Search This Blog

Showing posts with label Tanzu. Show all posts
Showing posts with label Tanzu. Show all posts

Tuesday, 16 June 2020

Deploying a Spring Boot application to Tanzu Application Service for Kubernetes using GitLab

In this demo I show how to deploy a simple Springboot application using GitLab pipeline onto Tanzu Application Service for Kubernetes (TAS4K8s).

Steps

Ensure you have Tanzu Application Service for Kubernetes (TAS4K8s) running as shown below
  
$ kapp list
Target cluster 'https://lemons.run.haas-236.pez.pivotal.io:8443' (nodes: a51852ac-e449-40ad-bde7-1beb18340854, 5+)

Apps in namespace 'default'

Name  Namespaces                                    Lcs   Lca
cf    (cluster),build-service,cf-blobstore,cf-db,   true  10d
      cf-system,cf-workloads,cf-workloads-staging,
      istio-system,kpack,metacontroller

Lcs: Last Change Successful
Lca: Last Change Age

1 apps

Succeeded

Ensure you have GitLab running. In this example it's installed on a Kubernetes cluster but it doesn't have to be. All that matters here is that GitLab can access the API endpoint of your TAS4K8s install
  
$ helm ls -A
NAME   NAMESPACE REVISION UPDATED                               STATUS   CHART        APP VERSION
gitlab gitlab    2        2020-05-15 13:22:15.470219 +1000 AEST deployed gitlab-3.3.4 12.10.5

1. First let's create a basic Springboot application with a simple RESTful endpoint as shown below. It's best to use the Spring Initializer to create this application. I simply used the web and lombok dependancies as shown below.

Note: Make sure you select java version 11.

Spring Initializer Web Interface


Using built in Spring Initializer in IntelliJ IDEA.


Here is my simple RESTful controller which simply output's todays date.
  
package com.example.demo;

import lombok.extern.slf4j.Slf4j;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RestController;

import java.util.Date;

@RestController
@Slf4j
public class FrontEnd {
    @GetMapping("/")
    public String index () {
        log.info("An INFO Message");
        return new Date().toString();
    }
}

2. Create an empty project in GitLab using the name "gitlab-TAS4K8s-boot-demo"



3. At this point this add our project files from step #1 above into the empty GitLab project repository. We do that as follows.

$ cd "existing project folder from step #1"
$ git init
$ git remote add origin http://gitlab.ci.run.haas-236.pez.pivotal.io/root/gitlab-tas4k8s-boot-demo.git
$ git add .
$ git commit -m "Initial commit"
$ git push -u origin master

Once done we now have out GitLab project repository with the files we created as part of the project setup


4. It's always worth running the code locally just to make sure it's working so if you like you can do that as follows

RUN:

$ ./mvnw spring-boot:run

CURL:

$ curl http://localhost:8080/
Tue Jun 16 10:46:26 AEST 2020

HTTPie:

papicella@papicella:~$
papicella@papicella:~$
papicella@papicella:~$ http :8080/
HTTP/1.1 200
Connection: keep-alive
Content-Length: 29
Content-Type: text/plain;charset=UTF-8
Date: Tue, 16 Jun 2020 00:46:40 GMT
Keep-Alive: timeout=60

Tue Jun 16 10:46:40 AEST 2020

5. Our GitLab project as no pipelines defined so let's create one as follows in the project root directory using the default pipeline name ".gitlab-ci.yml"

image: openjdk:11-jdk

stages:
  - build
  - deploy

build:
  stage: build
  script: ./mvnw package
  artifacts:
    paths:
      - target/demo-0.0.1-SNAPSHOT.jar

production:
  stage: deploy
  script:
  - curl --location "https://cli.run.pivotal.io/stable?release=linux64-binary&source=github" | tar zx
  - ./cf api https://api.system.run.haas-236.pez.pivotal.io --skip-ssl-validation
  - ./cf auth $CF_USERNAME $CF_PASSWORD
  - ./cf target -o apples-org -s development
  - ./cf push -f manifest.yaml
  only:
  - master


Note: We have not defined any tests in our pipeline which we should do but we haven't written any in this example.

6. For this pipeline to work we will need to do the following

- Add a manifest.yaml file in the project root to deploy our simple Springboot RESTful application

---
applications:
  - name: gitlab-TAS4K8s-boot-demo
    memory: 1024M
    instances: 1
    path: ./target/demo-0.0.1-SNAPSHOT.jar

- Alter the API endpoint to match your TAS4K8s endpoint

- ./cf api https://api.system.run.haas-236.pez.pivotal.io --skip-ssl-validation

- Alter the target to use your ORG and SPACE within TAs4K8s.

- ./cf target -o apples-org -s development

This command shows you what your current CF CLI is targeted to so you can ensure you edit it with correct details
  
$ cf target
api endpoint:   https://api.system.run.haas-236.pez.pivotal.io
api version:    2.150.0
user:           pas
org:            apples-org
space:          development

7. For the ".gitlab-ci.yml" to work we need to define two ENV variables for our username and password. Those two are as follows which is our login credentials to TAS4K8s

  • CF_USERNAME 
  • CF_PASSWORD

To do that we need to navigate to "Project Settings -> CI/CD - Variables" and fill in the appropriate details as shown below



8. Now let's add the two new files using git , add a commit message and push the changes

$ git add .gitlab-ci.yml
$ git add manifest.yaml
$ git commit -m "add pipeline configuration"
$ git push -u origin master

9. Navigate to GitLab UI "CI/CD -> Pipelines" and we should see our pipeline starting to run








10. If everything went well!!!



11. Finally our application will be deployed to TAS4K8s as shown below
  
$ cf apps
Getting apps in org apples-org / space development as pas...
OK

name                       requested state   instances   memory   disk   urls
gitlab-TAS4K8s-boot-demo   started           1/1         1G       1G     gitlab-tas4k8s-boot-demo.apps.system.run.haas-236.pez.pivotal.io
gitlab-tas4k8s-demo        started           1/1         1G       1G     gitlab-tas4k8s-demo.apps.system.run.haas-236.pez.pivotal.io
test-node-app              started           1/1         1G       1G     test-node-app.apps.system.run.haas-236.pez.pivotal.io

$ cf app gitlab-TAS4K8s-boot-demo
Showing health and status for app gitlab-TAS4K8s-boot-demo in org apples-org / space development as pas...

name:                gitlab-TAS4K8s-boot-demo
requested state:     started
isolation segment:   placeholder
routes:              gitlab-tas4k8s-boot-demo.apps.system.run.haas-236.pez.pivotal.io
last uploaded:       Tue 16 Jun 11:29:03 AEST 2020
stack:
buildpacks:

type:           web
instances:      1/1
memory usage:   1024M
     state     since                  cpu    memory         disk      details
#0   running   2020-06-16T01:29:16Z   0.1%   118.2M of 1G   0 of 1G

12. Access it as follows.

$ http http://gitlab-tas4k8s-boot-demo.apps.system.run.haas-236.pez.pivotal.io
HTTP/1.1 200 OK
content-length: 28
content-type: text/plain;charset=UTF-8
date: Tue, 16 Jun 2020 01:35:28 GMT
server: istio-envoy
x-envoy-upstream-service-time: 198

Tue Jun 16 01:35:28 GMT 2020

Of course if you wanted to create an API like service you could use the source code at this repo rather then the simple demo shown here using OpenAPI.

https://github.com/papicella/spring-book-service



More Information

Download TAS4K8s
https://network.pivotal.io/products/tas-for-kubernetes/

GitLab
https://about.gitlab.com/

Tuesday, 11 February 2020

Taking VMware Tanzu Mission Control for a test drive

You may or may not have heard of Tanzu Mission Control (TMC) part of the new VMware Tanzu offering which will help you build, run and manage modern apps. To find out more about Tanzu Mission Control here is the Blog link on that.

https://blogs.vmware.com/cloudnative/2019/08/26/vmware-tanzu-mission-control/

In this blog I show you how easily you can use TMC to monitor your existing k8s clusters. Keep in mind TMC can also create k8s clusters for you but here we will use the "Attach Cluster" part of TMC. Demo as follows

1. Of course you will need access account on TMC which for this demo I already have. Once logged in you will see a home screen as follows



2. In the right hand corner there is a "Attach Cluster" button click this to attach an existing cluster to TMC. Enter some cluster details , in this case I am attaching to a k8s cluster on GKE and giving it a name "pas-gke-cluster".


3. Click the "Register" button which takes you to a screen which allows you to install the VMware Tanzu Mission Control agent. This is simply done by using "kubectl apply ..." on your k8s cluster which allows an agent to communicate back to TMC itself. Everything is created in a namespace called "vmware-system-tmc"



4. Once you have run the "kubectl apply .." on your cluster you can verify the status of the pods and other components installed as follows

$ kubectl get all --namespace=vmware-system-tmc

Or you could just check the status of the various pods as shown below and assume everything else was created ok
  
$ kubectl get pods --namespace=vmware-system-tmc
NAME                                     READY   STATUS      RESTARTS   AGE
agent-updater-67bb5bb9c6-khfwh           1/1     Running     0          74m
agentupdater-workload-1581383460-5dsx9   0/1     Completed   0          59s
data-protection-657d8bf96c-v627g         1/1     Running     0          73m
extension-manager-857d46c6c-zfzbj        1/1     Running     0          74m
extension-updater-6ddd9858cf-lr88r       1/1     Running     0          74m
inspection-extension-789bb48b6-mnlqj     1/1     Running     0          73m
intent-agent-cfb49d788-cq8tk             1/1     Running     0          73m
policy-sync-extension-686c757989-jftjc   1/1     Running     0          73m
policy-webhook-5cdc7b87dd-8shlp          1/1     Running     0          73m
policy-webhook-5cdc7b87dd-fzz6s          1/1     Running     0          73m
sync-agent-84bd6c7bf7-rtzcn              1/1     Running     0          73m

5. Now at this point click on "Verify Connection" button to confirm the agent in your k8s cluster is able to communicate with TMC

6. Now let's search for out cluster on the "Clusters" page as shown below



7. Click on "pas-gke-cluster" and you will be taken to an Overview page as shown below. Ensure all green tick boxes are in place this may take a few minutes so refresh the page as needed



8. So this being an empty cluster I will create a deployment with 2 pods so we can see how TMC shows this workload in the UI. These "kubectl commands" should work on any cluster as the image is on Docker Hub

$ kubectl run pbs-deploy --image=pasapples/pbs-demo-image --replicas=2 --port=8080
$ kubectl expose deployment pbs-deploy --type=LoadBalancer --port=80 --target-port=8080 --name=pbs-demo-service

9. Test the workload (Although this isn't really required)

$ echo "http://`kubectl get svc pbs-demo-service -o jsonpath='{.status.loadBalancer.ingress[0].ip}'`/customers/1"
http://104.197.202.165/customers/1

$ http http://104.197.202.165/customers/1
HTTP/1.1 200
Content-Type: application/hal+json;charset=UTF-8
Date: Tue, 11 Feb 2020 01:43:26 GMT
Transfer-Encoding: chunked

{
    "_links": {
        "customer": {
            "href": "http://104.197.202.165/customers/1"
        },
        "self": {
            "href": "http://104.197.202.165/customers/1"
        }
    },
    "name": "pas",
    "status": "active"
}

10. Back on the TMC UI click on workloads. You should see our deployment as per below


11. Click on the deployment "pbs-deploy" to see the status of the pods created as part of the deployment replica set plus the YAML of the deployment itself


12. Of course this is just scratching the surface but from the other tabs you can see the cluster nodes, namespaces and other information as required not just for your workloads but also for the cluster itself




One thing to note here is when I attach a cluster as shown in this demo the life cycle of the cluster, for example upgrades, can't be managed / performed by TMC. In the next post I will show how "Create Cluster" will actually be able to control the life cycle of the cluster as well as this time TMC will actually create the cluster for us.

Stay tuned!!!

More Information

Introducing VMware Tanzu Mission Control to Bring Order to Cluster Chaos
https://blogs.vmware.com/cloudnative/2019/08/26/vmware-tanzu-mission-control/

VMware Tanzu Mission Control
https://cloud.vmware.com/tanzu-mission-control