September 2020 Summaries

Sentry alerts are now available in Microsoft Teams, allowing developers to receive notifications for critical issues directly within the platform. This integration enables teams to stay on top of inflight work, design solutions, and resolve issues in real-time using a single communication channel. With Sentry's alert system, users can assign, ignore, or resolve issues right from the chat, providing context to quickly address performance issues such as slow page loads or high error rates. To get started with this integration, developers can navigate to Sentry's settings, create a new alert rule, and configure the Microsoft Teams action to receive notifications when their rules are triggered.

To ensure critical issues don't get stuck in backlog purgatory, Sentry's PagerDuty integration supports Metric Alerts, allowing teams to monitor the overall health of their product or specific parts of it with filtered alerts on event tags and attributes. This integration notifies PagerDuty when event volume goes below a threshold, enabling on-call teams to fix issues immediately. With the added feature, teams can create new Metric Alert Rules in Sentry, configure thresholds for errors or transactions, and view active and resolved alerts, as well as jump into a curated Discover view to root-cause problems quickly.

This integration allows developers to automate issue tracking and sync issues between Linear and Sentry services, enabling teams to streamline their software projects by creating and assigning Linear Issues directly from any Sentry Issue or linking Sentry Issues to existing Linear Issues.

Finding and Fixing Django N+1 Problems` is a common issue in Django where the Object-relational mapper (ORM) causes multiple queries to be executed, leading to performance issues. The problem arises from parent-child relationships, where each row in the results of an initial query spawns another query for each child object. This can lead to overwhelming the database and taking down the application. To fix this issue, Django provides two methods: `select_related()` and `prefetch_related()`. These methods allow the ORM to follow one-to-many relationships and add them to the SQL query as a JOIN, or perform separate queries for each object and then join them in Python. By using these methods, developers can reduce the number of database queries and improve performance. The example code snippet demonstrates how to use `prefetch_related()` to fix an N+1 problem in a Django application, resulting in a significant reduction in page load time and query count.

Monitoring serverless apps can be challenging due to the lack of visibility into error details. Sentry provides code-level visibility for serverless projects, offering stack traces, user-agent information, and metadata around issues, allowing developers to quickly resolve problems. The platform also offers deep-linking to CloudWatch or Stackdriver logs from issue details screens. With its new SDKs for Python and Node, developers can start capturing errors and relevant information in minutes, streamlining the debugging process.