[FAB-816] Clean multichain integration interfaces

The multichain integration in https://gerrit.hyperledger.org/r/#/c/2721/
hooked the multi-chain manager into the default solo path.  This was
attempted in a way which kept the diff minimal, but correspondingly
produced some unpleasant artifacts in the code.

In particular, the integration used one catch-all interface in
multichain.Manager to supply support to the various components, this
meant that components were forced to drag in unnecessary imports and it
complicated mock testing with many panic-ing unimplemented functions.

This changeset breaks this interface into pieces, and pushes the
definition of the interface back into the components which depend on the
definition.

Also included in this changeset is substantial cleanup of the
solo/consenter_test.go file. This set of tests was originally written in
a way which depended on the end to end flow of the system, but as the
common components have been factored out, and as the solo 'consensus'
has been reduced to its simplest component, more targetted tests are now
needed.

Also included are some assorted linting fixes.

Change-Id: I7c5e20cd7b8c66eb51cc56ad539177ce81cbcbfc
Signed-off-by: Jason Yellick <jyellick@us.ibm.com>

Author: Jason Yellick

bddtests/docker-compose-with-orderer.yml

@@ -12,7 +12,7 @@ services:
       - ORDERER_GENERAL_MAXWINDOWSIZE=1000
       - ORDERER_GENERAL_ORDERERTYPE=solo
       - ORDERER_GENERAL_LISTENADDRESS=0.0.0.0
-      - ORDERER_GENERAL_LISTENPORT=5005
+      - ORDERER_GENERAL_LISTENPORT=7050
       - ORDERER_RAMLEDGER_HISTORY_SIZE=100
     working_dir: /opt/gopath/src/github.com/hyperledger/fabric/orderer
     command: orderer
@@ -29,7 +29,7 @@ services:
       - CORE_NEXT=true
       - CORE_PEER_ENDORSER_ENABLED=true
       - CORE_PEER_COMMITTER_ENABLED=true
-      - CORE_PEER_COMMITTER_LEDGER_ORDERER=orderer:5005
+      - CORE_PEER_COMMITTER_LEDGER_ORDERER=orderer:7050
     volumes:
         - /var/run/:/host/var/run/
     networks:
@@ -67,7 +67,7 @@ services:
     environment:
       - CORE_PEER_ID=vp2
       - CORE_PEER_PROFILE_ENABLED=true
-      - CORE_PEER_COMMITTER_LEDGER_ORDERER=orderer:5005
+      - CORE_PEER_COMMITTER_LEDGER_ORDERER=orderer:7050
       - CORE_PEER_GOSSIP_BOOTSTRAP=vp0:10000
       - CORE_PEER_GOSSIP_ORGLEADER=false
     command: peer node start

bddtests/steps/orderer_util.py

@@ -188,6 +188,8 @@ def getABStubForComposeService(self, context, composeService):
 		self.atomicBroadcastStubsDict[composeService] = newABStub
 		return newABStub
 
+# The default chain ID when the system is statically bootstrapped for testing
+TEST_CHAIN_ID = "**TEST_CHAINID**".encode()
 
 # Registerses a user on a specific composeService
 def registerUser(context, secretMsg, composeService):
@@ -217,7 +219,7 @@ def getUserRegistration(context, enrollId):
 def createDeliverUpdateMsg(Start, SpecifiedNumber, WindowSize):
     seek = ab_pb2.SeekInfo()
     startVal = seek.__getattribute__(Start)
-    seekInfo = ab_pb2.SeekInfo(Start = startVal, SpecifiedNumber = SpecifiedNumber, WindowSize = WindowSize)
+    seekInfo = ab_pb2.SeekInfo(Start = startVal, SpecifiedNumber = SpecifiedNumber, WindowSize = WindowSize, ChainID = TEST_CHAIN_ID)
     deliverUpdateMsg = ab_pb2.DeliverUpdate(Seek = seekInfo)
     return deliverUpdateMsg
 
@@ -227,7 +229,8 @@ def generateBroadcastMessages(numToGenerate = 1, timeToHoldOpen = 1):
     for i in range(0, numToGenerate):
         envelope = common_pb2.Envelope()
         payload = common_pb2.Payload(header = common_pb2.Header(chainHeader = common_pb2.ChainHeader()))
-        # TODO, appropriately set the header type
+        payload.header.chainHeader.chainID = TEST_CHAIN_ID
+        payload.header.chainHeader.type = common_pb2.ENDORSER_TRANSACTION
         payload.data = str("BDD test: {0}".format(datetime.datetime.utcnow()))
         envelope.payload = payload.SerializeToString()
         messages.append(envelope)

orderer/common/blockcutter/blockcutter.go

@@ -52,6 +52,7 @@ type receiver struct {
 	curBatch      []*cb.Envelope
 }
 
+// NewReceiverImpl creates a Receiver implementation based on the given batchsize, filters, and configtx manager
 func NewReceiverImpl(batchSize int, filters *broadcastfilter.RuleSet, configManager configtx.Manager) Receiver {
 	return &receiver{
 		batchSize:     batchSize,
@@ -106,9 +107,8 @@ func (r *receiver) Ordered(msg *cb.Envelope) ([][]*cb.Envelope, bool) {
 		secondBatch := []*cb.Envelope{msg}
 		if firstBatch == nil {
 			return [][]*cb.Envelope{secondBatch}, true
-		} else {
-			return [][]*cb.Envelope{firstBatch, secondBatch}, true
 		}
+		return [][]*cb.Envelope{firstBatch, secondBatch}, true
 	case broadcastfilter.Reject:
 		logger.Debugf("Rejecting message")
 		return nil, false

orderer/common/bootstrap/static/static.go

@@ -26,7 +26,10 @@ import (
 	ab "github.com/hyperledger/fabric/protos/orderer"
 )
 
-var TestChainID = "**TEST_CHAINID**"
+// TestChainID is the default chain ID which is used by all statically bootstrapped networks
+// This is necessary to allow test clients to connect without being rejected for targetting
+// a chain which does not exist
+const TestChainID = "**TEST_CHAINID**"
 
 const msgVersion = int32(1)
 
@@ -38,12 +41,20 @@ type bootstrapper struct {
 	batchSize     int32
 }
 
+const (
+	// DefaultBatchSize is the default value of BatchSizeKey
+	DefaultBatchSize = 10
+
+	// DefaultConsensusType is the default value of ConsensusTypeKey
+	DefaultConsensusType = "solo"
+)
+
 // New returns a new static bootstrap helper.
 func New() bootstrap.Helper {
 	return &bootstrapper{
 		chainID:       TestChainID,
-		consensusType: "solo",
-		batchSize:     10,
+		consensusType: DefaultConsensusType,
+		batchSize:     DefaultBatchSize,
 	}
 }
 

orderer/common/broadcast/broadcast.go

@@ -18,7 +18,6 @@ package broadcast
 
 import (
 	"github.com/hyperledger/fabric/orderer/common/broadcastfilter"
-	"github.com/hyperledger/fabric/orderer/multichain"
 	cb "github.com/hyperledger/fabric/protos/common"
 	ab "github.com/hyperledger/fabric/protos/orderer"
 	"github.com/op/go-logging"
@@ -38,18 +37,30 @@ type Handler interface {
 	Handle(srv ab.AtomicBroadcast_BroadcastServer) error
 }
 
+// SupportManager provides a way for the Handler to look up the Support for a chain
+type SupportManager interface {
+	GetChain(chainID string) (Support, bool)
+}
+
+// Support provides the backing resources needed to support broadcast on a chain
+type Support interface {
+	// Enqueue accepts a message and returns true on acceptance, or false on shutdown
+	Enqueue(env *cb.Envelope) bool
+
+	// Filters returns the set of broadcast filters for this chain
+	Filters() *broadcastfilter.RuleSet
+}
+
 type handlerImpl struct {
 	queueSize int
-	ml        multichain.Manager
-	exitChan  chan struct{}
+	sm        SupportManager
 }
 
 // NewHandlerImpl constructs a new implementation of the Handler interface
-func NewHandlerImpl(queueSize int, ml multichain.Manager) Handler {
+func NewHandlerImpl(sm SupportManager, queueSize int) Handler {
 	return &handlerImpl{
 		queueSize: queueSize,
-		ml:        ml,
-		exitChan:  make(chan struct{}),
+		sm:        sm,
 	}
 }
 
@@ -61,44 +72,45 @@ func (bh *handlerImpl) Handle(srv ab.AtomicBroadcast_BroadcastServer) error {
 	return b.queueEnvelopes(srv)
 }
 
-type msgAndChainSupport struct {
-	msg          *cb.Envelope
-	chainSupport multichain.ChainSupport
+type msgAndSupport struct {
+	msg     *cb.Envelope
+	support Support
 }
 
 type broadcaster struct {
-	bs    *handlerImpl
-	queue chan *msgAndChainSupport
+	bs       *handlerImpl
+	queue    chan *msgAndSupport
+	exitChan chan struct{}
 }
 
 func newBroadcaster(bs *handlerImpl) *broadcaster {
 	b := &broadcaster{
-		bs:    bs,
-		queue: make(chan *msgAndChainSupport, bs.queueSize),
+		bs:       bs,
+		queue:    make(chan *msgAndSupport, bs.queueSize),
+		exitChan: make(chan struct{}),
 	}
 	return b
 }
 
 func (b *broadcaster) drainQueue() {
-	for {
-		select {
-		case msgAndChainSupport, ok := <-b.queue:
-			if ok {
-				if !msgAndChainSupport.chainSupport.Chain().Enqueue(msgAndChainSupport.msg) {
-					return
-				}
-			} else {
-				return
-			}
-		case <-b.bs.exitChan:
+	defer close(b.exitChan)
+	for msgAndSupport := range b.queue {
+		if !msgAndSupport.support.Enqueue(msgAndSupport.msg) {
+			logger.Debugf("Consenter instructed us to shut down")
 			return
 		}
 	}
+	logger.Debugf("Exiting because the queue channel closed")
 }
 
 func (b *broadcaster) queueEnvelopes(srv ab.AtomicBroadcast_BroadcastServer) error {
 
 	for {
+		select {
+		case <-b.exitChan:
+			return nil
+		default:
+		}
 		msg, err := srv.Recv()
 		if err != nil {
 			return err
@@ -111,20 +123,20 @@ func (b *broadcaster) queueEnvelopes(srv ab.AtomicBroadcast_BroadcastServer) err
 			return srv.Send(&ab.BroadcastResponse{Status: cb.Status_BAD_REQUEST})
 		}
 
-		chainSupport, ok := b.bs.ml.GetChain(payload.Header.ChainHeader.ChainID)
+		support, ok := b.bs.sm.GetChain(payload.Header.ChainHeader.ChainID)
 		if !ok {
 			// XXX Hook in chain creation logic here
 			panic("Unimplemented")
 		}
 
-		action, _ := chainSupport.Filters().Apply(msg)
+		action, _ := support.Filters().Apply(msg)
 
 		switch action {
 		case broadcastfilter.Reconfigure:
 			fallthrough
 		case broadcastfilter.Accept:
 			select {
-			case b.queue <- &msgAndChainSupport{msg: msg, chainSupport: chainSupport}:
+			case b.queue <- &msgAndSupport{msg: msg, support: support}:
 				err = srv.Send(&ab.BroadcastResponse{Status: cb.Status_SUCCESS})
 			default:
 				err = srv.Send(&ab.BroadcastResponse{Status: cb.Status_SERVICE_UNAVAILABLE})