Intelligent DevOps Monitoring and Automated Incident Response using Anomaly Detection and Self‐Healing Pipelines

DevOps today focuses on continuous delivery and fast iterations. This chapter details an intelligent DevOps monitoring and incident response system (IDMIRS). This system executes real-time monitoring by aggregating logs, metrics and traces from apps, infrastructure and continuous integrated/continuous delivery pipelines and includes anomaly detection with machine learning triggers and basic workflows automated for self-healing tools. IDMIRS depends on getting operational data from a very complex ecosystem of sources. The main intelligence of IDMIRS is the anomaly detection engine. The feedback loop is the primary mechanism of IDMIRS. IDMIRS merges data-driven intelligence and self-healing in a DevOps process, which results in less downtime, faster remediation and a better resilient service. In addition to objective time savings, IDMIRS also provided subjective time savings by reducing the DevOps engineers' manual labor.