-
Notifications
You must be signed in to change notification settings - Fork 178
/
engine.go
700 lines (592 loc) · 25.8 KB
/
engine.go
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
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
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
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
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
package fetcher
import (
"context"
"fmt"
"github.com/rs/zerolog"
"go.opentelemetry.io/otel/attribute"
"github.com/onflow/flow-go/engine"
"github.com/onflow/flow-go/model/chunks"
"github.com/onflow/flow-go/model/flow"
"github.com/onflow/flow-go/model/flow/filter"
"github.com/onflow/flow-go/model/verification"
"github.com/onflow/flow-go/module"
"github.com/onflow/flow-go/module/mempool"
"github.com/onflow/flow-go/module/trace"
"github.com/onflow/flow-go/network"
"github.com/onflow/flow-go/state/protocol"
"github.com/onflow/flow-go/storage"
"github.com/onflow/flow-go/utils/logging"
)
// Engine implements the fetcher engine functionality. It works between a chunk consumer queue, and a verifier engine.
// Its input is an assigned chunk locator from the chunk consumer it is subscribed to.
//
// Its output is a verifiable chunk that it passes to the verifier engine.
//
// Fetcher engine is an AssignedChunkProcessor implementation: it receives assigned chunks to this verification node from the chunk consumer.
// The assigned chunks are passed on concurrent executions of its ProcessAssignedChunk method.
//
// On receiving an assigned chunk, the engine requests their chunk data pack through the requester that is attached to it.
// On receiving a chunk data pack response, the fetcher engine validates it, and shapes a verifiable chunk out of it, and passes it
// to the verifier engine.
type Engine struct {
// common
unit *engine.Unit
state protocol.State // used to verify the origin ID of chunk data response, and sealing status.
// monitoring
log zerolog.Logger
tracer module.Tracer
metrics module.VerificationMetrics
// memory and storage
pendingChunks mempool.ChunkStatuses // stores all pending chunks that their chunk data is requested from requester.
blocks storage.Blocks // used to for verifying collection ID.
headers storage.Headers // used for building verifiable chunk data.
results storage.ExecutionResults // used to retrieve execution result of an assigned chunk.
receipts storage.ExecutionReceipts // used to find executor ids of a chunk, for requesting chunk data pack.
// output interfaces
verifier network.Engine // used to push verifiable chunk down the verification pipeline.
requester ChunkDataPackRequester // used to request chunk data packs from network.
chunkConsumerNotifier module.ProcessingNotifier // used to notify chunk consumer that it is done processing a chunk.
stopAtHeight uint64
}
func New(
log zerolog.Logger,
metrics module.VerificationMetrics,
tracer module.Tracer,
verifier network.Engine,
state protocol.State,
pendingChunks mempool.ChunkStatuses,
headers storage.Headers,
blocks storage.Blocks,
results storage.ExecutionResults,
receipts storage.ExecutionReceipts,
requester ChunkDataPackRequester,
stopAtHeight uint64,
) *Engine {
e := &Engine{
unit: engine.NewUnit(),
metrics: metrics,
tracer: tracer,
log: log.With().Str("engine", "fetcher").Logger(),
verifier: verifier,
state: state,
pendingChunks: pendingChunks,
blocks: blocks,
headers: headers,
results: results,
receipts: receipts,
requester: requester,
stopAtHeight: stopAtHeight,
}
e.requester.WithChunkDataPackHandler(e)
return e
}
// WithChunkConsumerNotifier sets the processing notifier of fetcher.
// The fetcher engine uses this notifier to inform the chunk consumer that it is done processing a given chunk, and
// is ready to receive a new chunk to process.
func (e *Engine) WithChunkConsumerNotifier(notifier module.ProcessingNotifier) {
e.chunkConsumerNotifier = notifier
}
// Ready initializes the engine and returns a channel that is closed when the initialization is done
func (e *Engine) Ready() <-chan struct{} {
if e.chunkConsumerNotifier == nil {
e.log.Fatal().Msg("missing chunk consumer notifier callback in verification fetcher engine")
}
return e.unit.Ready(func() {
<-e.requester.Ready()
})
}
// Done terminates the engine and returns a channel that is closed when the termination is done
func (e *Engine) Done() <-chan struct{} {
return e.unit.Done(func() {
<-e.requester.Done()
})
}
// ProcessAssignedChunk is the entry point of fetcher engine.
// It pushes the assigned chunk down the pipeline.
// Through the pipeline the chunk data pack for this chunk is requested,
// a verifiable chunk is shaped for it,
// and is pushed to the verifier engine for verification.
//
// It should not be blocking since multiple chunk consumer workers might be calling it concurrently.
// It fetches the chunk data pack, once received, verifier engine will be verifying
// Once a chunk has been processed, it will call the processing notifier callback to notify
// the chunk consumer in order to process the next chunk.
func (e *Engine) ProcessAssignedChunk(locator *chunks.Locator) {
locatorID := locator.ID()
lg := e.log.With().
Hex("locator_id", logging.ID(locatorID)).
Hex("result_id", logging.ID(locator.ResultID)).
Uint64("chunk_index", locator.Index).
Logger()
e.metrics.OnAssignedChunkReceivedAtFetcher()
// retrieves result and chunk using the locator
result, err := e.results.ByID(locator.ResultID)
if err != nil {
// a missing result for a chunk locator is a fatal error potentially a database leak.
lg.Fatal().Err(err).Msg("could not retrieve result for chunk locator")
}
chunk := result.Chunks[locator.Index]
chunkID := chunk.ID()
lg = lg.With().
Hex("chunk_id", logging.ID(chunkID)).
Hex("block_id", logging.ID(chunk.ChunkBody.BlockID)).
Logger()
lg.Debug().Msg("result and chunk for locator retrieved")
requested, blockHeight, err := e.processAssignedChunkWithTracing(chunk, result, locatorID)
lg = lg.With().Uint64("block_height", blockHeight).Logger()
if err != nil {
lg.Fatal().Err(err).Msg("could not process assigned chunk")
}
lg.Info().Bool("requested", requested).Msg("assigned chunk processed successfully")
if requested {
e.metrics.OnChunkDataPackRequestSentByFetcher()
}
}
// processAssignedChunkWithTracing encapsulates the logic of processing assigned chunk with tracing enabled.
func (e *Engine) processAssignedChunkWithTracing(chunk *flow.Chunk, result *flow.ExecutionResult, chunkLocatorID flow.Identifier) (bool, uint64, error) {
span, _ := e.tracer.StartBlockSpan(e.unit.Ctx(), result.BlockID, trace.VERProcessAssignedChunk)
span.SetAttributes(attribute.Int("collection_index", int(chunk.CollectionIndex)))
defer span.End()
requested, blockHeight, err := e.processAssignedChunk(chunk, result, chunkLocatorID)
return requested, blockHeight, err
}
// processAssignedChunk receives an assigned chunk and its result and requests its chunk data pack from requester.
// Boolean return value determines whether chunk data pack was requested or not.
func (e *Engine) processAssignedChunk(chunk *flow.Chunk, result *flow.ExecutionResult, chunkLocatorID flow.Identifier) (bool, uint64, error) {
// skips processing a chunk if it belongs to a sealed block.
chunkID := chunk.ID()
sealed, blockHeight, err := e.blockIsSealed(chunk.ChunkBody.BlockID)
if err != nil {
return false, 0, fmt.Errorf("could not determine whether block has been sealed: %w", err)
}
if sealed {
e.chunkConsumerNotifier.Notify(chunkLocatorID) // tells consumer that we are done with this chunk.
return false, blockHeight, nil
}
// skip chunk if it verifies a block at or above stop height
if e.stopAtHeight > 0 && blockHeight >= e.stopAtHeight {
e.log.Warn().Msgf("Skipping chunk %s - height %d at or above stop height requested (%d)", chunkID, blockHeight, e.stopAtHeight)
e.chunkConsumerNotifier.Notify(chunkLocatorID) // tells consumer that we are done with this chunk.
return false, blockHeight, nil
}
// adds chunk status as a pending chunk to mempool.
status := &verification.ChunkStatus{
ChunkIndex: chunk.Index,
ExecutionResult: result,
BlockHeight: blockHeight,
}
added := e.pendingChunks.Add(status)
if !added {
return false, blockHeight, nil
}
err = e.requestChunkDataPack(chunk.Index, chunkID, result.ID(), chunk.BlockID)
if err != nil {
return false, blockHeight, fmt.Errorf("could not request chunk data pack: %w", err)
}
// requesting a chunk data pack is async, i.e., once engine reaches this point
// it gracefully waits (unblocking) for the requested
// till it either delivers us the requested chunk data pack
// or cancels our request (when chunk belongs to a sealed block).
//
// both these events happen through requester module calling fetchers callbacks.
// it is during those callbacks that we notify the consumer that we are done with this job.
return true, blockHeight, nil
}
// HandleChunkDataPack is called by the chunk requester module everytime a new requested chunk data pack arrives.
// The chunks are supposed to be deduplicated by the requester.
// So invocation of this method indicates arrival of a distinct requested chunk.
func (e *Engine) HandleChunkDataPack(originID flow.Identifier, response *verification.ChunkDataPackResponse) {
lg := e.log.With().
Hex("origin_id", logging.ID(originID)).
Hex("chunk_id", logging.ID(response.Cdp.ChunkID)).
Logger()
if response.Cdp.Collection != nil {
// non-system chunk data packs have non-nil collection
lg = lg.With().
Hex("collection_id", logging.ID(response.Cdp.Collection.ID())).
Logger()
lg.Info().Msg("chunk data pack arrived")
} else {
lg.Info().Msg("system chunk data pack arrived")
}
e.metrics.OnChunkDataPackArrivedAtFetcher()
// make sure we still need it
status, exists := e.pendingChunks.Get(response.Index, response.ResultID)
if !exists {
lg.Debug().Msg("could not fetch pending status from mempool, dropping chunk data")
return
}
resultID := status.ExecutionResult.ID()
lg = lg.With().
Hex("block_id", logging.ID(status.ExecutionResult.BlockID)).
Uint64("block_height", status.BlockHeight).
Hex("result_id", logging.ID(resultID)).
Uint64("chunk_index", status.ChunkIndex).
Bool("system_chunk", IsSystemChunk(status.ChunkIndex, status.ExecutionResult)).
Logger()
span, ctx := e.tracer.StartBlockSpan(context.Background(), status.ExecutionResult.BlockID, trace.VERFetcherHandleChunkDataPack)
defer span.End()
processed, err := e.handleChunkDataPackWithTracing(ctx, originID, status, response.Cdp)
if IsChunkDataPackValidationError(err) {
lg.Error().Err(err).Msg("could not validate chunk data pack")
return
}
if err != nil {
// TODO: byzantine fault
lg.Fatal().Err(err).Msg("could not handle chunk data pack")
return
}
if processed {
e.metrics.OnVerifiableChunkSentToVerifier()
// we need to report that the job has been finished eventually
e.chunkConsumerNotifier.Notify(status.ChunkLocatorID())
lg.Info().Msg("verifiable chunk pushed to verifier engine")
}
}
// handleChunkDataPackWithTracing encapsulates the logic of handling chunk data pack with tracing enabled.
//
// Boolean returned value determines whether the chunk data pack passed validation and its verifiable chunk
// submitted to verifier.
// The first returned value determines non-critical errors (i.e., expected ones).
// The last returned value determines the critical errors that are unexpected, and should lead program to halt.
func (e *Engine) handleChunkDataPackWithTracing(
ctx context.Context,
originID flow.Identifier,
status *verification.ChunkStatus,
chunkDataPack *flow.ChunkDataPack) (bool, error) {
// make sure the chunk data pack is valid
err := e.validateChunkDataPackWithTracing(ctx, status.ChunkIndex, originID, chunkDataPack, status.ExecutionResult)
if err != nil {
return false, NewChunkDataPackValidationError(originID,
status.ExecutionResult.ID(),
status.ChunkIndex,
chunkDataPack.ID(),
chunkDataPack.ChunkID,
chunkDataPack.Collection.ID(),
err)
}
processed, err := e.handleValidatedChunkDataPack(ctx, status, chunkDataPack)
if err != nil {
return processed, fmt.Errorf("could not handle validated chunk data pack: %w", err)
}
return processed, nil
}
// handleValidatedChunkDataPack receives a validated chunk data pack, removes its status from the memory, and pushes a verifiable chunk for it to
// verifier engine.
// Boolean return value determines whether verifiable chunk pushed to verifier or not.
func (e *Engine) handleValidatedChunkDataPack(ctx context.Context,
status *verification.ChunkStatus,
chunkDataPack *flow.ChunkDataPack) (bool, error) {
removed := e.pendingChunks.Remove(status.ChunkIndex, status.ExecutionResult.ID())
if !removed {
// we deduplicate the chunk data responses at this point, reaching here means a
// duplicate chunk data response is under process concurrently, so we give up
// on processing current one.
return false, nil
}
// pushes chunk data pack to verifier, and waits for it to be verified.
chunk := status.ExecutionResult.Chunks[status.ChunkIndex]
err := e.pushToVerifierWithTracing(ctx, chunk, status.ExecutionResult, chunkDataPack)
if err != nil {
return false, fmt.Errorf("could not push the chunk to verifier engine")
}
return true, nil
}
// validateChunkDataPackWithTracing encapsulates the logic of validating a chunk data pack with tracing enabled.
func (e *Engine) validateChunkDataPackWithTracing(ctx context.Context,
chunkIndex uint64,
senderID flow.Identifier,
chunkDataPack *flow.ChunkDataPack,
result *flow.ExecutionResult) error {
var err error
e.tracer.WithSpanFromContext(ctx, trace.VERFetcherValidateChunkDataPack, func() {
err = e.validateChunkDataPack(chunkIndex, senderID, chunkDataPack, result)
})
return err
}
// validateChunkDataPack validates the integrity of a received chunk data pack as well as the authenticity of its sender.
// Regarding the integrity: the chunk data pack should have a matching start state with the chunk itself, as well as a matching collection ID with the
// given collection.
//
// Regarding the authenticity: the chunk data pack should be coming from a sender that is an authorized execution node at the block of the chunk.
func (e *Engine) validateChunkDataPack(chunkIndex uint64,
senderID flow.Identifier,
chunkDataPack *flow.ChunkDataPack,
result *flow.ExecutionResult) error {
chunk := result.Chunks[chunkIndex]
// 1. chunk ID of chunk data pack should map the chunk ID on execution result
expectedChunkID := chunk.ID()
if chunkDataPack.ChunkID != expectedChunkID {
return fmt.Errorf("chunk ID of chunk data pack does not match corresponding chunk on execution result, expected: %x, got:%x",
expectedChunkID, chunkDataPack.ChunkID)
}
// 2. sender must be a authorized execution node at that block
blockID := chunk.BlockID
authorized := e.validateAuthorizedExecutionNodeAtBlockID(senderID, blockID)
if !authorized {
return fmt.Errorf("unauthorized execution node sender at block ID: %x, resultID: %x, chunk ID: %x",
blockID,
result.ID(),
chunk.ID())
}
// 3. start state must match
if chunkDataPack.StartState != chunk.ChunkBody.StartState {
return engine.NewInvalidInputErrorf("expecting chunk data pack's start state: %x, but got: %x, block ID: %x, resultID: %x, chunk ID: %x",
chunk.ChunkBody.StartState,
chunkDataPack.StartState,
blockID,
result.ID(),
chunk.ID())
}
// 3. collection id must match
err := e.validateCollectionID(chunkDataPack, result, chunk)
if err != nil {
return fmt.Errorf("could not validate collection: %w", err)
}
return nil
}
// validateCollectionID returns error for an invalid collection of a chunk data pack,
// and returns nil otherwise.
func (e Engine) validateCollectionID(
chunkDataPack *flow.ChunkDataPack,
result *flow.ExecutionResult,
chunk *flow.Chunk) error {
if IsSystemChunk(chunk.Index, result) {
return e.validateSystemChunkCollection(chunkDataPack)
}
return e.validateNonSystemChunkCollection(chunkDataPack, chunk)
}
// validateSystemChunkCollection returns nil if the system chunk data pack has a nil collection.
func (e Engine) validateSystemChunkCollection(chunkDataPack *flow.ChunkDataPack) error {
// collection of a system chunk should be nil
if chunkDataPack.Collection != nil {
return engine.NewInvalidInputErrorf("non-nil collection for system chunk, collection ID: %v, len: %d",
chunkDataPack.Collection.ID(), chunkDataPack.Collection.Len())
}
return nil
}
// validateNonSystemChunkCollection returns nil if the collection is matching the non-system chunk data pack.
// A collection is valid against a non-system chunk if it has a matching ID with the
// collection ID of corresponding guarantee of the chunk in the referenced block payload.
func (e Engine) validateNonSystemChunkCollection(chunkDataPack *flow.ChunkDataPack, chunk *flow.Chunk) error {
collID := chunkDataPack.Collection.ID()
block, err := e.blocks.ByID(chunk.BlockID)
if err != nil {
return fmt.Errorf("could not get block: %w", err)
}
if block.Payload.Guarantees[chunk.Index].CollectionID != collID {
return engine.NewInvalidInputErrorf("mismatch collection id with guarantee, expected: %v, got: %v",
block.Payload.Guarantees[chunk.Index].CollectionID,
collID)
}
return nil
}
// validateAuthorizedExecutionNodeAtBlockID validates sender ID of a chunk data pack response as an authorized
// execution node at the given block ID.
func (e Engine) validateAuthorizedExecutionNodeAtBlockID(senderID flow.Identifier, blockID flow.Identifier) bool {
snapshot := e.state.AtBlockID(blockID)
valid, err := protocol.IsNodeAuthorizedWithRoleAt(snapshot, senderID, flow.RoleExecution)
if err != nil {
e.log.Fatal().
Err(err).
Hex("block_id", logging.ID(blockID)).
Hex("sender_id", logging.ID(senderID)).
Msg("could not validate sender identity at specified block ID snapshot as execution node")
}
return valid
}
// NotifyChunkDataPackSealed is called by the ChunkDataPackRequester to notify the ChunkDataPackHandler that the specified chunk
// has been sealed and hence the requester will no longer request it.
//
// When the requester calls this callback method, it will never return a chunk data pack for this specified chunk to the handler (i.e.,
// through HandleChunkDataPack).
func (e *Engine) NotifyChunkDataPackSealed(chunkIndex uint64, resultID flow.Identifier) {
lg := e.log.With().
Uint64("chunk_index", chunkIndex).
Hex("result_id", logging.ID(resultID)).
Logger()
// we need to report that the job has been finished eventually
status, exists := e.pendingChunks.Get(chunkIndex, resultID)
if !exists {
lg.Debug().
Msg("could not fetch pending status for sealed chunk from mempool, dropping chunk data")
return
}
chunkLocatorID := status.ChunkLocatorID()
lg = lg.With().
Uint64("block_height", status.BlockHeight).
Hex("result_id", logging.ID(status.ExecutionResult.ID())).Logger()
removed := e.pendingChunks.Remove(chunkIndex, resultID)
e.chunkConsumerNotifier.Notify(chunkLocatorID)
lg.Info().
Bool("removed", removed).
Msg("discards fetching chunk of an already sealed block and notified consumer")
}
// pushToVerifierWithTracing encapsulates the logic of pushing a verifiable chunk to verifier engine with tracing enabled.
func (e *Engine) pushToVerifierWithTracing(
ctx context.Context,
chunk *flow.Chunk,
result *flow.ExecutionResult,
chunkDataPack *flow.ChunkDataPack) error {
var err error
e.tracer.WithSpanFromContext(ctx, trace.VERFetcherPushToVerifier, func() {
err = e.pushToVerifier(chunk, result, chunkDataPack)
})
return err
}
// pushToVerifier makes a verifiable chunk data out of the input and pass it to the verifier for verification.
//
// When this method returns without any error, it means that the verification of the chunk at the verifier engine is done (either successfully,
// or unsuccessfully)
func (e *Engine) pushToVerifier(chunk *flow.Chunk,
result *flow.ExecutionResult,
chunkDataPack *flow.ChunkDataPack) error {
header, err := e.headers.ByBlockID(chunk.BlockID)
if err != nil {
return fmt.Errorf("could not get block: %w", err)
}
vchunk, err := e.makeVerifiableChunkData(chunk, header, result, chunkDataPack)
if err != nil {
return fmt.Errorf("could not verify chunk: %w", err)
}
err = e.verifier.ProcessLocal(vchunk)
if err != nil {
return fmt.Errorf("verifier could not verify chunk: %w", err)
}
return nil
}
// makeVerifiableChunkData creates and returns a verifiable chunk data for the chunk data.
// The verifier engine, which is the last engine in the pipeline of verification, uses this verifiable
// chunk data to verify it.
func (e *Engine) makeVerifiableChunkData(chunk *flow.Chunk,
header *flow.Header,
result *flow.ExecutionResult,
chunkDataPack *flow.ChunkDataPack,
) (*verification.VerifiableChunkData, error) {
// system chunk is the last chunk
isSystemChunk := IsSystemChunk(chunk.Index, result)
endState, err := EndStateCommitment(result, chunk.Index, isSystemChunk)
if err != nil {
return nil, fmt.Errorf("could not compute end state of chunk: %w", err)
}
transactionOffset, err := TransactionOffsetForChunk(result.Chunks, chunk.Index)
if err != nil {
return nil, fmt.Errorf("cannot compute transaction offset for chunk: %w", err)
}
return &verification.VerifiableChunkData{
IsSystemChunk: isSystemChunk,
Chunk: chunk,
Header: header,
Result: result,
ChunkDataPack: chunkDataPack,
EndState: endState,
TransactionOffset: transactionOffset,
}, nil
}
// requestChunkDataPack creates and dispatches a chunk data pack request to the requester engine.
func (e *Engine) requestChunkDataPack(chunkIndex uint64, chunkID flow.Identifier, resultID flow.Identifier, blockID flow.Identifier) error {
agrees, disagrees, err := e.getAgreeAndDisagreeExecutors(blockID, resultID)
if err != nil {
return fmt.Errorf("could not segregate the agree and disagree executors for result: %x of block: %x", resultID, blockID)
}
header, err := e.headers.ByBlockID(blockID)
if err != nil {
return fmt.Errorf("could not get header for block: %x", blockID)
}
allExecutors, err := e.state.AtBlockID(blockID).Identities(filter.HasRole(flow.RoleExecution))
if err != nil {
return fmt.Errorf("could not fetch execution node ids at block %x: %w", blockID, err)
}
request := &verification.ChunkDataPackRequest{
Locator: chunks.Locator{
ResultID: resultID,
Index: chunkIndex,
},
ChunkDataPackRequestInfo: verification.ChunkDataPackRequestInfo{
ChunkID: chunkID,
Height: header.Height,
Agrees: agrees,
Disagrees: disagrees,
Targets: allExecutors,
},
}
e.requester.Request(request)
return nil
}
// getAgreeAndDisagreeExecutors segregates the execution nodes identifiers based on the given execution result id at the given block into agree and
// disagree sets.
// The agree set contains the executors who made receipt with the same result as the given result id.
// The disagree set contains the executors who made receipt with different result than the given result id.
func (e *Engine) getAgreeAndDisagreeExecutors(blockID flow.Identifier, resultID flow.Identifier) (flow.IdentifierList, flow.IdentifierList, error) {
receipts, err := e.receipts.ByBlockID(blockID)
if err != nil {
return nil, nil, fmt.Errorf("could not retrieve receipts for block: %v: %w", blockID, err)
}
agrees, disagrees := executorsOf(receipts, resultID)
return agrees, disagrees, nil
}
// blockIsSealed returns true if the block at specified height by block ID is sealed.
func (e Engine) blockIsSealed(blockID flow.Identifier) (bool, uint64, error) {
// TODO: as an optimization, we can keep record of last sealed height on a local variable.
header, err := e.headers.ByBlockID(blockID)
if err != nil {
return false, 0, fmt.Errorf("could not get block: %w", err)
}
lastSealed, err := e.state.Sealed().Head()
if err != nil {
return false, 0, fmt.Errorf("could not get last sealed: %w", err)
}
sealed := header.Height <= lastSealed.Height
return sealed, header.Height, nil
}
// executorsOf segregates the executors of the given receipts based on the given execution result id.
// The agree set contains the executors who made receipt with the same result as the given result id.
// The disagree set contains the executors who made receipt with different result than the given result id.
func executorsOf(receipts []*flow.ExecutionReceipt, resultID flow.Identifier) (flow.IdentifierList, flow.IdentifierList) {
var agrees flow.IdentifierList
var disagrees flow.IdentifierList
for _, receipt := range receipts {
executor := receipt.ExecutorID
if receipt.ExecutionResult.ID() == resultID {
agrees = append(agrees, executor)
} else {
disagrees = append(disagrees, executor)
}
}
return agrees, disagrees
}
// EndStateCommitment computes the end state of the given chunk.
func EndStateCommitment(result *flow.ExecutionResult, chunkIndex uint64, systemChunk bool) (flow.StateCommitment, error) {
var endState flow.StateCommitment
if systemChunk {
var err error
// last chunk in a result is the system chunk and takes final state commitment
endState, err = result.FinalStateCommitment()
if err != nil {
return flow.DummyStateCommitment, fmt.Errorf("can not read final state commitment, likely a bug:%w", err)
}
} else {
// any chunk except last takes the subsequent chunk's start state
endState = result.Chunks[chunkIndex+1].StartState
}
return endState, nil
}
// TransactionOffsetForChunk calculates transaction offset for a given chunk which is the index of the first
// transaction of this chunk within the whole block
func TransactionOffsetForChunk(chunks flow.ChunkList, chunkIndex uint64) (uint32, error) {
if int(chunkIndex) > len(chunks)-1 {
return 0, fmt.Errorf("chunk list out of bounds, len %d asked for chunk %d", len(chunks), chunkIndex)
}
var offset uint32 = 0
for i := 0; i < int(chunkIndex); i++ {
offset += uint32(chunks[i].NumberOfTransactions)
}
return offset, nil
}
// IsSystemChunk returns true if `chunkIndex` points to a system chunk in `result`.
// Otherwise, it returns false.
// In the current version, a chunk is a system chunk if it is the last chunk of the
// execution result.
func IsSystemChunk(chunkIndex uint64, result *flow.ExecutionResult) bool {
return chunkIndex == uint64(len(result.Chunks)-1)
}