grpc/grpc

gRPC gitter chat

Rex Tsao

@clarifysky

Hello, anyone who can help me to solve this problem: "code 14, failed to connecto to all addresses" ?

Rex Tsao

@clarifysky

I noticed that after I chmod the sock file to 0777, the problem gone.

Vilém Pavlíček

@vildapavlicek

Hello. I'm defining proto3 files for the first time and I'm wondering if it is okay to define empty message like message MyMessageReply{}? The use case is I need to send some data to server, process them server-side and then just let the client know when done returning reply or error, but I do not need to send back any data.

4 replies

Santiago Gil

@santigl_gitlab

Hi. I am trying to find the right way to abort a call from the server side in a client-streaming call

1 reply

I was expecting to make it so that the client will notice that the stream is closed when it calls writer(request2) and that returns false, but is that the right way?

Andrew Braunstein

@awbraunstein

I'm looking for some help with regards to this stack overflow question I asked. https://stackoverflow.com/questions/63200740/how-to-access-the-method-name-in-a-python-grpc-handler Specifically around accessing the service/method name of the incoming RPC from within the handler in Python. I know this is accessible via interceptors, but there doesn't seem to be a good way to propagate that information into the actual handler itself.

Lidi Zheng

@lidizheng

@awbraunstein I think you have found the right component (HandlerCallDetails ) in your StackOverflow description. HandlerCallDetails is used by the GenericRpcHandler: https://grpc.github.io/grpc/python/grpc.html#grpc.GenericRpcHandler. You can implement your own handler and supply it to the gRPC server. The ProtoBuf generated code hides this detail and it can be good reference to see how it is done.

1 reply

HelloWood

@helloworlde

Is there anyone known how to print binary log? I found the design document in https://github.com/grpc/proposal/blob/master/A16-binary-logging.md, but neither grpc-java nor grpc-go print the log, how could I open this setting? GRPC_BINARY_LOG_FILTER= and GRPC_BINARY_LOG_CONFIG= doesn't work at all.

10 replies

Edgar González

@gonzedge

:wave: Hello - I see the latest release of grpc is v1.30.2 https://github.com/grpc/grpc/releases/tag/v1.30.2. However, the latest version I see on pypi is v1.30.0: https://pypi.org/project/grpcio-tools/

10 replies

noam-hoze

@noam-hoze

Hello,
I have a question regarding grpc OSS compliance.
Who should I attribute the copyright to?
Searching in the repo (https://github.com/grpc/grpc-node/search?p=2&q=copyright&unscoped_q=copyright) you get:

I assume it should be the latest.

Also, you have the AUTHORS in the parent grpc-node (https://github.com/grpc/grpc-node/blob/master/AUTHORS) file which attributes it to Google.

If someone can help that would be much appreciated.

Thanks!

1 reply

PraneethaYanala

@PraneethaYanala

Hi All, we are building an application using grpc-java, I have few questions regarding the ManagedChannel, from the git issues and Stack Overflow I understood that it is recommended to use single channel for entire application, but I have few questions regarding it -
What are the scenarios during which more than one grpc Channel is recommended? (assuming that we don't need name resolver and load balancer)
Based on what factors should we decide how many channels we need (like number of concurrent requests or any other stats)?
If we need to have multiple channels when exactly should we open a new connection?
Do we have any library to support grpc connection pooling?

6 replies

DMR-8

@DMR-8

Hi All,
I am making a desktop application that has to stream data from incoming UDP ports to a number of applications on the system. Is Grpc a good solution for such inter process communication where we need real time streaming of data?

3 replies

ps142020

@ps142020

Hi, I am running C++ helloworld with client and server on different terminal and see "14: Socket closed" I see previously many folks faced this but did not see a root cause or solution, can someone help?

2 replies

HelloWood

@helloworlde

Does the grpc-java has metrics for retry? If so how could I use it, if not is there has lib to record retry metrics?

9 replies

DMR-8

@DMR-8

I want to use grpc for inter procees communication between applications on the same system. The UDP data is coming from the sensors and I have an application that forwards this stream of data via grpc to other applications. It is not a cloud solution. Would grpc still be advisable ?

2 replies

DMR-8

@DMR-8

That’s great to hear

Thank You

Roger Ingouacka

@Ze-ro-G

Hello I am trying to use swift grpc to do a simple request but I can't figure it out :( Do you know where I can find any help please ?

marcoscleison

@marcoscleison

Hi all. I am new to grpc. I could compile the lib and generate the C++ files. However, I am experimenting an issue with the integration with Cmake. Cmake is not compiling .grpc.pb.cc .pb.cc files trigging an error saying that the those extensions are not C++ recognized extensions. My cmake is 3.18.1, Ubuntu 16.04, clang++ 3.8. Could anyone to gime me a tip about the issue? Thank you very much.

HelloWood

@helloworlde

I found the repo https://github.com/grpc-ecosystem/grift.git, looks like this tool can generate gRPC code by thrift IDL file, is there any possible to make gRPC service and thrift call each other?

1 reply

Francois-air

@Francois-air

I'm looking to play around with the experimental python asyncio API, as I think it'll solve some issues on my end. Is there an example of full-duplex streaming-streaming usage (from the server side) with that API around somewhere (i.e. that can yield a reply while awaiting a request)?

2 replies

Loren Sanz

@mrsanz

Hey had a QQ about GRPC and node support. The docs for node are pretty bad and I've been around the horn a few times generating this and that... in the end I'm trying to generate the client stub for a pretty complex grpc protobuf.

Lidi Zheng

@lidizheng

@Francois-air Yes. That is possible. For the asyncio API, you can send in one coroutine, and receive (await) in another coroutine.

Loren Sanz

@mrsanz

I'd like to just use something simple like the npm package grpc-caller. but my protobuf has deeply linked dependencies

so I'm getting errors like -> Error: no such Type or Enum 'member.protobuf.MemberClassLabel' in Type audience.protobuf.Audience

I've tried static compilation but can't seem to generate a client stub as I'm not sure what flags to pass to protoc
And dynamic loading fails due to the above reason..
Thanks in advance for any help or advice anyone can offer

Loren Sanz

@mrsanz

So I guess I'll circle back around to my question. For a large collection of protobuf's what library for node should I be using to generate a grpc node client? probtobufjs, grpc (@grpc/grpc-loader), grpc-web or something else?

7 replies

Stephen Gutekanst

@slimsag

Questions about strongly typing JSON messages in gRPC

10 replies

HelloWood

@helloworlde

Is there has config for every reqeust timeout when open retry? Now the timeout config is worked for all request(origin request and retry request), and when client timeout with DEADLINE_EXCEEDED, the not completed retry request was canceled too, how could I set timeout config for every request @gonzedge
And timeout config for not retry method and retry method is different, is there support use different configuration?

Srinjoy Santra

@Srinjoy-Santra

This is a bit old question.
I need to send a JSON as request to a grpc API via Envoy. Currently, I am using setter functions from protoc auto generated code. Is there any way to directly convert json (even, deeply nested ones) to proto message and send it as request?

const cat = new Category();
    cat.setCode('XY')
    cat.setRatingsList([0.1, 0.6, 0.9])
    cat.setLiked(true)


    const catJson = {code: 'XY', ratings: [0.1, 0.4], liked: false}
    console.log(catJson)// CAN THIS BE DIRECTLY ENCODED AND SEND REQUEST??

    await request.setSample(cat) //Instead of using the setters!

More details here: https://stackoverflow.com/questions/60430081/how-to-convert-json-to-proto-message-format-in-javascript-client-side

1 reply

togis123

@togis123

I bringing up gRPC client-server C ++ application. I have following queries will designing the thread modelling and connection management

-> How do gRPC server knows that if a client got disconnected from the server. Is there any context maintained in the server for each client.?

1 reply

-> By using the gRPC keep alive server can ping the client as well or its other way alone.?

1 reply

Vineeth Sagar

@vsag96

When I am trying to build gRPC I get the following errors. I installed protobuf using protobuf-cpp-3.6.1.tar.gz What's the reason for this? the output of pkg-config --libs protobuf is -lprotobuf -lpthread. Can anyone suggest a fix?

/usr/local/include/google/protobuf/io/coded_stream.h:697: undefined reference to google::protobuf::io::EpsCopyOutputStream::WriteRawFallback(void const*, int, unsigned char*) /usr/local/include/google/protobuf/io/coded_stream.h:697: undefined reference tod google::protobuf::io::EpsCopyOutputStream::WriteRawFallback(void const*, int, unsigned char*) /usr/local/include/google/protobuf/io/coded_stream.h:697: undefined reference to google::protobuf::io::EpsCopyOutputStream::WriteRawFallback(void const*, int, unsigned char*)

Vineeth Sagar

@vsag96

The fix was to use compile using HAS_SYSTEM_PROTOBUF=false. I know make is deprecated. If there's a cmake alternative for this, other can chime in.

HelloWood

@helloworlde

Is there has config for every reqeust timeout when open retry? Now the timeout config is worked for all request(origin request and retry request), and when client timeout with DEADLINE_EXCEEDED, the not completed retry request was canceled too, how could I set timeout config for every request @gonzedge
And timeout config for not retry method and retry method is different, is there support use different configuration?
@ejona86 Does there has any config for retry?

William Chiu

@WilliamChiu

Hi all, I am trying to access a python grpc server running on google cloud run from a statically hosted site, but am running into CORS issues. Hitting the cloud run url using grpcurl seems to work fine, but grpc-web isn't able to connect. Would anybody have any tips for debugging? Thank you in advance

1 reply

HelloWood

@helloworlde

As my previous question, does there support retry at client when client is timeout? Now retry need response status of server side, but if there no reponse, the retry will not work, looks like this is not a good idea.

2 replies

Stefano Siaeh

@SSiaeh_twitter

Hi all....
I would like to mock my gRPC client so i need that it implements an interface.
Is possible edit csproj file in order to have my client implements an interface?

5 replies

Nithisha-BR

@Nithisha-BR

Hi All,

HelloWood

@helloworlde

How could I distinguish retry and hedging request? I found method io.grpc.ClientStreamTracer.Factory#newClientStreamTracer(io.grpc.ClientStreamTracer.StreamInfo, io.grpc.Metadata) info.getCallOptions().getOption(RETRY_KEY) to get config, but the CallOptions.Key<String> not override equals method, and it compare result is false, so I could get config! Is there possible to put request type in header or make io.grpc.internal.ServiceConfigInterceptor public in future?

2 replies

Nithisha-BR

@Nithisha-BR

Hi All, I need to use GRPC between C# client and CPP server. Can any one guide me the right place to start with.

1 reply

Ryan Ju

@ryan-ju

Morning, I'm using the Mali framework with Node grpc (not grpc-js), and I found that the server crashes when the client ungracefully disconnects:

Error: startBatch failed
    at handleError (/Users/ran/Projects/grpc-demo-service/node_modules/grpc/src/server.js:72:8)
    at ServerUnaryCall.<anonymous> (/Users/ran/Projects/grpc-demo-service/node_modules/grpc/src/server.js:573:5)
    at ServerUnaryCall.emit (events.js:315:20)
    at /Users/ran/Projects/grpc-demo-service/node_modules/grpc/src/server.js:486:14 {
  code: 8
}

I've tested this on Debian, Alpine and MacOS, and happens in all of them. This should be a very obvious issue, has anyone managed to fix it?

piyush-devnow-io

@piyush-devnow-io

Hi,

I have created a nodejs grpc server which provides perfect response upon calling from BloomGrpc client.
I am calling the same method from my spring boot application and it gives me the following error
io.grpc.StatusRuntimeException: UNIMPLEMENTED: RPC method not implemented <generated class name.method name>
the proto file is same in both cases

Can someone assist me on this.

2 replies

sivanmahesh

@sivanmahesh

Hi all, i am new to GRPC. Exciting concept. Had a very basic question.. Are there any documented use cases or can anyone tell me when do we use client streaming ? If i have a large file contents (say 500mb) that i need to send over to a service for processing, is it good to send chunks over client streaming or can i send it in a huge payload ?

1 reply

jevgenija

@jpantiuchina

Dear developers,

As part of a research team from Università della Svizzera italiana (Switzerland) and University of Sannio (Italy), we have analyzed refactoring pull requests in grpc/grpc-java repository and are looking for developers for a short 5-10 min survey (https://usi.eu.qualtrics.com/jfe/form/SV_cO6Ayah0D6q4eSF). Would you please spare your time by answering some questions about refactoring-related contributions? We would greatly appreciate your input — it would help us understand how developers can improve the quality of refactoring contributions, and benefit the development process. The responses will be anonymized and handled confidentially! Thank you a lot!

If you consider this message to be spam, I'm very sorry! There will be no follow-up to bother you.

Quynh Anh "Emma" Nguyen

@qaemma

Hi, beginner question: in grpc-java, is there a way to suppress Deadline Exceeded log messages? Thanks!

5 replies

HelloWood

@helloworlde

What's the best practice of initialBackoff and maxBackoff when retry, is there is good choice to set more close than zero when have many server? Looks like reduce delay is work well to improve TP999

coffee93-coder

@coffee93-coder

Hi! Is there way to generate phpunit mock class for grpc call?
I can only find google test code. (https://googleapis.github.io/gax-php/0.37.0/Google/ApiCore/Testing/MockUnaryCall.html)
I cannot find way to generate test code for create mock client and invoke my grpc request code.

2 replies

Ryan Ju

@ryan-ju

Hi, I did performance test with grpc-js vs REST (using Koa), and found grpc-js significantly slower than Koa. Does anyone have the same experience?

Also grpc-js is much slower than grpc server implemented in Go.

The performance is : gRPC in Go > Koa > grpc-js. Any ideas why this is the case?

1 reply

Where communities thrive

People

Repo info

Activity