All dashboards

Restate: Internals

Deep dive into Restate Server components for troubleshooting and performance analysis

Restate: Internals screenshot 1

Official deep-dive monitoring dashboard for Restate - the platform for building resilient applications with durable execution.

Overview

This dashboard provides detailed visibility into Restate Server internals for troubleshooting and performance analysis. Use it alongside the Restate: Overview dashboard to drill down into specific subsystems.

Sections

  • Errors & Diagnostics - Error rates, errors by node and partition, task status comparison
  • Ingress (HTTP & Kafka) - Request rates by service, latency percentiles, status breakdown, Kafka consumer lag
  • Invoker (Detailed) - Task rates by status, task duration percentiles, invoker capacity per node
  • Bifrost - Log System - Sequencer append rates and latency, store latency by target node, record cache hit rate
  • Partition Processor - Command processing rate by type, command latency (P99), applied LSN lag
  • RocksDB - Read/write ops, get/write latencies, block cache hit/miss, compaction activity, LSM tree structure, write stalls
  • Metadata Server - Get/put operations, latencies, Raft replication status

Requirements

  • Restate Server exposing metrics on /metrics (default port 5122)
  • Prometheus scraping Restate metrics

For high-level cluster health monitoring, use the companion Restate: Overview dashboard.

Revisions
RevisionDescriptionCreated

Get this dashboard

Import the dashboard template

or

Download JSON

Datasource
Dependencies
Resources
Docs: Importing dashboards Webinar: Getting started with Grafana dashboard design Webinar: Building advanced Grafana dashboards