November 2017 Summaries

Robotic Process Automation (RPA) is increasingly being adopted by IT departments to automate various tasks, offering significant cost reductions as highlighted by the example of UIPath, which demonstrated that 50 robots could perform the work of 800 full-time employees. However, the integration of RPA comes with challenges, such as managing and monitoring the performance of these bots, distinguishing between them, and automating their remediation when issues arise. Dynatrace offers a solution for monitoring RPA deployments by automatically detecting and mapping relationships between RPA servers and clients, capturing and baselining transactions, and identifying performance hotspots. Its capabilities extend to log file analysis and alert generation through ChatOps tools, providing full-stack visibility and root cause analysis for issues impacting RPA deployments. Dynatrace's ease of deployment and automatic detection make it a valuable tool for managing RPA projects, ensuring organizations optimize their automation investments with minimal manual intervention.

AI-driven solutions in DevOps, particularly those utilizing Dynatrace AI, can significantly enhance automated problem resolution, a concept discussed in a recent webinar and elaborated in a subsequent blog post. The post contemplates the appropriate terminology for this approach, oscillating between "auto-mitigation," "auto-remediation," and "self-healing." It describes how Dynatrace AI can detect issues, such as process crashes or database slowdowns, and trigger AWS Lambda functions to automatically address root causes rather than symptoms. For instance, the system can restart crashed processes or temporarily scale services to handle increased loads, thereby minimizing disruptions without human intervention. These automated responses are documented within the Dynatrace Problem history for transparency and accountability. The blog emphasizes the shift from traditional, reactive problem-solving methods to proactive, automated solutions, which leverage AI to reduce Mean Time To Repair (MTTR) by utilizing comprehensive data and dependency analysis.

Dynatrace has announced several updates and changes across various platforms and technologies, including beta support for Java and Apache HTTP Server on Solaris, and IIOP remote services for WebSphere Liberty and Glassfish. Enhancements include improved request body handling in Nginx and the introduction of .NET Method hotspot for .NET Core on Linux, with continued support for .NET 4.7.1. Node.js versions 0.10.x, 0.12.x, and 5.x are deprecated, with version 7.x set for deprecation in Dynatrace version 135, which also marks the end of support for PHP versions 5.3, 5.4, and 5.5. Real User Monitoring is now disabled for Apigee applications, and adjustments have been made to the memory usage chart for PaaS-based hosts, ensuring the display reflects the host's memory limit. Custom plugin metrics require display names, and Windows now supports Docker daemon discovery, with auto-tags for hosts defined via key/value pairs in a configuration file. Additionally, IBM MQ hostname configuration is now possible via a config file, and OpenStack's Keystone plugin supports SSL endpoints, while security updates include the deprecation of metrics retrieval over the Ceilometer API by default.