How I Use Cheap Observability to Find Hidden SaaS Bugs | Build in public 204
Quick Overview
The developer maintains software performance and resolves hidden bugs by implementing an observability strategy using GlitchTip, a self-hosted, cost-effective exception tracking service. By logging critical backend and frontend events and integrating them with GitHub, the developer effectively monitors system health, filters out noise, and identifies urgent issues that require immediate attention.
Key Points: The developer utilizes GlitchTip, a self-hosted, cost-effective exception tracker, to monitor application health and identify bugs. A bug-squashing initiative focuses on cleaning up edge cases in the developer's project, Channel Surfer. The developer employs sampling to manage data volume, logging only a percentage of user events to stay within storage limits. GitHub integration automates issue creation, allowing the developer to triage bugs based on their severity and frequency. Personal identifiable information is scrubbed from logs to maintain user privacy while preserving essential debugging data. The developer's current Monthly Recurring Revenue (MRR) exceeds £300, with a long-term goal of reaching £30,000 MRR.
Context: The video is a 'Build in Public' update from a software developer working on their project, Channel Surfer. The developer addresses the technical challenges of running a scalable, cost-effective software application, specifically focusing on how to maintain high-quality observability without incurring prohibitive cloud costs or data storage issues.
Detailed Analysis
The developer explains their approach to building and maintaining software, emphasizing the necessity of robust observability for identifying bugs before users encounter them. By self-hosting GlitchTip on a Virtual Private Server (VPS), the developer achieves significant cost savings compared to managed services. The workflow involves aggressive log filtering to remove noise, such as browser extensions or unusual device behavior, while retaining critical metadata like browser type and operating system. This data is then funneled into a GitHub-based triage system where issues are automatically flagged. The developer details the importance of 'looping'—an automated process where the system identifies new, recurring errors and notifies the developer to investigate. This approach allows the developer to focus their limited time on the most impactful bugs, ensuring the application remains stable while they work toward their revenue target of £30,000 MRR.