Kubesense

Features

KubeSense Features

KubeSense is an all-in-one observability platform for Kubernetes environments, offering advanced monitoring, logging, tracing, and metrics collection with minimal setup and high performance. With eBPF at its core, KubeSense provides real-time insights into infrastructure and application behavior, empowering teams to optimize performance, troubleshoot issues, and maintain system health.

Feature Overview

Application Performance Monitoring (APM)

Full-stack visibility into your services with auto-instrumented tracing, service maps, endpoint analysis, and database monitoring — all powered by eBPF.

  • Services — Monitor success rates, throughput, latency, and error trends
  • Service Map — Visualize service-to-service dependencies
  • Traces — Distributed tracing with protocol and status filtering
  • Endpoints — Per-endpoint performance analysis
  • Database Monitoring — MySQL, Redis, and MongoDB monitoring

Real User Monitoring (RUM)

Capture real-time performance data from mobile and web applications to understand user experience, crashes, and app performance.

  • Dashboard — Crash-free rates, session metrics, and performance trends
  • Sessions Explorer — Drill into individual user sessions

Logs

Centralized log management with powerful search, filtering, and log transformation pipelines.

  • Log Explorer — Search and analyze logs with faceted filtering
  • Log Pipelines — Parse, extract, and transform logs with pipeline rules

Infrastructure

Deep visibility into Kubernetes cluster resources with real-time monitoring of 11 resource types.

  • Kubernetes Resources — Nodes, Pods, Deployments, DaemonSets, StatefulSets, ReplicaSets, Jobs, CronJobs, ConfigMaps, PersistentVolumes, PersistentVolumeClaims
  • Kubernetes Events — Real-time event stream with type and namespace filtering

Dashboards & Metrics

Custom dashboarding and a powerful metrics query builder.

AgentSRE Intelligence

AI-powered observability with automated root cause analysis, error grouping, and anomaly detection.

Alerts

Threshold-based and pattern-based alerting with notification integrations.

  • Alert Rules — Configure evaluation intervals, conditions, and states
  • Integrations — Slack, Email, PagerDuty, Webhook, and Teams

LLM Observability (Coming Soon)

Monitoring for LLM applications — inference latency, token usage, cost tracking, and prompt analytics.

Service Levels (Coming Soon)

Define and track SLOs with error budget monitoring and burn rate alerts.

Data Streams (Coming Soon)

End-to-end visibility into asynchronous data pipelines and message queue health.

Supported Protocols

KubeSense uses eBPF to auto-detect and monitor 25+ protocols at the kernel level — no configuration needed.

Web & API — HTTP/1.0, HTTP/1.1, HTTP/2, gRPC, ConnectRPC, GraphQL, WebSocket

Databases — MySQL, MariaDB, PostgreSQL, MongoDB, Redis, Cassandra (CQL), Memcached, Elasticsearch

Messaging & Streaming — Kafka, RabbitMQ (AMQP), NATS, MQTT

Network & Infrastructure — DNS, TCP, UDP, TLS/SSL

Additional — SMTP, LDAP

Because eBPF operates at the kernel's network stack, KubeSense can observe any protocol traversing the network — including custom or proprietary protocols — at the TCP/UDP level. The protocols listed above have deep, application-layer parsing built in.

Supported Programming Languages

eBPF auto-instrumentation works with any language — no code changes required. Because eBPF captures telemetry at the kernel level, it works with Java, Python, Go, Node.js, .NET, Rust, C, C++, Ruby, PHP, Elixir, Scala, Kotlin, Swift, and any language that runs on Linux.

For application-level enrichment, KubeSense natively ingests data from all OpenTelemetry SDKs:

  • Java — Auto-instrumentation agent for Spring, Quarkus, Micronaut, and 100+ libraries
  • Python — Auto-instrumentation for Django, Flask, FastAPI, and more
  • JavaScript / Node.js — Auto-instrumentation for Express, NestJS, Next.js, and more
  • Go — Manual instrumentation with rich library support
  • .NET / C# — Auto-instrumentation for ASP.NET Core, Entity Framework, and more
  • Ruby — Instrumentation for Rails, Sinatra, and Rack
  • PHP — Instrumentation for Laravel, Symfony, and WordPress
  • C++ — Manual instrumentation with OTLP export
  • Rust — Community SDK via tracing-opentelemetry
  • Swift — OpenTelemetry Swift SDK for iOS and server-side Swift
  • Erlang / Elixir — OTel SDK for BEAM applications and Phoenix
  • Kotlin / Scala — Fully supported via the Java auto-instrumentation agent

Since KubeSense accepts standard OTLP over gRPC and HTTP, any language with an OpenTelemetry SDK or exporter is compatible.

back

Architecture

next

Installation