I have a web application with the following stack:
I have defined a few protobufs and compiled them into both Go and Dart successfully. When I run the Go server code, I am able to successfully make gRPC calls with Kreya, however when I try making the same call from Flutter using grpc/grpc_web.dart, though I keep running into the following error:
gRPC Error (code: 2, codeName: UNKNOWN, message: HTTP request completed without a status
(potential CORS issue), details: null, rawResponse: , trailers: {})
Here is my UI Code:
class FiltersService {
static ResponseFuture<Filters> getFilters() {
GrpcWebClientChannel channel =
GrpcWebClientChannel.xhr(Uri.parse('http://localhost:9000'));
FiltersServiceClient clientStub = FiltersServiceClient(
channel,
);
return clientStub.getFilters(Void());
}
}
Backend Code:
func StartServer() {
log.Println("Starting server")
listener, err := net.Listen("tcp", fmt.Sprintf(":%v", port))
if err != nil {
log.Fatalf("Unable to listen to port %v\n%v\n", port, err)
}
repositories.ConnectToMongoDB()
grpcServer = grpc.NewServer()
registerServices()
if err = grpcServer.Serve(listener); err != nil {
log.Fatalf("Failed to serve gRPC\n%v\n", err)
}
}
// Register services defined in protobufs to call from UI
func registerServices() {
cardsService := &services.CardsService{}
protos.RegisterCardsServiceServer(grpcServer, cardsService)
filtersService := &services.FiltersService{}
protos.RegisterFiltersServiceServer(grpcServer, filtersService)
}
As mentioned, the API call is successful when Kreya is used to make the call, however the Dart code keeps failing.
I have tried wrapping the gRPC server in the gRPC web proxy, however that also failed from both Dart and Kreya. Here is the code I tried:
func StartProxy() {
log.Println("Starting server")
listener, err := net.Listen("tcp", fmt.Sprintf(":%v", port))
if err != nil {
log.Fatalf("Unable to listen to port %v\n%v\n", port, err)
}
repositories.ConnectToMongoDB()
grpcServer = grpc.NewServer()
registerServices()
grpcWebServer := grpcweb.WrapServer(grpcServer)
httpServer := &http.Server{
Handler: http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
if r.ProtoMajor == 2 {
grpcWebServer.ServeHTTP(w, r)
} else {
w.Header().Set("Access-Control-Allow-Origin", "*")
w.Header().Set("Access-Control-Allow-Methods", "POST, GET, OPTIONS, PUT, DELETE")
w.Header().Set("Access-Control-Allow-Headers", "Accept, Content-Type, Content-Length, Accept-Encoding, X-CSRF-Token, Authorization, X-User-Agent, X-Grpc-Web")
w.Header().Set("grpc-status", "")
w.Header().Set("grpc-message", "")
if grpcWebServer.IsGrpcWebRequest(r) {
grpcWebServer.ServeHTTP(w, r)
}
}
}),
}
httpServer.Serve(listener)
}
func StartServer() {
StartProxy()
}
I am also aware of Envoy Proxy which can be used in place of this gRPC web proxy, however if I do that, I would be exposing the endpoints on Envoy as REST APIs, which would then forward the request as a gRPC call. From what I understand, this would require maintaining 2 versions of the data models - one for communication between the UI and Envoy (in JSON), and the other for communication between Envoy and the server (as protobuf). Is this the correct understanding? How can I move past this?
*** EDIT: ***
As per the suggestion in the comments, I have tried using Envoy in place of the go proxy. However, even now, I'm having trouble getting it to work. I'm now getting upstream connect error or disconnect/reset before headers. reset reason: overflow when trying to access the port exposed by Envoy 9001, though I can successfully call the backend service directly from kreya on port 9000.
Here is my envoy.yaml:
static_resources:
listeners:
- name: listener_0
address:
socket_address: { address: host.docker.internal, port_value: 9001 }
filter_chains:
- filters:
- name: envoy.filters.network.http_connection_manager
typed_config:
"@type": type.googleapis.com/envoy.extensions.filters.network.http_connection_manager.v3.HttpConnectionManager
codec_type: auto
stat_prefix: ingress_http
route_config:
name: local_route
virtual_hosts:
- name: local_service
domains: ["*"]
routes:
- match: { prefix: "/" }
route:
cluster: greeter_service
max_stream_duration:
grpc_timeout_header_max: 0s
cors:
allow_origin_string_match:
- prefix: "*"
allow_methods: GET, PUT, DELETE, POST, OPTIONS
allow_headers: keep-alive,user-agent,cache-control,content-type,content-transfer-encoding,x-accept-content-transfer-encoding,x-accept-response-streaming,x-user-agent,x-grpc-web,grpc-timeout
max_age: "1728000"
expose_headers: id,token,grpc-status,grpc-message
http_filters:
- name: envoy.filters.http.grpc_web
- name: envoy.filters.http.cors
- name: envoy.filters.http.router
clusters:
- name: greeter_service
connect_timeout: 0.25s
type: logical_dns
http2_protocol_options: {}
lb_policy: round_robin
# win/mac hosts: Use address: host.docker.internal instead of address: localhost in the line below
load_assignment:
cluster_name: cluster_0
endpoints:
- lb_endpoints:
- endpoint:
address:
socket_address:
address: host.docker.internal
port_value: 9000
I was able to resolve the issue by taking the suggestion in the comments and using Envoy to proxy instead of the go-proxy, though the solution didn't work purely out of the box according to the linked post.
Here is the working envoy.yaml
admin:
access_log_path: /tmp/admin_access.log
address:
socket_address:
address: 0.0.0.0
port_value: 9901
static_resources:
listeners:
- name: listener_0
address:
socket_address: { address: 0.0.0.0, port_value: 9000 }
filter_chains:
- filters:
- name: envoy.http_connection_manager
typed_config:
"@type": type.googleapis.com/envoy.extensions.filters.network.http_connection_manager.v3.HttpConnectionManager
codec_type: auto
stat_prefix: ingress_http
route_config:
name: local_route
virtual_hosts:
- name: local_service
domains: ["*"]
routes:
- match: { prefix: "/" }
route:
cluster: greeter_service
max_grpc_timeout: 0s
cors:
allow_origin_string_match:
- prefix: "*"
allow_methods: GET, PUT, DELETE, POST, OPTIONS
allow_headers: keep-alive,user-agent,cache-control,content-type,content-transfer-encoding,custom-header-1,x-accept-content-transfer-encoding,x-accept-response-streaming,x-user-agent,x-grpc-web,grpc-timeout
max_age: "1728000"
expose_headers: custom-header-1,grpc-status,grpc-message
http_filters:
- name: envoy.filters.http.grpc_web
- name: envoy.filters.http.cors
- name: envoy.filters.http.router
clusters:
- name: greeter_service
connect_timeout: 0.25s
type: logical_dns
lb_policy: round_robin
http2_protocol_options: {}
load_assignment:
cluster_name: cluster_0
endpoints:
- lb_endpoints:
- endpoint:
address:
socket_address:
address: host.docker.internal
port_value: 9001
The working Dockerfile
FROM envoyproxy/envoy:v1.20-latest
COPY ./envoy.yaml /etc/envoy/envoy.yaml
CMD /usr/local/bin/envoy -c /etc/envoy/envoy.yaml -l debug
And the commands used to run Envoy
docker build -t envoy .
The previous command has to be run in the same directory as the Dockerfile
docker run -p 9000:9000 -p 9901:9901 envoy
Where 9000 and 9901 are the ports I want to expose and be able to access externally (listed in the envoy.yaml)
**NOTE: **
Make sure to include http2_protocol_options: {}. Following some possible solutions online, I had removed it, which led to a connection reset due to protocol error. I had been stuck on this issue for hours until I saw that Envoy can forward a request as either HTTP/2 or HTTP/3, and decided to add it back, which finally allowed me to make a gRPC call using the gRPC Web client.
Hope this helps anyone else that may be coming across this issue.