Modern software engineering relies heavily on Continuous Integration and Continuous Delivery (CI/CD) pipelines, which allow teams to provide code quickly and consistently. However, there are unique engineering hurdles when combining heterogeneous DevOps tools like SonarQube, Jenkins, Docker, Kubernetes, HashiCorp Vault, and Prometheus into a single pipeline. Six major issues that arise in real-world CI/CD deployments are methodically identified and analyzed in this paper: extended pipeline execution time due to late quality-gate failures, dependency version conflicts, insecure secret management, environment configuration drift, flaky test-induced false negatives, and inadequate observability during rollbacks.Evidence-based mitigation techniques are suggested for each issue, backed by structured checklists and a comparative failure-rate study. The suggested remedies lower average pipeline failure rates from 32.5% to less than 9%, according to empirical data from 18 enterprise pipelines. For DevOps practitioners looking for dependable, secure, and maintainable delivery pipelines, the findings provide practical advice
Kiroula et al. (Fri,) studied this question.