March 2023 Summaries

Sentry has significantly enhanced its build process efficiency by integrating Nx caching with Lerna 6 in its JavaScript SDK monorepo. This improvement came in response to the fast-paced nature of the company's development environment, where they processed 165 pull requests and modified over 800 files within a month, resulting in substantial time savings. By adopting Nx's task runner, Sentry achieved an 87.5% reduction in minimum build times and a 25% decrease in average build times, alongside a 35% reduction in CI runtime. The key to these improvements was the strategic caching of build tasks, leveraging Lerna's capabilities to avoid redundant operations. Sentry's approach included defining dependencies between scripts and employing named inputs to optimize cache invalidation, balancing between speed and reliability to ensure that changes are correctly detected without unnecessary cache misses. The integration not only focused on build tasks but also extended to linting and unit tests, showcasing the versatility and impact of the caching strategy on development workflows.

Codecov has integrated with Sentry, allowing developers to see untested code causing errors directly in the Sentry Issue stack trace, making it faster and easier to debug and patch problems as they come up. This integration aims to unify pre-production and post-deploy data, reducing the time spent on building test coverage by nearly 50%. To get started, users need to have the GitHub integration for Sentry installed and have their test coverage data in Codecov, with a new offer exclusively for Sentry customers offering 5 seats of Codecov Pro for $29/month. The integration is easy to set up, with options ranging from doing nothing to enabling it immediately, and all Sentry customers will also get a 14-day free trial when they upgrade to Codecov Pro for Sentry.

Sentry uses Code Mappings to connect errors to the source code in a repository, providing features such as navigating directly to the source code from the issue details page and analyzing who touched the line that triggered the error to suggest an assignee. However, setting up Code Mappings was previously cumbersome, but Sentry has simplified this process by enabling automatic Code Mappings for customers with the GitHub integration installed in certain languages. To achieve this, Sentry had to overcome technical challenges such as scaling processing jobs and using GitHub APIs efficiently while adhering to rate limits. The feature is currently available for Python, JavaScript, Node, and Ruby languages, but more languages and integrations are planned.

Sentry has introduced a new feature that automatically detects and groups common performance problems with code, providing actionable context to address application slowdowns and latency issues. This feature brings the workflow of errors and exceptions to performance monitoring, allowing developers to quickly identify and fix performance issues. Sentry now auto-detects frontend performance issues such as N+1 database queries, uncompressed assets, and large render-blocking assets, as well as backend performance issues like consecutive database queries, MN+1 database queries, and slow database queries. Additionally, Sentry detects mobile performance issues, including file I/O on main thread errors. With this new feature, developers can take immediate action to triage and resolve performance issues, improving the overall user experience of their applications.

The New APM: Actionable, Affordable, and Actually Built For Developers` introduces a new approach to application monitoring that focuses on being actionable, affordable, and built specifically for developers. This new approach aims to simplify the process of finding and fixing latency issues by removing complexity and providing actionable insights. The new Sentry APM solution is designed to be more affordable than traditional legacy tools, with a focus on delivering value while reducing costs. It also includes features such as performance issues detection, session replay, and integration with popular tools like GitHub and Slack. The goal of this new approach is to make performance monitoring easy and accessible for developers, allowing them to quickly identify and fix issues that impact their application's performance.