April 2021 Summaries

Agile engineering teams often face service reliability issues primarily due to changes, as noted by Azure's CTO Mark Russinovich. To mitigate this, canary deployment is a strategy used, where a new software version is gradually exposed to a small user subset to monitor its impact, using tools like Kong Gateway and Spinnaker. At Razorpay, Kong Gateway is utilized to manage API traffic and ensure deployment reliability by splitting traffic between canary, baseline, and stable deployments. This setup allows for performance metrics comparison and rollback if anomalies are detected. Spinnaker, alongside Kayenta, further aids in automated canary analysis by scrutinizing error rates and latency, with the ability to terminate canary traffic if necessary. Kong's Admin API facilitates progressive deployments by allowing precise traffic control, starting at a low percentage and gradually increasing while monitoring performance. Additional capabilities of Kong Gateway include intelligent load balancing based on user authentication, consistent hashing to distribute traffic, and a Prometheus plugin for detailed metric analysis, aiding in the thorough evaluation of canary releases.

Kong has been recognized as a Great Place to Work® for the second consecutive year, with 94% of employees affirming its positive workplace environment, significantly surpassing the average for U.S.-based companies. This accolade highlights Kong's commitment to fostering an inclusive and supportive culture, as evidenced by anonymous employee feedback on leadership, business practices, and workplace conditions. The company's emphasis on diversity and inclusion is reflected in its core values, which include being global, authentic, and resourceful, enabling employees to thrive and feel welcomed regardless of their location or tenure. Survey results revealed that 100% of new hires felt welcomed, 99% of employees trust their colleagues, and 98% take pride in working at Kong. By prioritizing employee experience and leveraging continuous feedback, Kong has crafted a high-trust environment that encourages growth and engagement on a global scale.

Fastly's next-gen Web Application Firewall (WAF), formerly known as Signal Sciences, integrates with Kong Konnect to enhance the security of API services by blocking malicious requests. The collaboration leverages Kong Gateway's enterprise API management capabilities and focuses on Layer 7 application security. Fastly's system operates by inspecting incoming requests for potential threats, such as SQL injections, and sanitizes them by redacting sensitive information. The system also aggregates attack data through a cloud agent and uses threat intelligence feeds to inform its security measures. Practical applications of this integration include preventing malicious SQL injections, thwarting FinTech account enumeration attacks, and stopping account takeover attempts by setting customizable rules that define conditions for blocking or limiting requests. These rules can be tailored based on parameters such as IP address, request method, and response codes, and monitored through the Fastly management console, offering users detailed insights and control over the security of their API environments.

This tutorial outlines the process of creating a custom Kong Gateway plugin using the Go programming language, focusing on enhancing security by adding an extra layer between consumers and producers. The plugin identifies consumers through a consumer-key in the query string, and without it, an error message is returned. The process involves using the declarative configuration (decK), importing the Go Plugin Development Kit (PDK), and building a Docker image to run the plugin as a container. The plugin acts as middleware, checking for the correct API key for each request and redirecting URLs as necessary. The tutorial highlights the flexibility and expanded capabilities that custom plugins bring to Kong Gateway, encouraging users to explore further resources and engage with the Kong community for additional insights and support.

Kuma is an open-source control plane for service mesh that simplifies the collection of observability metrics across diverse cloud services and Kubernetes clusters. It integrates with Prometheus and Grafana for traffic metrics and performance monitoring, allowing users to quickly set up metrics collection through automatic service discovery. Additionally, Kuma supports integrations with tools like Splunk, Logstash, and tracing systems such as Jaeger and Zipkin, enabling comprehensive monitoring and troubleshooting of microservice behavior. Users can enhance their service consumption within the mesh using Kong Ingress Controller, which offers features like authentication and rate limiting. Kuma's universal design allows for automatic deployment of sidecar proxies in Kubernetes, supporting a seamless experience across various deployment environments. The tutorial emphasizes the benefits of automating service mesh observability with Kuma, highlighting its capacity to reduce costs, particularly through client-side load balancing in Kong Mesh, which has led to significant savings for some users.

Destination: Scale is a digital event aimed at helping organizations overcome challenges in scaling their platforms globally by leveraging microservices, service mesh, and Kubernetes. Industry leaders from companies like Typeform, Redhat, and Datadog will share insights on standardizing management technology, optimizing processes, and enhancing reusability to improve operational efficiencies. Key topics include breaking down monolithic architectures into loosely-coupled microservices, building monitoring strategies for global availability, and using observability tools to enhance system architecture intuition. Additionally, sessions will cover automating tests and deployments of custom plugins and implementing robust CI/CD pipelines to facilitate the delivery of applications across multi-cloud and on-premises environments. The event aims to empower global platform teams and developers to focus on building impactful features while ensuring efficient and scalable operations.

The tutorial on Kong Gateway offers a comprehensive guide on setting up and using Kong to manage and modernize microservices, particularly for applications transitioning from monolithic architectures. It provides instructions on downloading and installing Kong Gateway, adding services and routes, and implementing key authentication and proxy cache plugins through the Kong Manager interface. The tutorial emphasizes the benefits of using Kong Gateway as an abstraction layer to simplify and streamline service management, allowing developers to focus on building business logic while Kong handles common functionalities like authentication and caching. Practical examples illustrate how to configure services, routes, and plugins, with testing facilitated by the open-source tool Insomnia. The tutorial also highlights the architectural freedom and scalability Kong Gateway provides, enabling efficient microservice deployment and management without impacting client interactions.

As businesses increasingly shift towards digital and internet-based solutions, the demand for efficient API gateway governance strategies has become paramount, particularly as companies adopt microservices, Kubernetes, and service mesh technologies to accelerate product deployment. This shift results in more services and connections, necessitating robust API governance to manage the complexity and maintain control. Kong offers solutions for API gateway governance through automation, declarative configuration, and OpenID Connect, enabling organizations to implement standards-based authentication and authorization frameworks. By providing tools such as Kong Mesh, companies can establish zero-trust networks, ensuring secure and reliable communication between services without requiring developers to embed network reliability and security into their code. As the number of services grows, Kong's solutions facilitate the standardization of API experiences, promoting wider adoption and usage while allowing developers to manage their workspaces autonomously and efficiently.