Deployment example for Teamwork Cloud with services on Kubernetes

This deployment example is not intended for production use. Use it for testing purposes only.

This chapter provides information on how to deploy the product for testing purposes. Keep in mind that the guidelines provided below are not considered the best practices and do not take into account security settings and password encryption. This deployment has been tested with a specific Kubernetes version, CNI, container runtime, and OS. If you use other solutions, use them at your own risk.

Prerequisites

Kubernetes cluster (for more information, refer to https://kubernetes.io/):
- 1 virtual machine used as the control plane node and 3 nodes used as worker nodes
- Control plane node: 4 CPU cores, 8GB RAM
- Worker node: 16 CPU cores, at least 64GB RAM (depends on your deployment size)
Kubernetes version: v1.28
Container runtime: Containerd (for more information, refer to https://containerd.io/)
CNI: Calico
OS: Red Hat Enterprise Linux release 8.7 (Ootpa)
Load Balancer: MetalLB
Most cloud providers do not support MetalLB.
IP: 1 IP address with the DNS A record that points to this IP (IP from your network that is not used by your cluster nodes).
Ingress: ingress-nginx (with sticky session support).
Note that there are two nginx ingresses: ingress-nginx is maintained by the Kubernetes community, while nginx-ingress is maintained by F5 inc. The free F5 inc. version of nginx-ingress does not support “sticky sessions” which are required. Therefore, make sure you use ingress-nginx.
A Host with Docker or Podman to build images and push them to the image registry.
An image registry (repository) such as Harbor, Sonatype Nexus, Gitlab, or Docker registry server to push and pull the images you will build.
The applications needed to run Teamwork Cloud and services:
- Zookeeper 3.9.2 (for more information, refer to https://zookeeper.apache.org/)
- Apache Artemis MQ 2.32.0 (for more information, refer to https://activemq.apache.org/components/artemis/)
- Keda 2.13.2 CRD (deployed with the Simulation web application)
- MetalLB
- Ingress-Nginx
- Cassandra 4.1.4

The following figure displays the environment used in this deployment example.

In this environment, the K8S cluster admin host (Linux virtual machine) is used to build images, run the image repository, and manage the Kubernetes cluster.

In the workflow below:

admin.example.com refers to the K8S cluster admin host and should be changed to your K8S cluster admin FQDN.
webapp.example.com refers to the domain name through which Web Application Platform will be accessible.
All the steps are performed on the K8S cluster admin host.

To deploy the product for testing purposes

Install kubectl as described at https://kubernetes.io/docs/tasks/tools/install-kubectl-linux/.
Install the helm to the K8S cluster admin host as described at https://helm.sh/docs/intro/install.
Execute the following command to copy cluster config files to the K8S cluster admin host:
```
scp -r <user@control_plane_address>:.kube ~
Edit ~/.kube/config
```
Replace the localhost IP address with your control plane IP address or hostname, e.g., https://127.0.0.1:6443.
Prepare the required files and directories:
1. Download and extract the webapp-charts.zip file.
2. Download the installation package for your product version. For more information, refer to Downloading installation files.
3. Extract the packaged files, and inside the webapp-charts root directory (where you extracted the webapp-charts.zip file), extract the twcloud_<version>_no_install_linux64.zip file. After extracting the files, you should have the following folder structure: <webapp-charts_root>/CATIANoMagicServices/TeamworkCloud/.
  You will find all files mentioned in this chapter in twcloud_<version>_no_install_linux64.zip or on the software download website (except for Teamwork Cloud keystore and SSL certificate):
  - The <webapp-charts_root>/CATIANoMagicServices/WebAppPlatform/webapps directory contains application .war files.
  - The <webapp-charts_root>/CATIANoMagicServices/WebAppPlatform/shared/conf directory contains other configuration files.
  - The <webapp-charts_root>/CATIANoMagicServices/WebAppPlatform/conf directory contains server.xml.
  - The <webapp-charts_root>/CATIANoMagicServices/WebAppPlatform/shared/conf/data directory contains files for Cameo Collaborator and document-exporter.
  - The <webapp-charts_root>/CATIANoMagicServices/TeamworkCloud/configuration directory contains Teamwork Cloud configuration files.
  - The <webapp-charts_root>/CATIANoMagicServices/Utilities/Cassandra contains the .jar file for Cassandra's communication with Zookeeper.
Use the same SSL certificate and key that were used for the Teamwork Cloud server. You can copy the key and certificate files from the Teamwork Cloud server.
If you use only the .p12 file on the Teamwork Cloud server, execute the following command to extract the certificate and key from the .p12 file:
```
openssl pkcs12 -in INFILE.p12 -out tls.crt -nokeys
openssl pkcs12 -in INFILE.p12 -out tls.key -nodes –nocerts
```
The tls.crt and tls.key files will be used for ingress-nginx, so file names in ingress-nginx must be exactly the same.

Execute the following commands to prepare the webapp-charts root directory:

cd <webapp-charts_root>/webapp-charts
mkdir -p configs/{auth,cassandra,ssl,tomcat,twcloud,wap} configs/twcloud/ssl

Copy all files listed below into their directories:

auth
authserver.properties
cassandra
cassandra.yaml logback.xml
ssl
tls.crt tls.key
tomcat
catalina.properties server.xml
wap
revision.txt log4j2.properties webappplatform.properties
twcloud
application.conf logback.xml jvm.options
twcloud/ssl
keystore.p12 (from the Teamwork Cloud server) twamworkcloud.crt

Execute the following command to copy the generated Teamwork Cloud certificates:
```
cp keystore.p12 teamworkcloud.crt configs/twcloud/ssl
```

Edit the application.conf file as shown below:

sed -i 's/contact-points.*/contact-points = [${?CASSANDRA_SEED0}][${?CASSANDRA_SEED1}][${?CASSANDRA_SEED2}]/g' application.conf
sed -i "s/local.size.*/local.size = 3/g" application.conf
sed -i "s/remote.size.*/remote.size = 3/g" application.conf
sed -i "s/replication-factor.*/replication-factor = 3/g" application.conf

Edit the jvm.options file as shown below:

sed -i 's/-Xmx.*/-Xmx=${TWC_XMX}/g' jvm.options

Copy the application.conf, jvm.options, and logback.xml files to the configs/twcloud directory.
Execute the following command to copy the extracted Teamwork Cloud certificates:
```
cp tls.crt tls.key configs/ssl
```
Go to https://archive.apache.org/dist/cassandra/4.1.4/apache-cassandra-4.1.4-bin.tar.gz and download the Cassandra configuration file.
Execute the following command to extract the Cassandra .tar file:
```
tar -xzf apache-cassandra-4.1.4-bin.tar.gz
```

In the cassandra.yaml file edit Cassandra settings as shown below.

sed -i "s/# commitlog_total_space:.*/commitlog_total_space: 8192Mib/g" cassandra.yaml
sed -i "s/commitlog_segment_size:.*/commitlog_segment_size: 192Mib/g" cassandra.yaml
sed -i "s/read_request_timeout:.*/read_request_timeout: 1800000ms/g" cassandra.yaml
sed -i "s/range_request_timeout:.*/range_request_timeout: 1800000ms/g" cassandra.yaml
sed -i "s/^write_request_timeout:.*/write_request_timeout: 1800000ms/g" cassandra.yaml
sed -i "s/cas_contention_timeout:.*/cas_contention_timeout: 1000ms/g" cassandra.yaml
sed -i "s/truncate_request_timeout:.*/truncate_request_timeout: 1800000ms/g" cassandra.yaml
sed -i "s/request_timeout:.*/request_timeout: 1800000ms/g" cassandra.yaml
sed -i "s/batch_size_warn_threshold:.*/batch_size_warn_threshold: 3000Kib/g" cassandra.yaml
sed -i "s/batch_size_fail_threshold:.*/batch_size_fail_threshold: 5000Kib/g" cassandra.yaml
sed -i "s/# internode_application_send_queue_reserve_endpoint_capacity:.*/internode_application_send_queue_reserve_endpoint_capacity: 512MiB/g" cassandra.yaml
sed -i "s/# internode_application_receive_queue_reserve_endpoint_capacity:.*/internode_application_receive_queue_reserve_endpoint_capacity: 512MiB/g" cassandra.yaml

Execute the following command to copy files:

cp apache-cassandra-4.1.4/conf/cassandra.yaml logback.xml configs/cassandra

Execute the following command to create a directory for each application you are going to use:

mkdir -p imagebuild/{admin-console,authentication,collaborator,document-exporter,oslc,reports,resources,resource-usage-map,simulation,webapp,twc}

Copy all files listed below to their directories (except the artemis directory). You will find the Docker files for each service in dockerfiles.zip. After placing the Docker files in their appropriate directories, rename them to Dockerfile as shown below:

admin-console
admin.war Dockerfile ROOT
authentication
authentication.war Dockerfile ROOT
collaborator
collaborator.war Dockerfile data (The directory is located in <install_root>/WebAppPlatform/shared/conf/. Before copying, remove the document-exporter directory from the data directory.) ROOT
document-exporter
document-exporter.war Dockerfile data (This directory is located in <webapp-charts_root>/CATIANoMagicServices/WebAppPlatform/shared/conf/. Remove the collaborator directory from the data directory.) ROOT
element-chooser
element-chooser.war Dockerfile ROOT
oslc
oslc.war Dockerfile ROOT
reports
reports.war Dockerfile ROOT
resource-usage-map
resource-usage-map.war Dockerfile ROOT
resources
resources.war Dockerfile ROOT
simulation
simulation.war Dockerfile ROOT
webapp
webapp.war Dockerfile ROOT

Execute the following command to start the local Docker registry:
```
docker run -d -p 5000:5000 --restart=always --name registry registry:2
```
In this case, a local Docker registry is run on the K8S cluster admin host. It uses port 5000, so make sure this port is open on your host firewall and accessible to your Kubernetes cluster.
This local Docker registry should be used for testing purposes only and is not considered production-ready. In the production environment, run the image registry with TLS and authentication.
Add your repo URL (FQDN of the K8S cluster admin host) to the Docker /etc/docker/daemon.json file.
If the daemon.json file does not exist, create it. Assuming there are no other settings in the file, it should have the following content:
```
{
"insecure-registries": ["admin.example.com:5000"]
}
```
Execute the following command to restart the Docker service:
```
sudo systemctl restart docker
```
Execute the commands below for each service that you want to build images for inside their own directories, as shown in the examples in steps 24a, 24b, and 24c.
```
docker build -f Dockerfile -t {APP_NAME} .
docker tag {APP_NAME}:{VERSION} {IMAGE_REPO_URL}:5000/{APP_NAME}:{VERSION}
docker push {IMAGE_REPO_URL}/{APP_NAME}:{VERSION}
```
- {IMAGE_REPO_URL} - your image repository URL. It can differ depending on the image registry provider.
- {APP_NAME} - application name.
- {VERSION} - version to be used for the tag.
1. The following command example builds images for Admin Console:
```
cd imagebuild/admin-console
docker build -f Dockerfile -t admin .
```
2. The following command example tags the built image:
```
docker tag admin: latest admin.example.com:5000/admin:latest
```
3. The following command example pushes the image to the registry:
```
docker push admin.example.com:5000/admin:latest
```
To be able to pull images to your Kubernetes cluster nodes, add your image repository to the Containerd configuration on all cluster nodes.

Edit the /etc/containerd/config.toml file by adding the following lines to the "mirrors" part:

[plugins."io.containerd.grpc.v1.cri".registry.mirrors."admin.exmple.com:5000"]
              endpoint = ["http://admin.example.com:5000"]

      [plugins."io.containerd.grpc.v1.cri".registry.configs." admin.example.com"]
        [plugins."io.containerd.grpc.v1.cri".registry.configs." admin.example.com".tls]
         insecure_skip_verify = true

Keep in mind that alignment is very important, as shown in the example below.

Execute the following command to restart Containerd services on all cluster nodes:
```
sudo systemctl restart containerd
```

Execute the following commands to add dependency repos:

helm repo add kedacore https://kedacore.github.io/charts
helm repo add ingress-nginx https://kubernetes.github.io/ingress-nginx
helm repo add bitnami https://charts.bitnami.com/bitnami
helm repo update

cd ${webapp-charts_directory}

Execute the following command to update chart dependencies:
```
helm dependency update
```
Execute the command below to install Custom Resource Definitions (CRD). This will create all services and deployments.
```
helm install keda kedacore/keda --namespace keda --create-namespace --wait --version 2.13.2
```
- You can check the status of the services and pods with this command: kubectl get all
Keda and MetalLB should be added as separate resources.
Since CRDs (being a globally shared resource) are fragile, you have to assume that it is shared across multiple namespaces and groups of users once a CRD is installed. For that reason, installing, modifying, and deleting CRDs is a process that has ramifications for all users and systems of that cluster.

Execute the following command to deploy MetalLB:

helm install metallb bitnami/metallb --namespace metallb-system --create-namespace --wait

To check the MetalLB deployment status, execute the following command on the control plane node or the K8S cluster admin host:
```
kubectl get -n metallb-system all
```
You should see an output similar to this:
If you see that the MetalLB status is 'Running,' but READY is 0/1, give it some time to start and repeat the command.
Create configuration files for MetalLB:
Make sure to complete all the substeps of this step to create and apply both configurations. Otherwise, the product will not work properly.
1. Edit the metalLB/metallb_ipadresspool.yaml file as shown below:
```
apiVersion: metallb.io/v1beta1
kind: IPAddressPool
metadata:
  name: first-pool
  namespace: metallb-system
spec:
  addresses:
  - 192.168.1.1/32
  #- 192.168.10.0/24
  #- 192.168.9.1-192.168.9.5
  #- fc00:f853:0ccd:e799::/124
```
  - Your IT department should give you a reserved IP address or a range of IP addresses (depending on your needs) with a DNS A record that points to this IP address.
  - Configuration files should be formatted in .yaml.
2. Edit the metalLB/l2advertisement.yaml file as shown below. You can change the metadata name as needed. Keep in mind that the ipAddressPool name should match the specifications.
```
apiVersion: metallb.io/v1beta1
kind: L2Advertisement
metadata:
  name: example
  namespace: metallb-system
spec:
  ipAddressPools:
  - first-pool
```
3. Execute the following command to apply the MetalLB configuration:
```
kubectl apply -f metalLB/metallb_ipaddresspool.yaml
```
4. Execute the following command to check the applied configuration:
```
kubectl describe ipaddresspools.metallb.io first-pool -n metallb-system
```
5. Execute the following command to create the MetalLB advertisement:
```
kubectl apply -f metalLB/l2advertisement.yaml
```
6. Execute the following command to check the advertisement configuration:
```
kubectl describe l2advertisements.metallb.io example -n metallb-system
```
Find the Values.yaml file in the parent helm chart and provide the values of the parameters in the file to enable or disable specific applications or parts of the configuration.
Do one of the following:
- Execute this command in the helm parent chart directory to deploy services to the default namespace:
```
helm install webapp .
```
- Execute this command in the helm parent chart directory to create a namespace and deploy services in this namespace:
```
helm install webapp . --namespace=wap --create-namespace --wait
```
  This helm chart includes Zookeeper and Ingress-nginx as dependencies. They will be deployed automatically.
After all web applications are deployed, execute the following command to check their status:
```
kubectl get all
```
- All pods should be READY:1/1, STATUS:Running.
- The ingress-nginx service should be Type:LoadBalancer.
- EXTERNAL-IP should match the address that you provided in the MetalLB configuration.
If pods do not run, check for problems by executing this command:
```
kubectl describe pod pod_name
```

Execute the following command to check Ingress rules:

kubectl describe ingress

You should get an output similar to this:

Name:             ingress.resource
Namespace:        default
Address:          **.**.***.***
Default backend:  default-http-backend:80 (<error: endpoints "default-http-backend" not found>)
TLS:
  webapp-tls-secret terminates nm-wap10639.dsone.3ds.com
Rules:
  Host                       Path  Backends
  ----                       ----  --------
  nm-wap10639.dsone.3ds.com
                             /admin                webapp-adminconsole:8443 (10.233.105.39:8443)
                             /authentication       webapp-authentication:8443 (10.233.88.69:8443)
                             /collaborator         webapp-collaborator:8443 (10.233.88.75:8443)
                             /document-exporter    webapp-docexporter:8443 (10.233.105.22:8443)
                             /oslc                 webapp-oslc:8443 (10.233.105.23:8443)
                             /reports              webapp-reports:8443 (10.233.73.246:8443)
                             /resources            webapp-resources:8443 (10.233.88.67:8443)
                             /resource-usage-map   webapp-rum:8443 (10.233.105.40:8443)
                             /simulation           webapp-simulation:8443 (10.233.88.123:8443)
                             /webapp               webapp-webapp:8443 (10.233.105.29:8443)
Annotations:                 meta.helm.sh/release-name: webapp
                             meta.helm.sh/release-namespace: default
                             nginx.ingress.kubernetes.io/affinity: cookie
                             nginx.ingress.kubernetes.io/affinity-canary-behavior: sticky
                             nginx.ingress.kubernetes.io/affinity-mode: persistent
                             nginx.ingress.kubernetes.io/backend-protocol: https
                             nginx.ingress.kubernetes.io/proxy-body-size: 100m
                             nginx.ingress.kubernetes.io/proxy-connect-timeout: 600
                             nginx.ingress.kubernetes.io/proxy-read-timeout: 600
                             nginx.ingress.kubernetes.io/proxy-send-timeout: 600
                             nginx.ingress.kubernetes.io/send-timeout: 600
                             nginx.ingress.kubernetes.io/session-cookie-name: COOKIE
Events:                      <none>

Test the product deployment:
1. In an internet browser, go to domain_name/webapp (the DNS A record that points to the Ingress external IP).
  If you added a DNS record to bind the domain name to the IP address, you can use the domain name instead of the IP. The browser should show a warning because of the self-signed certificate. Accept it to proceed, and you will be redirected to the Teamwork Cloud Authentication web page.
2. Log in with your credentials, and you will be redirected back to Web Application Platform.

Content

Space Tools

In this environment, the K8S cluster admin host (Linux virtual machine) is used to build images, run the image repository, and manage the Kubernetes cluster.

Content

Space Tools

Breadcrumbs

Deployment example for Teamwork Cloud with services on Kubernetes

In this environment, the K8S cluster admin host (Linux virtual machine) is used to build images, run the image repository, and manage the Kubernetes cluster.