September 2020 Summaries

The author describes the creation of a consolidated API and a real-time visualization tool to efficiently monitor problem statuses across multiple Dynatrace environments. This tool, developed using Python Flask, flask-socketio, and D3.js, provides a high-level view of open problems, their types, durations, and cross-environment issues, all in real time, without storing data. By updating every 30 seconds and offering features like a replay mode, the visualization allows users to quickly understand the status of all environments. The tool proved its value when it quickly identified a datacenter-wide issue that other monitoring solutions missed, highlighting its utility in detecting both significant problems and short-lived noise. The visualization's success has prompted further exploration into optimizing Dynatrace's anomaly detection settings and operational processes.

The Dynatrace Managed release notes for version 1.204, published on September 25, 2020, detail various updates, new features, and resolved issues specific to Dynatrace Managed. Key features include automatic deletion of specific log files after two months to conserve disk space and enhancements to the Cluster Management Console, such as displaying version and transaction storage size details for each node. Backup processes have been improved to prevent full disk space utilization, with older data snapshots being removed once new backups are completed. The release supports installation on Ubuntu 20.04 and Oracle Linux 8, and addresses numerous issues across components like ActiveGate, Autonomous Cloud, and Cluster Management Console, resolving vulnerabilities and improving performance, user interface, and backup procedures. The release also includes updates and fixes to various APIs, user interface elements, and backup systems, ensuring improved functionality and security.

Releases of new mobile app versions are crucial milestones that significantly impact overall app health, often making it challenging to revert once users have installed them. The Dynatrace Software Intelligence Platform offers integrated mobile app monitoring, providing insights from user devices to back-end services, enhancing user experience and business growth. Recently, Dynatrace introduced a new Crash analysis page, which helps identify and address issues early with features like version cards that offer a quick overview of app versions and crash ratios. This workflow, supported by the Dynatrace Davis® AI causation engine, automates root cause analysis while allowing for manual investigation and tracking of long-term trends. Without requiring configuration, OneAgent for Mobile automatically detects version information, enabling seamless tracking of app performance across different devices and operating systems. Future updates promise even more detailed insights into app versions and user impact, allowing developers to focus on creating superior apps.

Dynatrace's OneAgent version 1.201 release notes, published on September 21, 2020, highlight updates in technology support, operating system compatibility, and various bug fixes across different modules. The release introduces support for GraalVM 19 in Java, and outlines the phased discontinuation of OpenTelemetry support for Go in future versions. Operating system support changes include the gradual phasing out of support for Amazon Linux, Debian 10, Windows Desktop 10 and 11, and Windows Server 2012 R2, with specific end dates provided. The release also addresses numerous resolved issues, such as memory allocation improvements, disk latency metrics, stability enhancements for NAS disk monitoring, and various fixes in modules like Java, mainframe, and JavaScript. Additionally, cumulative updates (129, 148, and 160) are included, providing fixes for specific components like the AI causation engine, IIB, and OS modules to enhance system reliability and performance.

Dynatrace has partnered with ServiceNow to enhance integration through the Service Graph Connector Program, which aims to improve IT Operations Management (ITOM) by offering comprehensive service mapping and automated workflows. This integration provides a complete view of multi-cloud environments, allowing for efficient issue detection and resolution by aligning with ServiceNow's Common Service Data Model 2.0 and improving data quality in the CMDB via the enhanced Identification and Reconciliation Engine (IRE). The Service Graph Connector enables real-time topology and dependency insights, facilitating immediate impact analysis and remediation action in ServiceNow, thus reducing resource strain and allowing teams to focus on strategic initiatives. The new features include enriched service-to-service relationships and a guided setup process, which streamline onboarding and integration management, ensuring a seamless transition and optimal use of the platform. Future updates promise further enhancements, including the use of the Dynatrace Environment v2 API to expose additional entity types, further enriching the CMDB and improving performance in collaboration with ServiceNow.

In a recent Performance Clinic, Stefano Doni, CTO and Co-Founder of Akamas, discussed how modern application development cycles have shifted to being measured in days, increasing the complexity and unpredictability of technology stacks. To address these challenges, Akamas leverages automation, AI, and machine learning to optimize software performance and reduce costs by automatically adjusting configurations based on real-time data. Akamas integrates with Dynatrace for comprehensive monitoring, allowing it to perform AI-driven experiments that optimize system configurations, such as JVM parameters, to achieve significant resource savings and performance gains. Akamas supports a wide range of technologies through its Optimization Pack library and allows for easy integration and customization, offering a flexible on-premises solution without requiring agents on hosts or applications. The platform's ability to automate and refine performance optimization demonstrates the potential for technology to handle the increasing demands of rapid software release cycles efficiently.

Keptn, an open-source project under the CNCF, aims to address challenges in software delivery by automating build validation using SLI/SLO-based quality gates, breaking monolithic pipelines into event-driven delivery choreography, and embracing event-driven auto-remediation with an SLO-based safety net. It integrates with various data sources like Prometheus, Dynatrace, and Neoload to validate metrics against predefined SLOs, facilitating continuous delivery by enabling teams to define their own quality criteria via dashboards. This approach simplifies the process for non-technical users and enhances existing DevOps toolchains, as demonstrated by users leveraging Keptn in Jenkins, Azure DevOps, and GitLab for streamlined build validation and performance test orchestration. The project encourages community involvement to drive innovation and offers a free trial of Dynatrace to explore these capabilities further.

Autodynatrace is a Python module designed to streamline the instrumentation of Python applications by automatically integrating with common libraries such as Flask, Django, Redis, and SQLAlchemy, using the Dynatrace OneAgent SDK. This module simplifies monitoring by requiring minimal code changes, allowing developers to efficiently collect transactional data and analyze application performance. The implementation of Autodynatrace involves a straightforward process, as demonstrated with a Django app, which can be instrumented by installing the module and making minor adjustments to the application settings. This approach enhances visibility into the application's flow, capturing detailed metrics like web requests, database interactions, and external API calls without extensive manual configuration. The open-source nature of Autodynatrace promises future expansions to support a wider array of Python modules, aiming to cover diverse use cases and reduce the time developers spend on monitoring setup.