echoserver

A grpc and http echo server returning connection and server information for debugging purpose. 🏓

Features

• supports grpc and http
• supports specifying a delay in the request
• rich information in the response, e.g.
  • version: useful in testing deployment roll out
  • server hostname: useful in testing load balancing
  • client ip and port
  • request host / authority header etc
• server request logs
• server graceful shutdown

Usage

Use docker to run the echo server, by default it listens on ports

• 8080: http
• 9090: grpc

docker run -p 8080:8080 -p 9090:9090 lapwingcloud/echoserver

request params

For both grpc and http it supports 2 parameters in the request. Note for http it needs to be passed in the request body as json.

• delaySeconds: if specified, the request will sleep the number of delaySeconds before returning the response
• payload: if specified, the server will add the same field with the same value in the response

example http request

$ curl -d '{"delaySeconds":0.1, "payload":"hello"}' localhost:8080
{
  "timestamp": "2024-02-04T01:57:36.115139978Z",
  "version": "dev",
  "hostname": "02f3d12d7951",
  "remoteIp": "172.17.0.1",
  "remotePort": 47082,
  "requestId": "7746d5b4-6ce8-4d7f-9a2c-833bbcffe9f9",
  "requestHost": "localhost:8080",
  "requestMethod": "POST",
  "requestPath": "/",
  "requestQuery": "",
  "requestTime": 0.100276072,
  "userAgent": "curl/7.81.0",
  "delaySeconds": 0.1,
  "payload": "hello"
}

example grpc request

$ grpcurl -plaintext -d '{"delaySeconds":0.1, "payload":"hello"}' localhost:9090 echo.Echo/Ping
{
  "timestamp": "2024-02-04T02:06:52.025236095Z",
  "hostname": "02f3d12d7951",
  "version": "dev",
  "remoteIp": "172.17.0.1",
  "remotePort": 35962,
  "requestId": "d0baf289-3ddd-4051-b67c-36df9549870f",
  "authority": "localhost:9090",
  "requestMethod": "/echo.Echo/Ping",
  "requestTime": 0.100468275,
  "userAgent": "grpcurl/dev-build (no version set) grpc-go/1.57.0",
  "delaySeconds": 0.1,
  "payload": "hello"
}

cli flags

$ docker run lapwingcloud/echoserver echoserver -h
Usage of echoserver:
  -grpc-bind string
        The grpc server listen address (default ":9090")
  -http-bind string
        The http server listen address (default ":8080")
  -log-format string
        The log format (text, json) (default "json")

Example

Deploy to Kubernetes

Here is an example kubernetes manifest that you can deploy to any cluster.

prerequisites

• the kubernetes cluster has an ingress controller (e.g. ingress-nginx)
• (optional) you already have a valid TLS certificate for the domain you want the echoserver to serve on

setup

First you need to replace the domain name in the example/deploy-k8s.yaml with your own domain.

Then run the kubectl commands below to create a namespace and create the k8s deployment, service, ingress objects.

kubectl create echoserver
kubectl -n echoserver apply -f example/deploy-k8s.yaml

Optionally, if you have a TLS certificate and the private key, create the TLS secret in kubernetes so the ingress controller can use it to serve https and grpc over TLS.

Note you can obtain free TLS certificates via tools like certbot or cert-manager using letsencrypt.

kubectl -n echoserver create secret tls echoserver-tls --cert fullchain.pem --key privkey.pem

Now the echoserver should be accessible from the domain you setup, e.g.

curl https://echoserver-http.lapwingcloud.com
grpcurl echoserver-grpc.lapwingcloud.com:443 echo.Echo/Ping

If you didn't setup the TLS, just add the insecure flags to ignore invalid certificate.

curl --insecure https://echoserver-http.lapwingcloud.com
grpcurl -insecure echoserver-grpc.lapwingcloud.com:443 echo.Echo/Ping

Note, nginx doesn't support grpc over plaintext, so you cannot use grpcurl -plaintext to access the ingress domain.