PILOT LIGHT DR (RAAS)
Authored by Nithin Kumar Shetty
Introduction
Disaster recovery in the cloud ensures business continuity by minimizing downtime and data loss during failures. It leverages automation, scalability, and multi-region redundancy for quick recovery and high availability.
Customers often face challenges balancing costs and ensuring business continuity. They come to us with specific needs for customized disaster recovery solutions.
Below is one of the scenarios;
A Life Science company wanted to achieve regional DR with minimal cost:
That’s when they turned to us for help. After carefully evaluating their needs, we recommended the Pilot Light Disaster recovery strategy for their critical applications to fail over seamlessly within the mentioned RTO.
In this guide, we’ll take you through the step-by-step process of setting up a Disaster Recovery strategy that ensures that non-critical applications remain resilient and available, even in the event of a regional failure.
Technical approach:
Pre-requisites and technology Stack:
Get CloudifyOps’s stories in your inbox
Join Medium for free to get updates from this writer.
The below-mentioned AWS resources built using Terraform are prerequisites for the DR site, and hence, it has to be pre-provisioned.
- DR Network (VPC, Subnets, IGW, NAT, Route tables).
- RDS — Cross-Region Backup to DR region.
- Backend S3 bucket (This will be a copy of the existing prod backend S3, cross-regional replication enabled from prod to DR)
- Frontend S3 bucket (This will be a copy of the existing prod frontend bucket, cross-regional replication enabled from prod to DR)
- Secrets (Secrets manually entered in the DR region environment)
- ECR repositories (Once a build is done and the image is pushed to the production ECR image, cross-region replication is in place to us-west-2)
- IAM Roles with required policy.
Code Pipelines:
- IAC Pipeline
- DR ENV Restore Pipeline
- Primary Region Rollback Pipeline
Production Environment was brought down after notifying.
- The IAC pipeline was run to establish the infrastructure in the DR region.
- The DR environment restore codebuild job was run to install the addons and support applications inside the cluster, like redis, backend elasticsearch and after that deploys the application helm chart into the destination cluster.
During the DR, the following resources were built:
- Additional Private subnets for hosting the EKS cluster and Ingress are enabled through ALB.
- DR ENV Restoration code build project was released to configure the cluster and deploy the application (helm chart).
- Manually created secrets with the DR, as we had to update the postgres endpoint and the S3 bucket.
- Updated the CNAME records (app.*****.com & appstatic.*****.com), with the DNS of the ALB of the DR environment.
Testing and Validation:
- Verified that the latest application images were replicated from the production ECR repository to the DR repository, ensuring availability of up-to-date images.
- Ensured that DNS records (e.g., app.*****.com and appstatic.****.com) were successfully updated to point to the ALB in the DR environment.
- Validated that manually created secrets, including PostgreSQL endpoint and S3 bucket details, were correctly integrated into the DR environment.
- Restoration Code Deployment Testing: Executed tests to ensure that the DR environment restoration code successfully configured the EKS cluster and deployed the application using Helm charts.
- Functional Testing: The development team conducted comprehensive testing of application functionalities such as image uploading and processing in the DR environment.
- Test Success Validation: Successful execution of tests validated that the application could perform essential tasks seamlessly in the DR environment, ensuring business continuity.
Conclusion:
Recovery Region in a pilot light strategy cannot serve requests until additional steps are taken.
As the client approached for cost-effective DR, pilot light DR provides Business continuity within the agreed SLA.
CloudifyOps demonstrates expertise in crafting cost-effective disaster recovery strategies like Pilot Light for business continuity. By leveraging AWS services and automation, they ensure resilience within agreed SLAs. For tailored DR solutions that balance cost and recovery, contact CloudifyOps to safeguard your critical applications.
