Deployment on EKS on

Architecture for EKS

Tue, 20 Feb 2024 00:00:00 +0000

The following architectural diagram illustrates the main components in the deployment of the product on EKS.

Component	Description
Ingress Controller	The Ingress controller acts as the single point of entry for the requests provided by a user.
Ingress rule	The Ingress rule routes the requests to the Classification service.
Classification pods	The classification service deployment contains the classification service pods. Based on the load of requests, the pods for the classification service will scale horizontally to meet the demand. New pods are created when a certain threshold is reached. The pods are fronted by a load balancer that directs requests from the Ingress controller. Based on the load and availability, the requests are distributed amongst the pods.
Service provider deployment	The Service deployment contain the images of the Presidio and the Roberta service providers. The requests sent by Classification service are received by these pods for the classification of data. Similar to the classification service deployment, these pods will scale horizontally to meet the request load demand.

Prerequisites

Tue, 20 Feb 2024 00:00:00 +0000

Ensure that the following points are completed before deploying the application.

The development package provided by Protegrity is obtained from the portal and extracted.
The following tools are required to install the product:
- AWS CLI: This must be configured with valid credentials. For more information, refer to Configuration and credentials precedence. The AWS account used must have the necessary permissions to create and manage EKS clusters and associated resources.
- kubectl: This is used to manage the Kubernetes cluster.

Configuring Environment Variables

Tue, 20 Feb 2024 00:00:00 +0000

Run the following steps to configure the environment variables.

Navigate to the eks directory.
Open the .env file and configure the following variables:

Variable	Description	Mandatory
CLUSTER_NAME	Name for the cluster. Ensure that a unique name is provided to the cluster.	No
CLASSIFICATION_DOCKER_IMAGE	Repository path where the docker image of Classification service is stored.	Yes
ROBERTA_DOCKER_IMAGE	Repository path where the docker image of Roberta service is stored.	Yes
PRESIDIO_DOCKER_IMAGE	Repository path where the docker image of Presidio service is stored.	Yes
SUBNET_IDS	Subnetwork IDs. Provide a comma-separated list. For example,`"subnet-YYYY\\,subnet-XXXXX"`	Yes
AWS_REGION	Region of AWS infrastructure.	Yes
CLUSTER_IAM_ROLE_ARN	IAM role ARN for the cluster. If left blank, CloudFormation will automatically create the role.	No
NODE_IAM_ROLE_ARN	IAM role ARN for the nodes. If left blank, CloudFormation will automatically create the role.	No

Save the changes.

Deploying the Cluster and the Application

Tue, 20 Feb 2024 00:00:00 +0000

Deploying the Application

Deploy the Data Discovery application on EKS.

Navigate to the eks directory.

cd /eks

Run the aws_deploy script.

./aws_deploy.sh

After the deployment is completed, the following message is displayed.

Service 'protegrity.classification-service is ready!
Your Ingress Controller is available at {DNS address}

Note the DNS address of the Ingress controller.

Verifying the Installation

Run the following cURL command to verify the installation.

curl -k https://{Ingress Controller DNS address}/readiness

curl -k -X POST https://{Ingress Controller DNS address}/pty/data-discovery/v1.0/classify -H 'Content-Type: text/plain' --data 'Jake lives in 35, Hampden Street'

The response generated will be similar to the following snippet.

Viewing the Logs

Tue, 20 Feb 2024 00:00:00 +0000

Viewing Logs

The application logs can be viewed using the following commands:

 kubectl logs classification-deployment-{version} -n protegrity -f
 kubectl logs roberta-provider-deployment-{version} -n protegrity -f
 kubectl logs presidio-provider-deployment-{version} -n protegrity -f

Note: The version of the images can be obtained using the following command: kubectl get pods -n <namespace-name>

Setting the Log Level

The log level can be updated in the value.yaml file.

 # Custom logging configuration for the application.
 #
 # EXAMPLE - default settings 
 # loggingConfig:
 # root:
 # level: DEBUG
 # loggers:
 # tornado.access:
 # level: WARNING
 #

To set the log level, run the following steps.

Removing the Application

Tue, 20 Feb 2024 00:00:00 +0000

To remove the cluster and application from EKS, run the following steps.

Navigate to the eks directory.

cd /eks

Run the aws_undeploy script.

./aws_undeploy.sh

Note: To remove the application completely, ensure that the images uploaded to the image repository are also deleted.

Mon, 01 Jan 0001 00:00:00 +0000

 ./aws_undeploy.sh

Mon, 01 Jan 0001 00:00:00 +0000

 ./aws_deploy.sh [cluster_name]

Mon, 01 Jan 0001 00:00:00 +0000

Navigate to the /deployment/eks directory.
Run the aws_undeploy.sh file.

Mon, 01 Jan 0001 00:00:00 +0000

Navigate to the /deployment/eks directory.
Run the aws_deploy.sh script.