November 2023 Summaries

Traditional monitoring tools focus mainly on infrastructure metrics, such as CPU usage, network latency, or server health, but fail to provide insight into the end-user experience or application performance from the user's point of view. This limitation can lead to issues being missed that have an immediate impact on users, like sluggish webpage loads, errors during specific user interactions, or challenges navigating the application. In contrast, synthetic monitoring actively models user interactions and workflows, simulating actual user behavior and offering insights into the genuine user experience. By testing applications and identifying problems before they affect real users, synthetic monitoring enhances the overall quality and reliability of digital services in the cloud. It provides a more comprehensive perspective that is aligned with the user's experience, helping to detect and resolve issues proactively, improving the application's adaptability and scalability, and enabling businesses to adopt a proactive approach to issue detection and prevention.

This summary highlights the challenges of end-to-end testing and synthetic monitoring in JavaScript-heavy applications, where asynchronous code can make tests unstable. To address this issue, the article proposes using Playwright's fixture feature to automatically fail end-to-end tests if JavaScript throws exceptions, making it easier to catch problems without directly testing all features. The article also shows how to extend Playwright's test object with custom fixtures and overrides, including a configurable `failOnJSError` option. Additionally, it emphasizes the importance of re structuring projects to be ready for a scalable fixture setup, providing a base file that exports an extended Playwright setup, making it reusable and standardized across tests. The ultimate goal is to get the most out of end-to-end testing by actively monitoring broken functionality while testing critical user flows.

What we don't want is a 10-15 minute speech about your CV, but rather a brief introduction including who you are, what you're doing now, why you started looking for a new job, and why you're interested in joining our company. We also expect clarity around your responsibilities and achievements, real examples to demonstrate them, and the ability to jump into conversations and drive discussions. It's okay to show vulnerability and ask questions that assess our company culture and expectations. Our ideal candidate is someone who can build their own opinion, make informed decisions, and share their thinking with us.

Prometheus is a popular open-source monitoring and alerting toolkit designed for reliability and scalability, bridging the gap between Prometheus and various applications/systems through exporters that collect specific metrics, format them, and expose them through HTTP endpoints for Prometheus to scrape. There are four types of Prometheus metrics: counter, gauge, histogram, and summary. Exporters should be optimized for efficiency, robust error handling implemented, security measures taken, and monitored internally with centralized monitoring systems. Existing exporters save development time but require community contributions, while custom exporters offer flexibility but require significant effort. Prometheuses exporters are available for a wide range of applications, including databases, web servers, messaging systems, and cloud services, allowing for comprehensive monitoring, analysis, visualization, and alerting in diverse computing environments. Checkly's Prometheus Exporter provides a seamless integration between Checkly's synthetic monitoring data and Prometheus, fetching crucial metrics from synthetic checks and exposing them in Prometheus-compatible formats, enabling users to gain real-time insights and monitor synthetic checks alongside other application and system metrics.