rpc: add limit for batch request items and response size (#26681)

This PR adds server-side limits for JSON-RPC batch requests. Before this change, batches
were limited only by processing time. The server would pick calls from the batch and
answer them until the response timeout occurred, then stop processing the remaining batch
items.

Here, we are adding two additional limits which can be configured:

- the 'item limit': batches can have at most N items
- the 'response size limit': batches can contain at most X response bytes

These limits are optional in package rpc. In Geth, we set a default limit of 1000 items
and 25MB response size.

When a batch goes over the limit, an error response is returned to the client. However,
doing this correctly isn't always possible. In JSON-RPC, only method calls with a valid
`id` can be responded to. Since batches may also contain non-call messages or
notifications, the best effort thing we can do to report an error with the batch itself is
reporting the limit violation as an error for the first method call in the batch. If a batch is
too large, but contains only notifications and responses, the error will be reported with
a null `id`.

The RPC client was also changed so it can deal with errors resulting from too large
batches. An older client connected to the server code in this PR could get stuck
until the request timeout occurred when the batch is too large. **Upgrading to a version
of the RPC client containing this change is strongly recommended to avoid timeout issues.**

For some weird reason, when writing the original client implementation, @fjl worked off of
the assumption that responses could be distributed across batches arbitrarily. So for a
batch request containing requests `[A B C]`, the server could respond with `[A B C]` but
also with `[A B] [C]` or even `[A] [B] [C]` and it wouldn't make a difference to the
client.

So in the implementation of BatchCallContext, the client waited for all requests in the
batch individually. If the server didn't respond to some of the requests in the batch, the
client would eventually just time out (if a context was used).

With the addition of batch limits into the server, we anticipate that people will hit this
kind of error way more often. To handle this properly, the client now waits for a single
response batch and expects it to contain all responses to the requests.

---------

Co-authored-by: Felix Lange <fjl@twurst.com>
Co-authored-by: Martin Holst Swende <martin@swende.se>

node/api.go

@@ -176,6 +176,10 @@ func (api *adminAPI) StartHTTP(host *string, port *int, cors *string, apis *stri
 		CorsAllowedOrigins: api.node.config.HTTPCors,
 		Vhosts:             api.node.config.HTTPVirtualHosts,
 		Modules:            api.node.config.HTTPModules,
+		rpcEndpointConfig: rpcEndpointConfig{
+			batchItemLimit:         api.node.config.BatchRequestLimit,
+			batchResponseSizeLimit: api.node.config.BatchResponseMaxSize,
+		},
 	}
 	if cors != nil {
 		config.CorsAllowedOrigins = nil
@@ -250,6 +254,10 @@ func (api *adminAPI) StartWS(host *string, port *int, allowedOrigins *string, ap
 		Modules: api.node.config.WSModules,
 		Origins: api.node.config.WSOrigins,
 		// ExposeAll: api.node.config.WSExposeAll,
+		rpcEndpointConfig: rpcEndpointConfig{
+			batchItemLimit:         api.node.config.BatchRequestLimit,
+			batchResponseSizeLimit: api.node.config.BatchResponseMaxSize,
+		},
 	}
 	if apis != nil {
 		config.Modules = nil

node/config.go

@@ -197,6 +197,12 @@ type Config struct {
 	// AllowUnprotectedTxs allows non EIP-155 protected transactions to be send over RPC.
 	AllowUnprotectedTxs bool `toml:",omitempty"`
 
+	// BatchRequestLimit is the maximum number of requests in a batch.
+	BatchRequestLimit int `toml:",omitempty"`
+
+	// BatchResponseMaxSize is the maximum number of bytes returned from a batched rpc call.
+	BatchResponseMaxSize int `toml:",omitempty"`
+
 	// JWTSecret is the path to the hex-encoded jwt secret.
 	JWTSecret string `toml:",omitempty"`
 

node/defaults.go

@@ -46,17 +46,19 @@ var (
 
 // DefaultConfig contains reasonable default settings.
 var DefaultConfig = Config{
-	DataDir:             DefaultDataDir(),
-	HTTPPort:            DefaultHTTPPort,
-	AuthAddr:            DefaultAuthHost,
-	AuthPort:            DefaultAuthPort,
-	AuthVirtualHosts:    DefaultAuthVhosts,
-	HTTPModules:         []string{"net", "web3"},
-	HTTPVirtualHosts:    []string{"localhost"},
-	HTTPTimeouts:        rpc.DefaultHTTPTimeouts,
-	WSPort:              DefaultWSPort,
-	WSModules:           []string{"net", "web3"},
-	GraphQLVirtualHosts: []string{"localhost"},
+	DataDir:              DefaultDataDir(),
+	HTTPPort:             DefaultHTTPPort,
+	AuthAddr:             DefaultAuthHost,
+	AuthPort:             DefaultAuthPort,
+	AuthVirtualHosts:     DefaultAuthVhosts,
+	HTTPModules:          []string{"net", "web3"},
+	HTTPVirtualHosts:     []string{"localhost"},
+	HTTPTimeouts:         rpc.DefaultHTTPTimeouts,
+	WSPort:               DefaultWSPort,
+	WSModules:            []string{"net", "web3"},
+	BatchRequestLimit:    1000,
+	BatchResponseMaxSize: 25 * 1000 * 1000,
+	GraphQLVirtualHosts:  []string{"localhost"},
 	P2P: p2p.Config{
 		ListenAddr: ":30303",
 		MaxPeers:   50,

node/node.go

@@ -101,10 +101,11 @@ func New(conf *Config) (*Node, error) {
 	if strings.HasSuffix(conf.Name, ".ipc") {
 		return nil, errors.New(`Config.Name cannot end in ".ipc"`)
 	}
-
+	server := rpc.NewServer()
+	server.SetBatchLimits(conf.BatchRequestLimit, conf.BatchResponseMaxSize)
 	node := &Node{
 		config:        conf,
-		inprocHandler: rpc.NewServer(),
+		inprocHandler: server,
 		eventmux:      new(event.TypeMux),
 		log:           conf.Logger,
 		stop:          make(chan struct{}),
@@ -403,6 +404,11 @@ func (n *Node) startRPC() error {
 		openAPIs, allAPIs = n.getAPIs()
 	)
 
+	rpcConfig := rpcEndpointConfig{
+		batchItemLimit:         n.config.BatchRequestLimit,
+		batchResponseSizeLimit: n.config.BatchResponseMaxSize,
+	}
+
 	initHttp := func(server *httpServer, port int) error {
 		if err := server.setListenAddr(n.config.HTTPHost, port); err != nil {
 			return err
@@ -412,6 +418,7 @@ func (n *Node) startRPC() error {
 			Vhosts:             n.config.HTTPVirtualHosts,
 			Modules:            n.config.HTTPModules,
 			prefix:             n.config.HTTPPathPrefix,
+			rpcEndpointConfig:  rpcConfig,
 		}); err != nil {
 			return err
 		}
@@ -425,9 +432,10 @@ func (n *Node) startRPC() error {
 			return err
 		}
 		if err := server.enableWS(openAPIs, wsConfig{
-			Modules: n.config.WSModules,
-			Origins: n.config.WSOrigins,
-			prefix:  n.config.WSPathPrefix,
+			Modules:           n.config.WSModules,
+			Origins:           n.config.WSOrigins,
+			prefix:            n.config.WSPathPrefix,
+			rpcEndpointConfig: rpcConfig,
 		}); err != nil {
 			return err
 		}
@@ -441,26 +449,29 @@ func (n *Node) startRPC() error {
 		if err := server.setListenAddr(n.config.AuthAddr, port); err != nil {
 			return err
 		}
+		sharedConfig := rpcConfig
+		sharedConfig.jwtSecret = secret
 		if err := server.enableRPC(allAPIs, httpConfig{
 			CorsAllowedOrigins: DefaultAuthCors,
 			Vhosts:             n.config.AuthVirtualHosts,
 			Modules:            DefaultAuthModules,
 			prefix:             DefaultAuthPrefix,
-			jwtSecret:          secret,
+			rpcEndpointConfig:  sharedConfig,
 		}); err != nil {
 			return err
 		}
 		servers = append(servers, server)
+
 		// Enable auth via WS
 		server = n.wsServerForPort(port, true)
 		if err := server.setListenAddr(n.config.AuthAddr, port); err != nil {
 			return err
 		}
 		if err := server.enableWS(allAPIs, wsConfig{
-			Modules:   DefaultAuthModules,
-			Origins:   DefaultAuthOrigins,
-			prefix:    DefaultAuthPrefix,
-			jwtSecret: secret,
+			Modules:           DefaultAuthModules,
+			Origins:           DefaultAuthOrigins,
+			prefix:            DefaultAuthPrefix,
+			rpcEndpointConfig: sharedConfig,
 		}); err != nil {
 			return err
 		}

node/rpcstack.go

@@ -41,15 +41,21 @@ type httpConfig struct {
 	CorsAllowedOrigins []string
 	Vhosts             []string
 	prefix             string // path prefix on which to mount http handler
-	jwtSecret          []byte // optional JWT secret
+	rpcEndpointConfig
 }
 
 // wsConfig is the JSON-RPC/Websocket configuration
 type wsConfig struct {
-	Origins   []string
-	Modules   []string
-	prefix    string // path prefix on which to mount ws handler
-	jwtSecret []byte // optional JWT secret
+	Origins []string
+	Modules []string
+	prefix  string // path prefix on which to mount ws handler
+	rpcEndpointConfig
+}
+
+type rpcEndpointConfig struct {
+	jwtSecret              []byte // optional JWT secret
+	batchItemLimit         int
+	batchResponseSizeLimit int
 }
 
 type rpcHandler struct {
@@ -297,6 +303,7 @@ func (h *httpServer) enableRPC(apis []rpc.API, config httpConfig) error {
 
 	// Create RPC server and handler.
 	srv := rpc.NewServer()
+	srv.SetBatchLimits(config.batchItemLimit, config.batchResponseSizeLimit)
 	if err := RegisterApis(apis, config.Modules, srv); err != nil {
 		return err
 	}
@@ -328,6 +335,7 @@ func (h *httpServer) enableWS(apis []rpc.API, config wsConfig) error {
 	}
 	// Create RPC server and handler.
 	srv := rpc.NewServer()
+	srv.SetBatchLimits(config.batchItemLimit, config.batchResponseSizeLimit)
 	if err := RegisterApis(apis, config.Modules, srv); err != nil {
 		return err
 	}

node/rpcstack_test.go

@@ -339,8 +339,10 @@ func TestJWT(t *testing.T) {
 		ss, _ := jwt.NewWithClaims(method, testClaim(input)).SignedString(secret)
 		return ss
 	}
-	srv := createAndStartServer(t, &httpConfig{jwtSecret: []byte("secret")},
-		true, &wsConfig{Origins: []string{"*"}, jwtSecret: []byte("secret")}, nil)
+	cfg := rpcEndpointConfig{jwtSecret: []byte("secret")}
+	httpcfg := &httpConfig{rpcEndpointConfig: cfg}
+	wscfg := &wsConfig{Origins: []string{"*"}, rpcEndpointConfig: cfg}
+	srv := createAndStartServer(t, httpcfg, true, wscfg, nil)
 	wsUrl := fmt.Sprintf("ws://%v", srv.listenAddr())
 	htUrl := fmt.Sprintf("http://%v", srv.listenAddr())