Documentation

Backend

System Metrics

CPU, memory, heap, and event loop of each running instance

The System Metrics page shows how much CPU and memory each instance of your services uses, and how healthy the Node.js runtime is.

System Metrics page with CPU, memory, V8 heap, and event loop charts

What you see

Pick an instance in the Instance selector at the top right. Instances are identified by service and host name.

CardShowsMetrics
CPU UtilizationProcess CPU usage, split between user and system timeprocess.cpu.utilization
Memory UsageMemory used by the processprocess.memory.usage
V8 Heap MemoryJavaScript heap used, and its limitv8js.memory.heap.used, v8js.memory.heap.limit
Event LoopEvent loop utilization, and P99 delaynodejs.eventloop.utilization, nodejs.eventloop.delay.p99

A high event loop delay means your process is blocked by synchronous work, which slows down every request it handles.

Setup

The AdonisJS agent collects these metrics by default. Set hostMetrics: false in config/monocle.ts to turn off CPU and memory collection.

Each instance needs a service.instance.id resource attribute. The agent uses the HOSTNAME environment variable, which is set automatically in Docker and Kubernetes. Without it, each restart appears as a new instance.

Only instances that reported in the last 15 minutes are listed. If your service is stopped, this page shows no data even for past time ranges.

Search documentation

Search pages and sections of the Monocle documentation