5/14/2024 • VirtueCloud

In the ever-evolving landscape of AWS migrations, every project presents its own set of hurdles and lessons. Recently, our team embarked on a significant migration endeavor, transitioning a client's production server from an old AWS account to a new one. This meticulous process involved moving various services like EC2, Cognito, SQS, S3, and, crucially, RDS.
As we progressed through the migration steps guided by AWS documentation and our DevOps expertise,we successfully migrated all the resources from the old account and tested all the system is in place or not.
Everything was going smoothly post-migration until the production database abruptly ceased functioning two months later. Upon investigation, we discovered that the RDS was indicating "inaccessible-encryption-credentials" as its status.
Upon attempting to resolve the “inaccessible-encryption-credentials” issue, we discovered that the encryption key shared from the old account was no longer accessible due to the deactivation of the account by the vendor. This left our RDS stranded, unable to access the necessary encryption credentials, and consequently rendered it inoperable.
Despite our diligent efforts, including seeking assistance from AWS support, we faced limitations in reinstating the encryption key or transferring its ownership to the new account due to AWS's stringent security policies. With limited options, we had to resort to reactivating the old AWS account with the vendor's cooperation. Upon reactivation of the old account and resharing the encryption key with the new account, the RDS was successfully revived, and normal operations resumed.
This experience shed light on crucial gaps in our understanding of encryption key management post-migration. It highlighted the need for proactive key management, clear documentation, effective vendor communication, and continuous learning.
While our journey to migrate the production server was largely successful, the encryption key hiccup underscored the importance of holistic planning and proactive risk management in AWS migrations. By incorporating these lessons into our practices, we aim to navigate future migrations with even greater confidence and resilience.
Related articles you might find interesting

Why platform teams are re-routing north-south traffic through the Kubernetes Gateway API, what HTTPRoute changes on the ground, and how to migrate without a big-bang rewrite.


Hassle-Free ECS: Terraform Automation + CI/CD Pipeline