August 2021 Summaries

Today, we’re announcing that network performance monitoring is now generally available as part of New Relic, allowing users to monitor and analyze their entire stack in one place, including applications, infrastructure, digital experience, and network data. This feature enables users to quickly identify if system performance issues are due to the network or other factors, providing context for faster resolution with network engineering teams. Network performance monitoring is automatically curated into golden signals, which can be used to detect anomalies and provide alerting capabilities. The feature offers scalability and power to ingest, store, and query high-cardinality network data, allowing users to ask ad hoc questions of their data. Users can start using network performance monitoring today without additional licenses or SKUs, and it is included in the free tier for New Relic customers.

New Relic has announced the availability of network performance monitoring as part of its platform, achieved through a partnership with network observability provider Kentik. This integration allows users to correlate and analyze various telemetry data, including applications, infrastructure, digital experience, and network data, all in one place. By providing network engineering teams with the necessary context for faster problem resolution, users can quickly determine if network issues are affecting system performance and engage the appropriate teams. Network data is parsed into "golden signals" such as latency, packet loss, and jitter, with automatic anomaly detection to highlight potential issues. The high cardinality of network data enables users to ask exploratory questions and isolate specific network conversations by source, destination, and TCP port. Existing New Relic customers can access this feature without additional licenses, and new users can sign up for free access, with data costs applicable beyond the free tier.

The author of this text highlights the importance of diversity in tech companies, particularly when it comes to people with disabilities, who are often overlooked and underrepresented. The author, who is deaf, shares their personal experiences and adaptations that have helped them succeed in a world built for hearing people. They argue that people with disabilities are not problems to be solved but rather valuable problem solvers who bring unique perspectives and strengths to the table. However, the tech industry often fails to include disability in its diversity efforts, leading to missed opportunities and exclusions. The author offers practical advice on how companies can actively recruit and include people with disabilities, such as offering accommodations during the interview process, including disability in public diversity reporting, being proactive in providing accommodations, and supporting affinity groups. By doing so, companies can create a more inclusive culture that values diversity and promotes accessibility for all employees.

Diversity discussions in the tech industry often overlook people with disabilities, despite their unique perspectives and problem-solving abilities that can enhance team creativity and resilience. With one in four Americans having a disability, their inclusion is crucial for understanding a significant portion of potential customers. The text highlights the underrepresentation of people with disabilities in employment, citing a disparity in workforce participation rates and earnings compared to those without disabilities. It emphasizes the importance of integrating disability into diversity, equity, and inclusion (DEI) initiatives and public diversity reporting, suggesting proactive measures such as offering accommodations during recruitment and employment, celebrating relevant events like the anniversary of the Americans with Disabilities Act, and creating supportive affinity groups. The author, who identifies as deaf, shares personal experiences that demonstrate how adapting to a predominantly hearing world has enhanced their skills, underlining that the real challenge lies in the "user interface" of societal structures rather than the disabilities themselves. Ultimately, fostering an inclusive tech industry requires a cultural shift that recognizes and addresses the needs and contributions of people with disabilities, going beyond technological solutions to embrace a more comprehensive and empathetic approach.

The OpenTelemetry Collector is a powerful tool for building telemetry data pipelines, allowing users to collect, process, and export data from various sources in a highly configurable and extensible manner. It can accept different formats of telemetry data, such as Prometheus, Jaeger, Kafka, and Zipkin, and apply common processing before exporting it to a backend observability platform. The Collector has two main deployment methods: agent and gateway. An agent is a program that collects telemetry data from a host and forwards it to a collector, while a gateway receives data from multiple agents and sends it to configured backends. The OpenTelemetry Collector can also be used as an agent on every host, enriching the data with host metadata before forwarding it on. It is highly configurable and extensible, allowing users to adapt it to their environment and add custom exporters as needed. The collector has two main deployment methods: agent and gateway. An agent is a program that collects telemetry data from a host and forwards it to a collector, while a gateway receives data from multiple agents and sends it to configured backends. The OpenTelemetry Collector can also be used as an agent on every host, enriching the data with host metadata before forwarding it on. It is highly configurable and extensible, allowing users to adapt it to their environment and add custom exporters as needed.

The blog post discusses the OpenTelemetry Collector, a versatile tool for managing telemetry data pipelines, highlighting its ability to aggregate, process, and export data from various sources like Prometheus, Jaeger, Kafka, and Zipkin to a backend observability platform such as New Relic. It explains how the Collector can be deployed as an agent on hosts to gather telemetry data, enrich it with host metadata, and forward it to a gateway cluster for further processing. The article includes demonstrations of setting up the Collector with Docker and shows how to instrument Java applications using OpenTelemetry, adhering to semantic conventions for consistent data output. The series aims to demystify OpenTelemetry by offering a comprehensive guide from foundational concepts to practical implementation, ultimately encouraging users to leverage OpenTelemetry's capabilities for enhanced observability across diverse software environments.

The latest State of DevOps report by Puppet marks its 10-year anniversary and encompasses the firsthand experience of over 2,600 professionals worldwide across various organizations. The survey found that 83% of organizations have implemented DevOps practices, with a significant increase in highly evolved practices from 10% to 18% over four years. However, the majority of organizations (78%) are still "stuck in the middle" and face barriers such as legacy architectures and a shortage of skilled resources. Automation is widespread in highly evolved organizations, but its value is not always recognized by mid- and low-evolution companies. The report highlights the importance of culture, communication, and proper documentation in facilitating DevOps practices. It also emphasizes the need to incorporate security into DevOps practices, regardless of the name used. Overall, the report provides actionable takeaways for organizations looking to evolve their DevOps practices and improve their overall software development and deployment processes.

Puppet's State of DevOps report, marking its ten-year anniversary, presents insights from over 2,600 professionals worldwide, highlighting the evolution and challenges of DevOps practices across various industries. The report underscores that while 83% of organizations have adopted DevOps practices, only 18% have highly evolved systems, with many stuck at intermediate levels due to barriers like legacy architectures and a skills shortage. Automation and autonomous self-service are identified as pivotal yet underutilized elements, with a cultural shift towards valuing DevOps practices deemed essential for progress. Additionally, the report discusses the integration of security into DevOps, stressing its importance despite debates over naming conventions. The findings suggest that embracing internal platforms and calculated risk-taking can significantly enhance DevOps capabilities, while effective communication and documentation can facilitate self-service and broader adoption.

New Relic has released several feature enhancements for its Dashboards product to help users explore, visualize, and act on their data more efficiently. The company has partnered with Formidable to provide ready-to-use custom visualizations from the Victory open source charting library, giving users more options to visualize their data in a few clicks. New Relic has also reduced 95th percentile dashboard loading time by over 60%, making it faster for users to access insights and beautiful dashboards. Additionally, charts now have more precise querying error messages, making it easier for users to troubleshoot issues. The company has also introduced new features such as the ability to borrow visualizations from other parts of the platform and duplicating widgets and dashboards with a single click. These enhancements aim to make data visualizations and dashboards more useful, faster, and easier for users.

New Relic One has introduced various enhancements to its Dashboards feature, aimed at improving data visualization for users by making it more versatile, faster, and easier to use. These updates include a partnership with Formidable to incorporate the Victory open-source charting library, offering new visualization types such as circular progress bars, scatter plots, and range charts. Additionally, the platform's performance has been optimized, reducing dashboard loading times by over 60%, and improving error message clarity to facilitate quicker troubleshooting. The new "Add to dashboard" feature allows users to borrow visualizations from other platform areas, and duplicating widgets and dashboards has been simplified. Users are encouraged to provide feedback and engage with the community through the Explorers Hub, and further learning resources, such as a Nerdlog episode on Victory charts, are available.

New Relic One provides real-time insights into infrastructure and application performance by aggregating telemetry data in one place. However, a recent service interruption highlighted how automation and redundancy protocols can sometimes exacerbate issues. The company uses a cell-based architecture for its Telemetry Data Platform, which allows for scale and reliability but also presents complexity management challenges. Apache Kafka is used extensively for streaming and processing data, but a broker becoming unresponsive can cause data to be stalled behind it. Infrastructure as code is used to provision resources and manage configurations, but this can lead to issues during incident response when engineers need to act quickly. A combination of human error, inadequate safety mechanisms, and timing issues led to the widespread disruption. The company has learned several key lessons from the incident, including the importance of respecting cell isolation, ensuring emergency tools are safe, and continually evaluating incident response processes. New Relic is committed to continuously improving its technology, tooling, and processes to ensure world-class services for its customers.

New Relic experienced a significant service disruption on July 29, impacting data collection and alerting for customers in the US region, due to a technology failure in their Kafka systems exacerbated by automation and redundancy protocols. The incident revealed weaknesses in their cell-based architecture and emergency response tools, which failed to prevent the issue from spreading beyond the affected cell. The failure was compounded by a series of human errors and misjudgments, such as an unsafe change to data retention settings that was applied across all cells, leading to disk space shortages and further broker failures. Despite having safety layers in place, the incident highlighted the need for better isolation during emergencies and more effective incident response processes. New Relic has committed to learning from this incident, improving their systems, and maintaining transparency with their customers to regain trust and ensure future reliability.

Improving JavaScript skills is crucial for developers, as it's the most commonly used language in web development. To enhance your skills, focus on functional programming concepts, such as map(), reduce(), and filter(). Keeping up with the latest specs, including ES6, ES7, and ES8 features, is also essential. Understanding asynchronous programming using promises, async/await, and generators can help you write more efficient code. Knowing where the language ends and the framework begins is vital when working with frameworks like React. Prototypal inheritance is another important concept to grasp, as it differs from classical inheritance used in languages like Ruby or C#. Using a good code editor with extensions, such as VS Code, ESLint, and Bracket Pair Colorizer 2, can make coding more efficient. Testing your code, even when learning, is crucial for writing reliable software. Building an environment from scratch using webpack can help manage large projects. Teaching others what you've learned is key to deep understanding, while being humble and asking questions shows that you're willing to learn. By following these tips, you'll be well on your way to becoming a proficient JavaScript developer.

Improving JavaScript skills offers significant professional advantages due to its ubiquitous use in web applications, with the language's popularity largely stemming from its role as the browser's language. The author, Franz Knupfer, shares insights from his experience as Director of Curriculum at Epicodus, focusing on key strategies to enhance JavaScript proficiency. These strategies include understanding both functional and object-oriented programming paradigms, staying updated with the latest ECMAScript specifications, mastering asynchronous programming, and distinguishing between JavaScript and framework-specific features like those in React. Additionally, the importance of grasping prototypal inheritance, utilizing code editors effectively, and adopting test-driven development is emphasized. The author also underscores the value of teaching others, which can deepen one's understanding, and encourages a humble approach to learning by asking questions. The article concludes with an invitation to explore DevOps and observability, offering resources for those interested in expanding their skill set beyond JavaScript.

OpenTelemetry Java is a language-specific implementation of OpenTelemetry that provides an API and SDK for instrumenting code, as well as a Java agent that can automatically capture telemetry data. The SDK defines the implementation of the API, which can be used to configure how instrumentation data is processed and exported. Manual instrumentation allows developers to add custom instrumentation using the APIs and SDK, while auto-instrumentation uses the Java agent to modify bytecode and capture data without manual configuration. The OpenTelemetry project provides a range of examples and documentation for instrumenting Java applications with tracing, metrics, and other features.

OpenTelemetry Java provides a framework for instrumenting Java applications to capture telemetry data, featuring an API and SDK for manual instrumentation and a Java agent for automatic instrumentation. The API is used by developers to add custom instrumentation, while the SDK processes and exports the collected data. The Java agent offers an auto-instrumentation option by modifying bytecode of popular libraries to collect telemetry data with minimal configuration. This blog post guides readers on manually instrumenting applications using OpenTelemetry's API and SDK, and demonstrates auto-instrumentation with the Java agent, using examples from the newrelic-opentelemetry-examples GitHub repository. It discusses the setup of custom tracing and metrics, highlights semantic conventions for better data interpretation, and emphasizes configuring the SDK early in the application lifecycle to ensure proper data processing and exporting. The post is part of a series exploring OpenTelemetry, with future installments focused on sending data to backend platforms for visualization and analysis.

Cloud native observability is a crucial component of cloud computing that provides continuous and real-time access to data, allowing users to identify potential issues before they cause major problems. It offers a holistic approach to system monitoring, capturing every aspect of the system, from internal processes to hardware and software performance. Cloud native observability tools are designed to work in a full-stack manner, providing a digital panopticon that allows for real-time analysis of past, current, and potential future events. These tools can accelerate DevOps, enable AI-driven detection, and provide real-time data availability, making cloud infrastructure more efficient and secure. New Relic is an open-source platform that integrates data from widely used solutions such as Prometheus, FluentD, Jaeger, and Pixie, providing a streamlined interface for users to monitor and troubleshoot their cloud-native applications.

Cloud native observability is vital for effective cloud computing as it offers real-time, comprehensive monitoring and analysis of complex cloud infrastructures, distinguishing itself from traditional observability tools that are inadequate for serverless environments. This modern observability approach integrates with open-source tools such as Prometheus, FluentD, Jaeger, and Pixie, supported by organizations like the Cloud Native Computing Foundation (CNCF) and companies like New Relic, to provide enhanced capabilities for logging, monitoring, and troubleshooting cloud native applications. Unlike pre-cloud tools that managed data through discrete servers, cloud native observability provides a holistic, continuous view of data across transient cloud environments, allowing for proactive issue resolution with the aid of AI and automation. It is crucial for DevOps efficiency and security, enabling the identification and prevention of potential problems before they escalate, thus ensuring system reliability and preventing downtime.