Browse Source

swarm: Fix T.Fatal inside a goroutine in tests (#18409)

* swarm/storage: fix T.Fatal inside a goroutine

* swarm/network/simulation: fix T.Fatal inside a goroutine

* swarm/network/stream: fix T.Fatal inside a goroutine

* swarm/network/simulation: consistent failures in TestPeerEventsTimeout

* swarm/network/simulation: rename sendRunSignal to triggerSimulationRun
Janoš Guljaš 6 years ago
parent
commit
d70c4faf20

+ 5 - 2
swarm/network/simulation/events_test.go

@@ -81,6 +81,7 @@ func TestPeerEventsTimeout(t *testing.T) {
 	events := sim.PeerEvents(ctx, sim.NodeIDs())
 
 	done := make(chan struct{})
+	errC := make(chan error)
 	go func() {
 		for e := range events {
 			if e.Error == context.Canceled {
@@ -90,14 +91,16 @@ func TestPeerEventsTimeout(t *testing.T) {
 				close(done)
 				return
 			} else {
-				t.Fatal(e.Error)
+				errC <- e.Error
 			}
 		}
 	}()
 
 	select {
 	case <-time.After(time.Second):
-		t.Error("no context deadline received")
+		t.Fatal("no context deadline received")
+	case err := <-errC:
+		t.Fatal(err)
 	case <-done:
 		// all good, context deadline detected
 	}

+ 11 - 10
swarm/network/simulation/http_test.go

@@ -73,7 +73,8 @@ func TestSimulationWithHTTPServer(t *testing.T) {
 	//this time the timeout should be long enough so that it doesn't kick in too early
 	ctx, cancel2 := context.WithTimeout(context.Background(), 5*time.Second)
 	defer cancel2()
-	go sendRunSignal(t)
+	errC := make(chan error, 1)
+	go triggerSimulationRun(t, errC)
 	result = sim.Run(ctx, func(ctx context.Context, sim *Simulation) error {
 		log.Debug("This run waits for the run signal from `frontend`...")
 		//ensure with a Sleep that simulation doesn't terminate before the signal is received
@@ -83,10 +84,13 @@ func TestSimulationWithHTTPServer(t *testing.T) {
 	if result.Error != nil {
 		t.Fatal(result.Error)
 	}
+	if err := <-errC; err != nil {
+		t.Fatal(err)
+	}
 	log.Debug("Test terminated successfully")
 }
 
-func sendRunSignal(t *testing.T) {
+func triggerSimulationRun(t *testing.T, errC chan error) {
 	//We need to first wait for the sim HTTP server to start running...
 	time.Sleep(2 * time.Second)
 	//then we can send the signal
@@ -94,16 +98,13 @@ func sendRunSignal(t *testing.T) {
 	log.Debug("Sending run signal to simulation: POST /runsim...")
 	resp, err := http.Post(fmt.Sprintf("http://localhost%s/runsim", DefaultHTTPSimAddr), "application/json", nil)
 	if err != nil {
-		t.Fatalf("Request failed: %v", err)
+		errC <- fmt.Errorf("Request failed: %v", err)
+		return
 	}
-	defer func() {
-		err := resp.Body.Close()
-		if err != nil {
-			log.Error("Error closing response body", "err", err)
-		}
-	}()
 	log.Debug("Signal sent")
 	if resp.StatusCode != http.StatusOK {
-		t.Fatalf("err %s", resp.Status)
+		errC <- fmt.Errorf("err %s", resp.Status)
+		return
 	}
+	errC <- resp.Body.Close()
 }

+ 27 - 7
swarm/network/stream/delivery_test.go

@@ -19,9 +19,11 @@ package stream
 import (
 	"bytes"
 	"context"
+	"errors"
 	"fmt"
 	"os"
 	"sync"
+	"sync/atomic"
 	"testing"
 	"time"
 
@@ -500,7 +502,7 @@ func testDeliveryFromNodes(t *testing.T, nodes, chunkCount int, skipCheck bool)
 
 	log.Info("Starting simulation")
 	ctx := context.Background()
-	result := sim.Run(ctx, func(ctx context.Context, sim *simulation.Simulation) error {
+	result := sim.Run(ctx, func(ctx context.Context, sim *simulation.Simulation) (err error) {
 		nodeIDs := sim.UpNodeIDs()
 		//determine the pivot node to be the first node of the simulation
 		pivot := nodeIDs[0]
@@ -553,14 +555,13 @@ func testDeliveryFromNodes(t *testing.T, nodes, chunkCount int, skipCheck bool)
 		}
 		pivotFileStore := item.(*storage.FileStore)
 		log.Debug("Starting retrieval routine")
+		retErrC := make(chan error)
 		go func() {
 			// start the retrieval on the pivot node - this will spawn retrieve requests for missing chunks
 			// we must wait for the peer connections to have started before requesting
 			n, err := readAll(pivotFileStore, fileHash)
 			log.Info(fmt.Sprintf("retrieved %v", fileHash), "read", n, "err", err)
-			if err != nil {
-				t.Fatalf("requesting chunks action error: %v", err)
-			}
+			retErrC <- err
 		}()
 
 		log.Debug("Watching for disconnections")
@@ -570,11 +571,19 @@ func testDeliveryFromNodes(t *testing.T, nodes, chunkCount int, skipCheck bool)
 			simulation.NewPeerEventsFilter().Drop(),
 		)
 
+		var disconnected atomic.Value
 		go func() {
 			for d := range disconnections {
 				if d.Error != nil {
 					log.Error("peer drop", "node", d.NodeID, "peer", d.PeerID)
-					t.Fatal(d.Error)
+					disconnected.Store(true)
+				}
+			}
+		}()
+		defer func() {
+			if err != nil {
+				if yes, ok := disconnected.Load().(bool); ok && yes {
+					err = errors.New("disconnect events received")
 				}
 			}
 		}()
@@ -595,6 +604,9 @@ func testDeliveryFromNodes(t *testing.T, nodes, chunkCount int, skipCheck bool)
 		if !success {
 			return fmt.Errorf("Test failed, chunks not available on all nodes")
 		}
+		if err := <-retErrC; err != nil {
+			t.Fatalf("requesting chunks: %v", err)
+		}
 		log.Debug("Test terminated successfully")
 		return nil
 	})
@@ -675,7 +687,7 @@ func benchmarkDeliveryFromNodes(b *testing.B, nodes, chunkCount int, skipCheck b
 	}
 
 	ctx := context.Background()
-	result := sim.Run(ctx, func(ctx context.Context, sim *simulation.Simulation) error {
+	result := sim.Run(ctx, func(ctx context.Context, sim *simulation.Simulation) (err error) {
 		nodeIDs := sim.UpNodeIDs()
 		node := nodeIDs[len(nodeIDs)-1]
 
@@ -702,11 +714,19 @@ func benchmarkDeliveryFromNodes(b *testing.B, nodes, chunkCount int, skipCheck b
 			simulation.NewPeerEventsFilter().Drop(),
 		)
 
+		var disconnected atomic.Value
 		go func() {
 			for d := range disconnections {
 				if d.Error != nil {
 					log.Error("peer drop", "node", d.NodeID, "peer", d.PeerID)
-					b.Fatal(d.Error)
+					disconnected.Store(true)
+				}
+			}
+		}()
+		defer func() {
+			if err != nil {
+				if yes, ok := disconnected.Load().(bool); ok && yes {
+					err = errors.New("disconnect events received")
 				}
 			}
 		}()

+ 12 - 2
swarm/network/stream/intervals_test.go

@@ -19,9 +19,11 @@ package stream
 import (
 	"context"
 	"encoding/binary"
+	"errors"
 	"fmt"
 	"os"
 	"sync"
+	"sync/atomic"
 	"testing"
 	"time"
 
@@ -117,7 +119,7 @@ func testIntervals(t *testing.T, live bool, history *Range, skipCheck bool) {
 		t.Fatal(err)
 	}
 
-	result := sim.Run(ctx, func(ctx context.Context, sim *simulation.Simulation) error {
+	result := sim.Run(ctx, func(ctx context.Context, sim *simulation.Simulation) (err error) {
 		nodeIDs := sim.UpNodeIDs()
 		storer := nodeIDs[0]
 		checker := nodeIDs[1]
@@ -162,11 +164,19 @@ func testIntervals(t *testing.T, live bool, history *Range, skipCheck bool) {
 			return err
 		}
 
+		var disconnected atomic.Value
 		go func() {
 			for d := range disconnections {
 				if d.Error != nil {
 					log.Error("peer drop", "node", d.NodeID, "peer", d.PeerID)
-					t.Fatal(d.Error)
+					disconnected.Store(true)
+				}
+			}
+		}()
+		defer func() {
+			if err != nil {
+				if yes, ok := disconnected.Load().(bool); ok && yes {
+					err = errors.New("disconnect events received")
 				}
 			}
 		}()

+ 17 - 6
swarm/network/stream/snapshot_sync_test.go

@@ -21,6 +21,7 @@ import (
 	"os"
 	"runtime"
 	"sync"
+	"sync/atomic"
 	"testing"
 	"time"
 
@@ -213,11 +214,13 @@ func testSyncingViaGlobalSync(t *testing.T, chunkCount int, nodeCount int) {
 		simulation.NewPeerEventsFilter().Drop(),
 	)
 
+	var disconnected atomic.Value
 	go func() {
 		for d := range disconnections {
-			log.Error("peer drop", "node", d.NodeID, "peer", d.PeerID)
-			t.Fatal("unexpected disconnect")
-			cancelSimRun()
+			if d.Error != nil {
+				log.Error("peer drop", "node", d.NodeID, "peer", d.PeerID)
+				disconnected.Store(true)
+			}
 		}
 	}()
 
@@ -226,6 +229,9 @@ func testSyncingViaGlobalSync(t *testing.T, chunkCount int, nodeCount int) {
 	if result.Error != nil {
 		t.Fatal(result.Error)
 	}
+	if yes, ok := disconnected.Load().(bool); ok && yes {
+		t.Fatal("disconnect events received")
+	}
 	log.Info("Simulation ended")
 }
 
@@ -395,11 +401,13 @@ func testSyncingViaDirectSubscribe(t *testing.T, chunkCount int, nodeCount int)
 		simulation.NewPeerEventsFilter().Drop(),
 	)
 
+	var disconnected atomic.Value
 	go func() {
 		for d := range disconnections {
-			log.Error("peer drop", "node", d.NodeID, "peer", d.PeerID)
-			t.Fatal("unexpected disconnect")
-			cancelSimRun()
+			if d.Error != nil {
+				log.Error("peer drop", "node", d.NodeID, "peer", d.PeerID)
+				disconnected.Store(true)
+			}
 		}
 	}()
 
@@ -514,6 +522,9 @@ func testSyncingViaDirectSubscribe(t *testing.T, chunkCount int, nodeCount int)
 		return result.Error
 	}
 
+	if yes, ok := disconnected.Load().(bool); ok && yes {
+		t.Fatal("disconnect events received")
+	}
 	log.Info("Simulation ended")
 	return nil
 }

+ 12 - 2
swarm/network/stream/syncer_test.go

@@ -18,11 +18,13 @@ package stream
 
 import (
 	"context"
+	"errors"
 	"fmt"
 	"io/ioutil"
 	"math"
 	"os"
 	"sync"
+	"sync/atomic"
 	"testing"
 	"time"
 
@@ -129,7 +131,7 @@ func testSyncBetweenNodes(t *testing.T, nodes, chunkCount int, skipCheck bool, p
 	if err != nil {
 		t.Fatal(err)
 	}
-	result := sim.Run(ctx, func(ctx context.Context, sim *simulation.Simulation) error {
+	result := sim.Run(ctx, func(ctx context.Context, sim *simulation.Simulation) (err error) {
 		nodeIDs := sim.UpNodeIDs()
 
 		nodeIndex := make(map[enode.ID]int)
@@ -143,11 +145,19 @@ func testSyncBetweenNodes(t *testing.T, nodes, chunkCount int, skipCheck bool, p
 			simulation.NewPeerEventsFilter().Drop(),
 		)
 
+		var disconnected atomic.Value
 		go func() {
 			for d := range disconnections {
 				if d.Error != nil {
 					log.Error("peer drop", "node", d.NodeID, "peer", d.PeerID)
-					t.Fatal(d.Error)
+					disconnected.Store(true)
+				}
+			}
+		}()
+		defer func() {
+			if err != nil {
+				if yes, ok := disconnected.Load().(bool); ok && yes {
+					err = errors.New("disconnect events received")
 				}
 			}
 		}()

+ 83 - 39
swarm/storage/netstore_test.go

@@ -20,6 +20,8 @@ import (
 	"bytes"
 	"context"
 	"crypto/rand"
+	"errors"
+	"fmt"
 	"io/ioutil"
 	"sync"
 	"testing"
@@ -114,19 +116,24 @@ func TestNetStoreGetAndPut(t *testing.T) {
 	defer cancel()
 
 	c := make(chan struct{}) // this channel ensures that the gouroutine with the Put does not run earlier than the Get
+	putErrC := make(chan error)
 	go func() {
 		<-c                                // wait for the Get to be called
 		time.Sleep(200 * time.Millisecond) // and a little more so it is surely called
 
 		// check if netStore created a fetcher in the Get call for the unavailable chunk
 		if netStore.fetchers.Len() != 1 || netStore.getFetcher(chunk.Address()) == nil {
-			t.Fatal("Expected netStore to use a fetcher for the Get call")
+			putErrC <- errors.New("Expected netStore to use a fetcher for the Get call")
+			return
 		}
 
 		err := netStore.Put(ctx, chunk)
 		if err != nil {
-			t.Fatalf("Expected no err got %v", err)
+			putErrC <- fmt.Errorf("Expected no err got %v", err)
+			return
 		}
+
+		putErrC <- nil
 	}()
 
 	close(c)
@@ -134,6 +141,10 @@ func TestNetStoreGetAndPut(t *testing.T) {
 	if err != nil {
 		t.Fatalf("Expected no err got %v", err)
 	}
+
+	if err := <-putErrC; err != nil {
+		t.Fatal(err)
+	}
 	// the retrieved chunk should be the same as what we Put
 	if !bytes.Equal(recChunk.Address(), chunk.Address()) || !bytes.Equal(recChunk.Data(), chunk.Data()) {
 		t.Fatalf("Different chunk received than what was put")
@@ -200,14 +211,18 @@ func TestNetStoreGetTimeout(t *testing.T) {
 	defer cancel()
 
 	c := make(chan struct{}) // this channel ensures that the gouroutine does not run earlier than the Get
+	fetcherErrC := make(chan error)
 	go func() {
 		<-c                                // wait for the Get to be called
 		time.Sleep(200 * time.Millisecond) // and a little more so it is surely called
 
 		// check if netStore created a fetcher in the Get call for the unavailable chunk
 		if netStore.fetchers.Len() != 1 || netStore.getFetcher(chunk.Address()) == nil {
-			t.Fatal("Expected netStore to use a fetcher for the Get call")
+			fetcherErrC <- errors.New("Expected netStore to use a fetcher for the Get call")
+			return
 		}
+
+		fetcherErrC <- nil
 	}()
 
 	close(c)
@@ -220,6 +235,10 @@ func TestNetStoreGetTimeout(t *testing.T) {
 		t.Fatalf("Expected context.DeadLineExceeded err got %v", err)
 	}
 
+	if err := <-fetcherErrC; err != nil {
+		t.Fatal(err)
+	}
+
 	// A fetcher was created, check if it has been removed after timeout
 	if netStore.fetchers.Len() != 0 {
 		t.Fatal("Expected netStore to remove the fetcher after timeout")
@@ -243,20 +262,29 @@ func TestNetStoreGetCancel(t *testing.T) {
 	ctx, cancel := context.WithTimeout(context.Background(), 1*time.Second)
 
 	c := make(chan struct{}) // this channel ensures that the gouroutine with the cancel does not run earlier than the Get
+	fetcherErrC := make(chan error, 1)
 	go func() {
 		<-c                                // wait for the Get to be called
 		time.Sleep(200 * time.Millisecond) // and a little more so it is surely called
 		// check if netStore created a fetcher in the Get call for the unavailable chunk
 		if netStore.fetchers.Len() != 1 || netStore.getFetcher(chunk.Address()) == nil {
-			t.Fatal("Expected netStore to use a fetcher for the Get call")
+			fetcherErrC <- errors.New("Expected netStore to use a fetcher for the Get call")
+			return
 		}
+
+		fetcherErrC <- nil
 		cancel()
 	}()
 
 	close(c)
+
 	// We call Get with an unavailable chunk, so it will create a fetcher and wait for delivery
 	_, err := netStore.Get(ctx, chunk.Address())
 
+	if err := <-fetcherErrC; err != nil {
+		t.Fatal(err)
+	}
+
 	// After the context is cancelled above Get should return with an error
 	if err != context.Canceled {
 		t.Fatalf("Expected context.Canceled err got %v", err)
@@ -286,46 +314,55 @@ func TestNetStoreMultipleGetAndPut(t *testing.T) {
 	ctx, cancel := context.WithTimeout(context.Background(), 3*time.Second)
 	defer cancel()
 
+	putErrC := make(chan error)
 	go func() {
 		// sleep to make sure Put is called after all the Get
 		time.Sleep(500 * time.Millisecond)
 		// check if netStore created exactly one fetcher for all Get calls
 		if netStore.fetchers.Len() != 1 {
-			t.Fatal("Expected netStore to use one fetcher for all Get calls")
+			putErrC <- errors.New("Expected netStore to use one fetcher for all Get calls")
+			return
 		}
 		err := netStore.Put(ctx, chunk)
 		if err != nil {
-			t.Fatalf("Expected no err got %v", err)
+			putErrC <- fmt.Errorf("Expected no err got %v", err)
+			return
 		}
+		putErrC <- nil
 	}()
 
+	count := 4
 	// call Get 4 times for the same unavailable chunk. The calls will be blocked until the Put above.
-	getWG := sync.WaitGroup{}
-	for i := 0; i < 4; i++ {
-		getWG.Add(1)
+	errC := make(chan error)
+	for i := 0; i < count; i++ {
 		go func() {
-			defer getWG.Done()
 			recChunk, err := netStore.Get(ctx, chunk.Address())
 			if err != nil {
-				t.Fatalf("Expected no err got %v", err)
+				errC <- fmt.Errorf("Expected no err got %v", err)
 			}
 			if !bytes.Equal(recChunk.Address(), chunk.Address()) || !bytes.Equal(recChunk.Data(), chunk.Data()) {
-				t.Fatalf("Different chunk received than what was put")
+				errC <- errors.New("Different chunk received than what was put")
 			}
+			errC <- nil
 		}()
 	}
 
-	finishedC := make(chan struct{})
-	go func() {
-		getWG.Wait()
-		close(finishedC)
-	}()
+	if err := <-putErrC; err != nil {
+		t.Fatal(err)
+	}
+
+	timeout := time.After(1 * time.Second)
 
 	// The Get calls should return after Put, so no timeout expected
-	select {
-	case <-finishedC:
-	case <-time.After(1 * time.Second):
-		t.Fatalf("Timeout waiting for Get calls to return")
+	for i := 0; i < count; i++ {
+		select {
+		case err := <-errC:
+			if err != nil {
+				t.Fatal(err)
+			}
+		case <-timeout:
+			t.Fatalf("Timeout waiting for Get calls to return")
+		}
 	}
 
 	// A fetcher was created, check if it has been removed after cancel
@@ -448,7 +485,7 @@ func TestNetStoreGetCallsOffer(t *testing.T) {
 	defer cancel()
 
 	// We call get for a not available chunk, it will timeout because the chunk is not delivered
-	chunk, err := netStore.Get(ctx, chunk.Address())
+	_, err := netStore.Get(ctx, chunk.Address())
 
 	if err != context.DeadlineExceeded {
 		t.Fatalf("Expect error %v got %v", context.DeadlineExceeded, err)
@@ -542,16 +579,12 @@ func TestNetStoreFetchFuncCalledMultipleTimes(t *testing.T) {
 		t.Fatalf("Expected netStore to have one fetcher for the requested chunk")
 	}
 
-	// Call wait three times parallelly
-	wg := sync.WaitGroup{}
-	for i := 0; i < 3; i++ {
-		wg.Add(1)
+	// Call wait three times in parallel
+	count := 3
+	errC := make(chan error)
+	for i := 0; i < count; i++ {
 		go func() {
-			err := wait(ctx)
-			if err != nil {
-				t.Fatalf("Expected no err got %v", err)
-			}
-			wg.Done()
+			errC <- wait(ctx)
 		}()
 	}
 
@@ -570,7 +603,12 @@ func TestNetStoreFetchFuncCalledMultipleTimes(t *testing.T) {
 	}
 
 	// wait until all wait calls return (because the chunk is delivered)
-	wg.Wait()
+	for i := 0; i < count; i++ {
+		err := <-errC
+		if err != nil {
+			t.Fatal(err)
+		}
+	}
 
 	// There should be no more fetchers for the delivered chunk
 	if netStore.fetchers.Len() != 0 {
@@ -606,23 +644,29 @@ func TestNetStoreFetcherLifeCycleWithTimeout(t *testing.T) {
 		t.Fatalf("Expected netStore to have one fetcher for the requested chunk")
 	}
 
-	// Call wait three times parallelly
-	wg := sync.WaitGroup{}
-	for i := 0; i < 3; i++ {
-		wg.Add(1)
+	// Call wait three times in parallel
+	count := 3
+	errC := make(chan error)
+	for i := 0; i < count; i++ {
 		go func() {
-			defer wg.Done()
 			rctx, rcancel := context.WithTimeout(context.Background(), 100*time.Millisecond)
 			defer rcancel()
 			err := wait(rctx)
 			if err != context.DeadlineExceeded {
-				t.Fatalf("Expected err %v got %v", context.DeadlineExceeded, err)
+				errC <- fmt.Errorf("Expected err %v got %v", context.DeadlineExceeded, err)
+				return
 			}
+			errC <- nil
 		}()
 	}
 
 	// wait until all wait calls timeout
-	wg.Wait()
+	for i := 0; i < count; i++ {
+		err := <-errC
+		if err != nil {
+			t.Fatal(err)
+		}
+	}
 
 	// There should be no more fetchers after timeout
 	if netStore.fetchers.Len() != 0 {