Self-Healing CI/CD Pipelines: Automating Kubernetes Deployments with Terraform and Ansible

Shinde, Prasanna Pankaj (2025) Self-Healing CI/CD Pipelines: Automating Kubernetes Deployments with Terraform and Ansible. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

In fast developing world of cloud-native DevOps, the dependability, and robustness of the CI/CD pipelines are essential. The traditional CI/CD systems are reactive, tool-centric, and highly manual prone when it comes to fault recovery. This study presents an integrated, intelligent, self-healing, CI/CD platform which incorporates Kubernetes, Terraform, Ansible, Prometheus and Open Policy Agent (OPA) as the systems to identify, categorise, and process failure rates, in an autonomous manner, in real time. The development system consists with five main modules, including Failure Classification Engine, Trust-Aware Deployment Scoring, Dynamic Policy enforcement, Deployment Health Scoring (DHS), and Explainable Rollbacks. All of these elements are all that is required to make the decisions be smart, adaptive to the context and be governed in a smart fashion, the roll back to be automated with full transparency and to modify policies on the fly. Deploying Prometheus as the observability toolset, the system constantly monitors pipeline activity and acts upon it with a sort of recovery protocol with selective fire as it is known. The framework, with simulations of controlled failures including changes of software and infrastructure levels of anomalies, the stability of a deployment, and policy compliance. The paper addresses the ability of open-source technologies to be choreographed to create scalable, cost-effective, autonomous CI/CD pipelines that enable minimum downtime, minimum human intervention, and improvement in kinetic confidence in deployment outcomes. The study has valuable implications to DevOps organizations as a resilient, cloud-native automation that does not come with the limitation of proprietary automation.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Emani, Sai UNSPECIFIED
Subjects:	T Technology > T Technology (General) > Information Technology > Cloud computing T Technology > T Technology (General) > Information Technology
Divisions:	School of Computing > Master of Science in Cloud Computing
Depositing User:	Ciara O'Brien
Date Deposited:	31 Mar 2026 09:24
Last Modified:	31 Mar 2026 09:24
URI:	https://norma.ncirl.ie/id/eprint/9267

Actions (login required)

View Item