Deploying a SpringBoot App with an application.properties using a docker-composer.yml file and limiting resources

We saw how to deploy a Spring Boot application with its application.properties plus a .p12 key using  an SSL certificate to secure Apache Tomcat or Netty in here https://springboot-vuejs-reactjs.blogspot.com/2019/12/springboot-with-applicationproperties.html

now let's do the same thing but using a docker-composer.yml with volumes instead of using images
so that this is going to give us the advantage about updating the .jar application and restarting the container without rebuilding the whole image.

let's create a folder in which you are going to have your .jar app, the application.properties and .p12 key to secure with SSL your SpringBoot app and any other files you need to run your application,
as in the previous post if you need a mini SpringBoot app to test it, grab this https://github.com/juandavidmarin368/SpringBoot-Docker

I've got a domain name with the one I do the tests with SSL and it is http://sslbackendtest.xyz/
I've created a file inside that folder called service_one.yml and this is what let run the java app on a docker container

javaapp:   image: openjdk:8   container_name: service_one   ports:       - '7075:7075'   volumes:       - /Data/service_1:/Data/service_1   working_dir: /Data/service_1   command: "java -jar PersisterDataDocker-0.0.1-SNAPSHOT.jar"

.

this is an application.properties to have SSL on SpringBoot

server.port:7075 server.ssl.enabled=true security.require-ssl=true server.ssl.key-store=sslbackendtest_xyz.p12 server.ssl.key-store-password=2k$C8=HJDHs4qQ/< server.ssl.key-alias=sslbackendtest.xyz server.ssl.key-password=2k$C8=HJDHs4qQ/< spring.datasource.driver = com.mysql.jdbc.Driver spring.datasource.url = jdbc:mysql://172.30.0.1:3306/docker?serverTimezone=UTC spring.datasource.username = datbaseUser spring.datasource.password = TjH=yrt/2-fgdS2454- spring.datasource.driver-class-name=com.mysql.jdbc.Driver

now to run it and to see in action on the console in order to check if everything worked OK, run this 

docker-compose -f service_one.yml up

now to run int in Detached mode or in the background run this:

docker-compose -f service_one.yml up -d

to see the containers running docker container ls

if you've got a new .jar version of the same app, just replace in the host folder the jar by the new one

and run docker container restart service_one 

service_one was the name given to the container if you need to stop it run

docker container stop service_one

LIMITING RESOURCES TO THE DOCKER CONTAINER

what about if you need to give it limited resources to the container because of many reasons, 

for instance what about if your application starts taking the whole ram memory and the whole processors you've got?, it would get the whole OS host down, and to avoid that if we've identified which services are taking a lot of resources by limiting their resources.

I am using Docker-compose 3 for the purpose of the example and it is as easy as adding  this block of
instructions

deploy:     resources:         limits:           cpus: "0.4"           memory: 400M

in which I am saying the application is going to use 40% of the whole CPUs and 400Mb ram memory

version: '3' services: java:   image: openjdk:8   container_name: service_one   deploy:     resources:         limits:           cpus: "0.4"           memory: 400M   ports:       - '7075:7075'   volumes:       - /Data/service_1:/Data/service_1   working_dir: /Data/service_1   command: "java -jar PersisterDataDocker-0.0.1-SNAPSHOT.jar"

now let's test it in Linux Centos 7 as the host OS

1: let's run the docker-compose  /usr/local/bin/docker-compose --compatibility -f service_one.yml up
2: let's open 2 more terminals on the host OS
3: in one of the terminals let's go to the container shell with this command
docker exec -it --user root service_one bash we are now logged in as root into the container service_one 
once we are there let's run apt-get update && apt-get install stress and after it has ended up let's run 

stress --cpu 8 --io 2 --vm 1 --vm-bytes 300mb 
with the last instruction, we are using the binary stress with the one we are simulating the resources of the CPU and ram memory in this case we are saying to it, take the whole permitted CPU and around 300mb
4: let's open another terminal in the host OS and let's run yum install htop after that run htop

We can see now how the whole CPU is being used around 40% plus taking approximately the ram memory we allow to use it