September 2024 Summaries

The Bulk Cancellation UI is a new addition to Inngest's recovery tool suite, designed to help users manage and recover from incidents with ease. This feature allows users to cancel running functions in bulk by specifying a specific time range and selecting one or more function states. It provides a faster, more scalable cancellation strategy that automatically cancels function runs, even when the function is paused. The Bulk Cancellation UI also ensures better visibility and control over cancellations. This feature is ideal for scenarios where users need to end up to millions of function runs all at once or cancel operations based on specific conditions such as time range or function state.

Inngest has introduced two significant updates to enhance observability for developers building applications with durable execution. The first update is a new trace experience for runs, which includes a dynamic waterfall view and expanded run details inspired by OpenTelemetry tracing. This makes it easier to visualize complex workflows and troubleshoot faster. The second update is a holistic top-down monitoring dashboard that provides a comprehensive snapshot of the entire application's health, allowing users to quickly identify anomalies across their Inngest applications. These enhancements aim to improve observability and give developers the tools they need to build resilient, durable workflows.

Introducing the new open-source SDK, Inngest Workflow Kit, which enables developers to create user-defined workflows in hours instead of days. The kit includes a full-stack package that simplifies the development process for both front end and back end. It leverages Inngest's Durable Functions, allowing customizable workflow actions with automatic retries, waits, and more. Additionally, it provides prebuilt React components to build a Workflow Editor UI, making it easier for end-users to configure their workflows. The combination of the Durable Execution Engine and the Workflow Kit's building blocks makes it faster than ever to develop marketing automation, approval workflows, or AI workflows.

Neon has integrated with Inngest to enable triggering durable functions directly from changes in a Neon Postgres database. This integration allows developers to decouple triggers from application logic, making it easier to build durable workflows. The setup process can be done automatically by Inngest or manually by the user. Database events can then be used to trigger Inngest functions, which execute in the developer's own codebase and server environment. This integration provides clear benefits such as decoupling function triggers from application logic, eliminating the need for certain architecture patterns, and leveraging specific Inngest features like fan-out, batching, and flow control. Developers can now create robust, scalable systems with ease using this new integration.

Inngest has announced its self-hosting capability alongside its open source project's 1.0 release. This new feature is highly requested by the developer community as it provides more flexibility in using Inngest. The first self-hosting release focuses on simplicity of deployment, with a single command and service setup. Future updates will include support for Postgres database persistence, event key and signing key management, and multi-node Inngest server support. This new feature enhances the developer's control over their infrastructure while using Inngest.

Durable Functions are a solution for managing long-running tasks that need to withstand interruptions and failures. They provide a powerful framework for building resilient, stateful applications, removing the need for manual error and retry management. In this article, we explored what Durable Functions really are, the problems they solve, and how JavaScript developers can get started with Durable Functions through Inngest's Javascript SDK. Durable Functions enhance the efficiency and reliability of long-running tasks compared to traditional approaches by providing robust state management, advanced error handling, efficient long-running process management, and enhanced scalability. They allow developers to focus on the app's main functions instead of infrastructure concerns.

Rate limiting, debouncing, and throttling are three different ways to control the frequency at which functions must be invoked over a given period. These concepts are essential in any queueing system. Rate limiting is commonly used with public APIs like OpenAI or Shopify API to limit the number of requests per hour or day. Debouncing provides protection against spikes of invocations by preserving the last occurrence of the given window. Throttling, on the other hand, buffers queued function runs to match a configured frequency and time window, often used when dealing with third-party API rate limits.