Conversation
- Replaced blocking `std::thread::spawn` + `recv_timeout` per request with a persistent background thread and Tokio runtime. - Implemented `tokio::sync::mpsc::channel` (bounded: 100) to queue insights. - `send_insights` now uses `try_send` to avoid blocking, dropping batches if the queue is full (load shedding). - Preserves synchronous `InsightExporter` trait signature while offloading I/O. - Added `tests/webhook_performance.rs` to verify non-blocking behavior. Co-authored-by: Tuntii <121901995+Tuntii@users.noreply.github.com>
|
👋 Jules, reporting for duty! I'm here to lend a hand with this pull request. When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down. I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job! For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with New to Jules? Learn more at jules.google/docs. For security, I will only act on instructions from the user who triggered this task. |
- Applied `cargo fmt` to `crates/rustapi-extras/src/insight/export.rs` and `crates/rustapi-extras/tests/webhook_performance.rs`. - Added `#![cfg(feature = "webhook")]` to `crates/rustapi-extras/tests/webhook_performance.rs` to prevent compilation errors when the `webhook` feature (and thus the `insight` module) is disabled. Co-authored-by: Tuntii <121901995+Tuntii@users.noreply.github.com>
Tuntii
deleted the
perf/webhook-exporter-non-blocking-6390766288707008754
branch
1 participant
This PR optimizes
WebhookExporterto address severe performance degradation caused by blocking thread operations in async contexts (middleware).Previously,
WebhookExporterspawned a new thread and a new Tokio runtime for every batch of insights, blocking the caller until the HTTP request completed (or timed out).This change:
WebhookExporter::new.send_insightsto be fire-and-forget (non-blocking). If the channel is full, the batch is dropped and a warning is logged to prevent backpressure from blocking the application.tracinginstead of returning them to the caller (as the caller has already moved on).Performance Impact:
exportblocked for ~500ms (or timeout duration) when the webhook endpoint was slow.exporttakes ~50µs (microseconds), returning immediately.This ensures that logging/telemetry does not impact the critical path of request handling.
PR created automatically by Jules for task 6390766288707008754 started by @Tuntii