Using Dashboards

This guide describes the canned Rumi Monitor dashboards and how to use them for monitoring.

Dashboard Overview

Rumi Monitor ships with three operational dashboards, each providing a different view of the system:

Dashboard	Focus	Primary Tags
Container Stats	Individual container health and performance	system_name, vm_name
Application Stats	Application-level messaging and processing metrics	system_name, vm_name, app_name
Host Stats	Physical host resource utilization	host_name

Container Stats Dashboard

This dashboard provides container-level metrics for monitoring the health and performance of individual Rumi containers.

Panels

• Apps Summary — Table showing application state, role, HA policy, and loaded time for each application in the container.

• CPU Load — Process CPU load and host system CPU load over time.

• Threads — Thread count and processor count.

• Heap Memory — Heap used and non-heap used memory over time.

• Host Memory — Physical memory size and usage.

• Garbage Collection — GC counts and durations by collector.

• JIT Compiler — JIT compilation time.

• Heartbeat Logger — Heartbeat logging statistics.

• Admin Client — Admin client connection statistics.

• Trace Log — Trace log output.

Usage

1. Select a system from the dropdown.

2. Select a container from the dropdown.

3. Panels display real-time metrics for the selected container.

4. Use the time range selector to focus on specific time periods.

Application Stats Dashboard

This dashboard provides application-level metrics for monitoring messaging, store operations, and processing latency.

Panels

• Store Operations — Hit rate, miss rate, eviction rate, and entry counts.

• Message Throughput — Inbound and outbound message counts by type.

• Disruptor Stats — Ring buffer utilization and backpressure.

• Transaction Latencies — End-to-end transaction latency histograms and percentiles.

• Messaging Latencies — Per-message-type latency statistics.

• Custom Gauges — User-defined application gauges.

• Custom Counters — User-defined application counters.

• Bus Statistics — Per-bus messaging statistics.

Usage

1. Select a system, container, and application.

2. For bus-level metrics, select a specific bus.

3. Use the time range to correlate latency spikes with throughput changes.

Host Stats Dashboard

This dashboard provides host-level metrics for monitoring physical resource utilization.

Panels

• Disk Usage — Disk space utilization and I/O rates.

• CPU Usage — System and process CPU load.

• Memory Usage — Physical memory utilization.

• Thread Counts — Per-container thread counts on the host.

• Trace Log — Host-level trace output.

Usage

1. Select a host from the dropdown.

2. Optionally filter by system to focus on specific deployments on that host.

3. Use host metrics to identify resource contention between containers sharing a host.