cloud_project_1400

Docker microservices

here we have three web service that can use in docker or Separately :

• Auth
• DataAnalytic
• Global

all these services developed by golang

Auth service

authentication of users

Global service

fetch some data about games

Data Analytic service

draw charts that you want

Running

we offer two method for up and running these services.

• normal method with docker run
• fast method with docker-compose

but first we should build our own services

build a new image

we need a dockerfile , now we create it:

docker build --no-cache --build-arg SERVICE={{SERVICE_NAME}} -t {{SERVICE_NAME}}:latest  . 

we provide better way to reach image :

./build.sh {{SERVICE_NAME}}

a Dockerfile must provided behind of build.sh file

Dockerfile

Dockerfile contain following instructions :

FROM golang:1.17

this mean we want to have image with base golang version 1.17

ARG SERVICE=default
ENV SERVICEN=$SERVICE

get SERVICE variable that set in docker build command and set it to env variable

RUN git clone --depth 1 https://github.com/hofarah/cloud_project_1400.git /go/src/cloud_1400 --branch master

get files of repository

ENV GO111MODULE=on

we need set go modules to on for our build

WORKDIR /go/src/cloud_1400

set our workdir to /go/src/cloud_1400 that we set it in last step

RUN go mod tidy
RUN go mod download

tidy and download our dependancy libraries.

RUN cd services/$SERVICE && go build -o main .

build our service

EXPOSE 7575
CMD /go/src/cloud_1400/services/${SERVICEN##*/}/main

expose on port 7575 and set command that run when container start

now you have image of your service.

1. docker run method

Follow the instructions below :

$ docker run [OPTIONS] IMAGE[:TAG|@DIGEST] 

you can use -d option to run in background

2. docker-compose method

in this method we can run multiple container with one command. we can use docker-compose up -d. this command will automaticly run and manage the container we mentioned in a docker-compose.yml

it works simply and if you want to remove the containers you can use docker-compose down. it will stop and remove containers. docker-compose is like rum method but you can use it many times.

Redis cli

in this project we caching all data. so,there is no concern

Jaeger tracing

you can see your traces of services in http://${SERVER_IP}:16686

Prometheus

• get load request
• failed request
• success request
• threads
• memory allocations
• gc calling and ...

you can create graph and queries for your services. just go on http://${SERVER_IP}:9000

all configs are mutable ,you can edit configs and change ports.