NoError(t, err)
79 80

		loader := node.loader
81 82 83 84 85 86 87
		assert.NotNil(t, loader)

		req := &querypb.LoadSegmentsRequest{
			Base: &commonpb.MsgBase{
				MsgType: commonpb.MsgType_WatchQueryChannels,
				MsgID:   rand.Int63(),
			},
X
xige-16 已提交
88 89
			DstNodeID: 0,
			Schema:    schema,
90 91 92 93 94 95 96 97 98 99
			Infos: []*querypb.SegmentLoadInfo{
				{
					SegmentID:    defaultSegmentID,
					PartitionID:  defaultPartitionID,
					CollectionID: defaultCollectionID,
					BinlogPaths:  fieldBinlog,
				},
			},
		}

100
		err = loader.loadSegment(req, segmentTypeSealed)
101 102
		assert.Error(t, err)
	})
103 104 105 106 107 108 109 110 111 112 113 114 115

	t.Run("test load segment with nil base message", func(t *testing.T) {
		node, err := genSimpleQueryNode(ctx)
		assert.NoError(t, err)

		loader := node.loader
		assert.NotNil(t, loader)

		req := &querypb.LoadSegmentsRequest{}

		err = loader.loadSegment(req, segmentTypeSealed)
		assert.Error(t, err)
	})
B
bigsheeper 已提交
116 117
}

118 119 120 121 122
func TestSegmentLoader_loadSegmentFieldsData(t *testing.T) {
	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()

	runLoadSegmentFieldData := func(dataType schemapb.DataType) {
123
		node, err := genSimpleQueryNode(ctx)
124
		assert.NoError(t, err)
125 126
		loader := node.loader
		assert.NotNil(t, loader)
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

		fieldUID := genConstantField(uidField)
		fieldTimestamp := genConstantField(timestampField)

		schema := &schemapb.CollectionSchema{
			Name:   defaultCollectionName,
			AutoID: true,
			Fields: []*schemapb.FieldSchema{},
		}

		constFieldID := FieldID(105)
		constFieldName := "const-field-test"
		constField := &schemapb.FieldSchema{
			FieldID: constFieldID,
			Name:    constFieldName,
		}

		switch dataType {
		case schemapb.DataType_Bool:
			constField.DataType = schemapb.DataType_Bool
		case schemapb.DataType_Int8:
			constField.DataType = schemapb.DataType_Int8
		case schemapb.DataType_Int16:
			constField.DataType = schemapb.DataType_Int16
		case schemapb.DataType_Int32:
			constField.DataType = schemapb.DataType_Int32
		case schemapb.DataType_Int64:
			constField.DataType = schemapb.DataType_Int64
		case schemapb.DataType_Float:
			constField.DataType = schemapb.DataType_Float
		case schemapb.DataType_Double:
			constField.DataType = schemapb.DataType_Double
		case schemapb.DataType_FloatVector:
			constField.DataType = schemapb.DataType_FloatVector
		case schemapb.DataType_BinaryVector:
			constField.DataType = schemapb.DataType_BinaryVector
		}

		schema.Fields = append(schema.Fields, constField)

167
		err = loader.historicalReplica.removeSegment(defaultSegmentID)
168 169 170 171
		assert.NoError(t, err)

		col := newCollection(defaultCollectionID, schema)
		assert.NotNil(t, col)
172
		segment, err := newSegment(col,
173 174 175
			defaultSegmentID,
			defaultPartitionID,
			defaultCollectionID,
176
			defaultDMLChannel,
177 178
			segmentTypeSealed,
			true)
179
		assert.Nil(t, err)
180 181 182 183 184 185 186

		schema.Fields = append(schema.Fields, fieldUID)
		schema.Fields = append(schema.Fields, fieldTimestamp)

		binlog, err := saveBinLog(ctx, defaultCollectionID, defaultPartitionID, defaultSegmentID, defaultMsgLength, schema)
		assert.NoError(t, err)

187
		err = loader.loadFiledBinlogData(segment, binlog)
188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206
		assert.NoError(t, err)
	}

	t.Run("test bool", func(t *testing.T) {
		runLoadSegmentFieldData(schemapb.DataType_Bool)
		runLoadSegmentFieldData(schemapb.DataType_Int8)
		runLoadSegmentFieldData(schemapb.DataType_Int16)
		runLoadSegmentFieldData(schemapb.DataType_Int32)
		runLoadSegmentFieldData(schemapb.DataType_Int64)
		runLoadSegmentFieldData(schemapb.DataType_Float)
		runLoadSegmentFieldData(schemapb.DataType_Double)
	})
}

func TestSegmentLoader_invalid(t *testing.T) {
	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()

	t.Run("test no collection", func(t *testing.T) {
207
		node, err := genSimpleQueryNode(ctx)
208
		assert.NoError(t, err)
209 210
		loader := node.loader
		assert.NotNil(t, loader)
211

212
		err = node.historical.replica.removeCollection(defaultCollectionID)
213 214 215 216 217 218 219
		assert.NoError(t, err)

		req := &querypb.LoadSegmentsRequest{
			Base: &commonpb.MsgBase{
				MsgType: commonpb.MsgType_WatchQueryChannels,
				MsgID:   rand.Int63(),
			},
X
xige-16 已提交
220
			DstNodeID: 0,
221 222 223 224 225 226 227 228 229
			Infos: []*querypb.SegmentLoadInfo{
				{
					SegmentID:    defaultSegmentID,
					PartitionID:  defaultPartitionID,
					CollectionID: defaultCollectionID,
				},
			},
		}

230
		err = loader.loadSegment(req, segmentTypeSealed)
231 232 233
		assert.Error(t, err)
	})

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
	//t.Run("test no collection 2", func(t *testing.T) {
	//	historical, err := genSimpleHistorical(ctx)
	//	assert.NoError(t, err)
	//
	//	err = historical.replica.removeCollection(defaultCollectionID)
	//	assert.NoError(t, err)
	//
	//	err = historical.loader.loadSegmentInternal(defaultCollectionID, nil, nil)
	//	assert.Error(t, err)
	//})
	//
	//t.Run("test no vec field", func(t *testing.T) {
	//	historical, err := genSimpleHistorical(ctx)
	//	assert.NoError(t, err)
	//
	//	err = historical.replica.removeCollection(defaultCollectionID)
	//	assert.NoError(t, err)
	//
	//	schema := &schemapb.CollectionSchema{
	//		Name:   defaultCollectionName,
	//		AutoID: true,
	//		Fields: []*schemapb.FieldSchema{
	//			genConstantField(constFieldParam{
	//				id:       FieldID(100),
	//				dataType: schemapb.DataType_Int8,
	//			}),
	//		},
	//	}
	//	err = historical.loader.historicalReplica.addCollection(defaultCollectionID, schema)
	//	assert.NoError(t, err)
	//
	//	err = historical.loader.loadSegmentInternal(defaultCollectionID, nil, nil)
	//	assert.Error(t, err)
	//})
268 269

	t.Run("test no vec field 2", func(t *testing.T) {
270
		node, err := genSimpleQueryNode(ctx)
271
		assert.NoError(t, err)
272 273
		loader := node.loader
		assert.NotNil(t, loader)
274

275
		err = node.historical.replica.removeCollection(defaultCollectionID)
276 277 278 279 280 281 282 283 284 285 286 287
		assert.NoError(t, err)

		schema := &schemapb.CollectionSchema{
			Name:   defaultCollectionName,
			AutoID: true,
			Fields: []*schemapb.FieldSchema{
				genConstantField(constFieldParam{
					id:       FieldID(100),
					dataType: schemapb.DataType_Int8,
				}),
			},
		}
288
		loader.historicalReplica.addCollection(defaultCollectionID, schema)
289 290 291 292 293 294

		req := &querypb.LoadSegmentsRequest{
			Base: &commonpb.MsgBase{
				MsgType: commonpb.MsgType_WatchQueryChannels,
				MsgID:   rand.Int63(),
			},
X
xige-16 已提交
295 296
			DstNodeID: 0,
			Schema:    schema,
297 298 299 300 301 302 303 304
			Infos: []*querypb.SegmentLoadInfo{
				{
					SegmentID:    defaultSegmentID,
					PartitionID:  defaultPartitionID,
					CollectionID: defaultCollectionID,
				},
			},
		}
305
		err = loader.loadSegment(req, segmentTypeSealed)
306 307 308
		assert.Error(t, err)
	})
}
309 310 311 312 313

func TestSegmentLoader_checkSegmentSize(t *testing.T) {
	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()

314
	node, err := genSimpleQueryNode(ctx)
315
	assert.NoError(t, err)
316 317
	loader := node.loader
	assert.NotNil(t, loader)
318

319
	err = loader.checkSegmentSize(defaultSegmentID, []*querypb.SegmentLoadInfo{{SegmentID: defaultSegmentID, SegmentSize: 1024}}, runtime.GOMAXPROCS(0))
320 321 322 323 324 325 326 327
	assert.NoError(t, err)

	//totalMem, err := getTotalMemory()
	//assert.NoError(t, err)
	//err = historical.loader.checkSegmentSize(defaultSegmentID, map[UniqueID]int64{defaultSegmentID: int64(totalMem * 2)})
	//assert.Error(t, err)
}

328 329 330 331 332 333 334 335 336 337 338 339 340
func TestSegmentLoader_testLoadGrowing(t *testing.T) {
	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()

	t.Run("test load growing segments", func(t *testing.T) {
		node, err := genSimpleQueryNode(ctx)
		assert.NoError(t, err)
		loader := node.loader
		assert.NotNil(t, loader)

		collection, err := node.historical.replica.getCollectionByID(defaultCollectionID)
		assert.NoError(t, err)

341 342
		segment, err := newSegment(collection, defaultSegmentID+1, defaultPartitionID, defaultCollectionID, defaultDMLChannel, segmentTypeGrowing, true)
		assert.Nil(t, err)
343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359

		insertMsg, err := genSimpleInsertMsg()
		assert.NoError(t, err)

		err = loader.loadGrowingSegments(segment, insertMsg.RowIDs, insertMsg.Timestamps, insertMsg.RowData)
		assert.NoError(t, err)
	})

	t.Run("test invalid insert data", func(t *testing.T) {
		node, err := genSimpleQueryNode(ctx)
		assert.NoError(t, err)
		loader := node.loader
		assert.NotNil(t, loader)

		collection, err := node.historical.replica.getCollectionByID(defaultCollectionID)
		assert.NoError(t, err)

360 361
		segment, err := newSegment(collection, defaultSegmentID+1, defaultPartitionID, defaultCollectionID, defaultDMLChannel, segmentTypeGrowing, true)
		assert.Nil(t, err)
362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393

		insertMsg, err := genSimpleInsertMsg()
		assert.NoError(t, err)

		insertMsg.RowData = nil

		err = loader.loadGrowingSegments(segment, insertMsg.RowIDs, insertMsg.Timestamps, insertMsg.RowData)
		assert.Error(t, err)
	})
}

func TestSegmentLoader_testLoadGrowingAndSealed(t *testing.T) {
	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()

	schema := genSimpleInsertDataSchema()
	fieldBinlog, err := saveBinLog(ctx, defaultCollectionID, defaultPartitionID, defaultSegmentID, defaultMsgLength, schema)
	assert.NoError(t, err)

	t.Run("test load growing and sealed segments", func(t *testing.T) {
		node, err := genSimpleQueryNode(ctx)
		assert.NoError(t, err)

		loader := node.loader
		assert.NotNil(t, loader)

		segmentID1 := UniqueID(100)
		req1 := &querypb.LoadSegmentsRequest{
			Base: &commonpb.MsgBase{
				MsgType: commonpb.MsgType_WatchQueryChannels,
				MsgID:   rand.Int63(),
			},
X
xige-16 已提交
394 395
			DstNodeID: 0,
			Schema:    schema,
396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414
			Infos: []*querypb.SegmentLoadInfo{
				{
					SegmentID:    segmentID1,
					PartitionID:  defaultPartitionID,
					CollectionID: defaultCollectionID,
					BinlogPaths:  fieldBinlog,
				},
			},
		}

		err = loader.loadSegment(req1, segmentTypeSealed)
		assert.NoError(t, err)

		segmentID2 := UniqueID(101)
		req2 := &querypb.LoadSegmentsRequest{
			Base: &commonpb.MsgBase{
				MsgType: commonpb.MsgType_WatchQueryChannels,
				MsgID:   rand.Int63(),
			},
X
xige-16 已提交
415 416
			DstNodeID: 0,
			Schema:    schema,
417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438
			Infos: []*querypb.SegmentLoadInfo{
				{
					SegmentID:    segmentID2,
					PartitionID:  defaultPartitionID,
					CollectionID: defaultCollectionID,
					BinlogPaths:  fieldBinlog,
				},
			},
		}

		err = loader.loadSegment(req2, segmentTypeGrowing)
		assert.NoError(t, err)

		segment1, err := loader.historicalReplica.getSegmentByID(segmentID1)
		assert.NoError(t, err)

		segment2, err := loader.streamingReplica.getSegmentByID(segmentID2)
		assert.NoError(t, err)

		assert.Equal(t, segment1.getRowCount(), segment2.getRowCount())
	})
}
439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511

func TestSegmentLoader_testLoadSealedSegmentWithIndex(t *testing.T) {
	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()
	//generate schema
	fieldUID := genConstantField(uidField)
	fieldTimestamp := genConstantField(timestampField)
	fieldVec := genFloatVectorField(simpleVecField)
	fieldInt := genConstantField(simpleConstField)

	schema := &schemapb.CollectionSchema{ // schema for insertData
		Name:   defaultCollectionName,
		AutoID: true,
		Fields: []*schemapb.FieldSchema{
			fieldUID,
			fieldTimestamp,
			fieldVec,
			fieldInt,
		},
	}

	// generate insert binlog
	fieldBinlog, err := saveBinLog(ctx, defaultCollectionID, defaultPartitionID, defaultSegmentID, defaultMsgLength, schema)
	assert.NoError(t, err)

	segmentID := UniqueID(100)
	// generate index file for segment
	indexPaths, err := generateIndex(segmentID)
	assert.NoError(t, err)
	indexInfo := &querypb.VecFieldIndexInfo{
		FieldID:        simpleVecField.id,
		EnableIndex:    true,
		IndexName:      indexName,
		IndexID:        indexID,
		BuildID:        buildID,
		IndexParams:    funcutil.Map2KeyValuePair(genSimpleIndexParams()),
		IndexFilePaths: indexPaths,
	}

	// generate segmentLoader
	node, err := genSimpleQueryNode(ctx)
	assert.NoError(t, err)
	loader := node.loader
	assert.NotNil(t, loader)

	req := &querypb.LoadSegmentsRequest{
		Base: &commonpb.MsgBase{
			MsgType: commonpb.MsgType_WatchQueryChannels,
			MsgID:   rand.Int63(),
		},
		DstNodeID: 0,
		Schema:    schema,
		Infos: []*querypb.SegmentLoadInfo{
			{
				SegmentID:    segmentID,
				PartitionID:  defaultPartitionID,
				CollectionID: defaultCollectionID,
				BinlogPaths:  fieldBinlog,
				IndexInfos:   []*querypb.VecFieldIndexInfo{indexInfo},
			},
		},
	}

	err = loader.loadSegment(req, segmentTypeSealed)
	assert.NoError(t, err)

	segment, err := node.historical.replica.getSegmentByID(segmentID)
	assert.NoError(t, err)
	vecFieldInfo, err := segment.getVectorFieldInfo(simpleVecField.id)
	assert.NoError(t, err)
	assert.NotNil(t, vecFieldInfo)
	assert.Equal(t, true, vecFieldInfo.indexInfo.EnableIndex)
}