Petera/handle resourceexhaused ingestion

bootstrap/bootstrap.go

@@ -8,7 +8,6 @@ import (
 	"time"
 
 	pebbleDB "github.com/cockroachdb/pebble"
-	"github.com/onflow/flow-evm-gateway/metrics"
 	"github.com/onflow/flow-go-sdk/access"
 	"github.com/onflow/flow-go-sdk/access/grpc"
 	"github.com/onflow/flow-go/fvm/environment"
@@ -21,9 +20,12 @@ import (
 	"github.com/rs/zerolog"
 	"github.com/sethvargo/go-limiter/memorystore"
 	grpcOpts "google.golang.org/grpc"
+	"google.golang.org/grpc/codes"
+	"google.golang.org/grpc/status"
 
 	"github.com/onflow/flow-evm-gateway/api"
 	"github.com/onflow/flow-evm-gateway/config"
+	"github.com/onflow/flow-evm-gateway/metrics"
 	"github.com/onflow/flow-evm-gateway/models"
 	errs "github.com/onflow/flow-evm-gateway/models/errors"
 	"github.com/onflow/flow-evm-gateway/services/ingestion"
@@ -33,6 +35,19 @@ import (
 	"github.com/onflow/flow-evm-gateway/storage/pebble"
 )
 
+const (
+	// DefaultMaxMessageSize is the default maximum message size for gRPC responses
+	DefaultMaxMessageSize = 1024 * 1024 * 1024
+
+	// DefaultResourceExhaustedRetryDelay is the default delay between retries when the server returns
+	// a ResourceExhausted error.
+	DefaultResourceExhaustedRetryDelay = 100 * time.Millisecond
+
+	// DefaultResourceExhaustedMaxRetryDelay is the default max request duration when retrying server
+	// ResourceExhausted errors.
+	DefaultResourceExhaustedMaxRetryDelay = 30 * time.Second
+)
+
 type Storages struct {
 	Storage      *pebble.Storage
 	Registers    *pebble.RegisterStorage
@@ -452,7 +467,13 @@ func setupCrossSporkClient(config config.Config, logger zerolog.Logger) (*reques
 	// create access client with cross-spork capabilities
 	currentSporkClient, err := grpc.NewClient(
 		config.AccessNodeHost,
-		grpc.WithGRPCDialOptions(grpcOpts.WithDefaultCallOptions(grpcOpts.MaxCallRecvMsgSize(1024*1024*1024))),
+		grpc.WithGRPCDialOptions(
+			grpcOpts.WithDefaultCallOptions(grpcOpts.MaxCallRecvMsgSize(DefaultMaxMessageSize)),
+			grpcOpts.WithUnaryInterceptor(retryInterceptor(
+				DefaultResourceExhaustedMaxRetryDelay,
+				DefaultResourceExhaustedRetryDelay,
+			)),
+		),
 	)
 	if err != nil {
 		return nil, fmt.Errorf(
@@ -487,6 +508,44 @@ func setupCrossSporkClient(config config.Config, logger zerolog.Logger) (*reques
 	return client, nil
 }
 
+// retryInterceptor is a gRPC client interceptor that retries the request when the server returns
+// a ResourceExhausted error
+func retryInterceptor(maxDuration, pauseDuration time.Duration) grpcOpts.UnaryClientInterceptor {
+	return func(
+		ctx context.Context,
+		method string,
+		req, reply interface{},
+		cc *grpcOpts.ClientConn,
+		invoker grpcOpts.UnaryInvoker,
+		opts ...grpcOpts.CallOption,
+	) error {
+		start := time.Now()
+		attempts := 0
+		for {
+			err := invoker(ctx, method, req, reply, cc, opts...)
+			if err == nil {
+				return nil
+			}
+
+			if status.Code(err) != codes.ResourceExhausted {
+				return err
+			}
+
+			attempts++
+			duration := time.Since(start)
+			if duration >= maxDuration {
+				return fmt.Errorf("request failed (attempts: %d, duration: %v): %w", attempts, duration, err)
+			}
+
+			select {
+			case <-ctx.Done():
+				return ctx.Err()
+			case <-time.After(pauseDuration):
+			}
+		}
+	}
+}
+
 // setupStorage creates storage and initializes it with configured starting cadence height
 // in case such a height doesn't already exist in the database.
 func setupStorage(
@@ -570,9 +629,9 @@ func setupStorage(
 			Stringer("fvm_address_for_evm_storage_account", storageAddress).
 			Msgf("database initialized with cadence height: %d", cadenceHeight)
 	}
-	//else {
+	// else {
 	//	// TODO(JanezP): verify storage account owner is correct
-	//}
+	// }
 
 	return db, &Storages{
 		Storage:      store,

bootstrap/bootstrap_test.go

@@ -0,0 +1,88 @@
+package bootstrap
+
+import (
+	"context"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+	"google.golang.org/grpc"
+	"google.golang.org/grpc/codes"
+	"google.golang.org/grpc/status"
+)
+
+func TestRetryInterceptor(t *testing.T) {
+	expecterErr := status.Error(codes.ResourceExhausted, "resource exhausted")
+	interceptor := retryInterceptor(100*time.Millisecond, 10*time.Millisecond)
+
+	testCases := []struct {
+		name           string
+		invoker        func(callCount int) error
+		maxRequestTime time.Duration
+		callCount      int // expect exact count
+		minCallCount   int // min, for when using a timeout
+		expectedErr    error
+	}{
+		{
+			name: "no error",
+			invoker: func(callCount int) error {
+				return nil
+			},
+			maxRequestTime: 10 * time.Millisecond,
+			callCount:      1,
+			expectedErr:    nil,
+		},
+		{
+			name: "succeeds on 3rd attempt",
+			invoker: func(callCount int) error {
+				if callCount >= 3 {
+					return nil
+				}
+				return expecterErr
+			},
+			maxRequestTime: 40 * time.Millisecond,
+			callCount:      3,
+			expectedErr:    nil,
+		},
+		{
+			name: "fails after timeout",
+			invoker: func(callCount int) error {
+				return expecterErr
+			},
+			maxRequestTime: 150 * time.Millisecond, // add a buffer for test slowness
+			minCallCount:   10,
+			expectedErr:    expecterErr,
+		},
+	}
+
+	for _, tc := range testCases {
+		tc := tc
+		t.Run(tc.name, func(t *testing.T) {
+			t.Parallel()
+
+			callCount := 0
+			invoker := func(context.Context, string, any, any, *grpc.ClientConn, ...grpc.CallOption) error {
+				callCount++
+				return tc.invoker(callCount)
+			}
+
+			start := time.Now()
+			err := interceptor(
+				context.Background(), "", nil, nil, nil,
+				invoker,
+			)
+			if tc.expectedErr != nil {
+				assert.ErrorIs(t, err, tc.expectedErr)
+			} else {
+				assert.NoError(t, err)
+			}
+
+			if tc.minCallCount > 0 {
+				assert.GreaterOrEqual(t, callCount, tc.minCallCount)
+			} else {
+				assert.Equal(t, callCount, tc.callCount)
+			}
+			assert.LessOrEqual(t, time.Since(start), tc.maxRequestTime)
+		})
+	}
+}