Can you tell me if your Node app is healthy?
summary
Defines precise health metrics for Node: goes beyond CPU/memory to measure event loop latency and tick frequency, shows that a Prometheus client misses long synchronous blocks because it polls on a scheduled tick, and demonstrates replacing Prometheus+Grafana with nodejs-dashboard plus pushing blocked measurements into InfluxDB. Reports spikes up to 2.4s and CPU hitting 100% when parsing big JSON, describes the integration built to capture those spikes, and shows the approach works while costing extra CPU so sample or toggle metrics in production.
AI-written summary · about
chapters (7)
- 0:00Start
- 1:45Cpu Consumption and Memory Consumption
- 2:22Event Loop
- 8:12Prometheus
- 9:51Node Js Dashboard
- 10:48Cpu Graph
- 14:52Demo
related talks
26mObserving Node.js: Using Metrics to Improve your Application Performance
20mMonitoring, And Visualizing your Node.JS server with Open Source tools
25mMy node.js process is on fire
22mExamining Observability in Node.js
13mMy Node.js process is on fire
01h 04mNode.js performance analytics and best practices
25mNode js applications diagnostics under the hood
01h 02mProfiling Node Applications