Kubernetes observability that uses eBPF to capture service traffic, metrics and CPU profiles automatically; a CNCF sandbox project.
Pricing and hosting
- Pricing
- Free
- Open source
- Yes (Apache-2.0)
- Self-hosted
- Yes
- Launched
- 2020
- Status
- Active
- Website
- px.dev
Plans and prices change; check the tool's own site before you decide.
What it does
Kubernetes and container monitoring checklist
4 of 10 confirmed in Pixie's own documentation. The same checklist is used for every tool in Kubernetes.
- Cluster, node and pod metrics Yes
- Kubernetes events Not confirmed
- Pod logs Not confirmed
- Cluster view Yes
- Requests and limits advice Not confirmed
- Change timeline Not confirmed
- Network visibility Yes
- Alerting No
- Multi-cluster Not confirmed
- Runs in your cluster Yes
FAQ states Pixie does not offer alerting (only via API, e.g. Slackbot tutorial). Stores data in-cluster; Pixie Cloud can be self-hosted.
Checked against: github.com, docs.px.dev, docs.px.dev (2)
Tracing and APM checklist
4 of 10 confirmed in Pixie's own documentation. The same checklist is used for every tool in Tracing and APM.
- Distributed tracing Not confirmed
- OTLP traces ingest Not confirmed
- Service map Yes
- Auto-instrumentation Yes
- Database query analysis Yes
- Error capture Not confirmed
- Latency and error alerts Not confirmed
- Trace to log correlation Not confirmed
- Sampling controls Not confirmed
- Code-level detail Yes
eBPF protocol tracing of requests per service; README does not describe distributed traces across services.
Checked against: github.com, docs.px.dev
Continuous profiling checklist
5 of 9 confirmed in Pixie's own documentation. The same checklist is used for every tool in Profiling.
- Always-on in production Yes
- CPU profiles Yes
- Memory or allocation profiles Not confirmed
- eBPF, no code changes Yes
- Several languages Yes
- Flame graph UI Yes
- Profile comparison Not confirmed
- Linked to traces Not confirmed
- OpenTelemetry profiles Not confirmed
Always-on CPU profiler for Go, C++, Rust and Java.
Checked against: docs.px.dev, github.com
Alternatives to Pixie
Other kubernetes and container monitoring tools. See all 25.
-
Botkube Kubernetes ChatOps bot that posts cluster events and alerts to Slack, Discord or Mattermost and lets you run kubectl from chat. -
Checkmk Infrastructure monitoring for servers, networks, cloud and containers with automatic service discovery; the base edition is open source and paid editions add more.
-
Coroot Open source observability that uses eBPF to build service maps and collect metrics, logs, traces and profiles without code changes, with root-cause hints. -
Dozzle Small web UI that streams and searches live Docker container logs across hosts, with no log storage. Optional Dozzle Cloud adds grouped alerts and log analysis. -
Freelens Community-maintained open source Kubernetes desktop IDE, forked from OpenLens, for managing clusters and viewing workloads, metrics and logs. -
groundcover Kubernetes observability that uses an eBPF sensor and runs in your own cloud account, covering logs, traces, metrics and APM without code changes.
-
Headlamp Kubernetes web UI and desktop app, now a Kubernetes SIG UI project, for browsing cluster resources, logs and events, extensible with plugins. -
Inspektor Gadget CNCF project of eBPF-based tools (gadgets) for inspecting and debugging Kubernetes workloads and Linux hosts, from network traces to process and file activity.
-
K8sGPT CLI and operator that scans Kubernetes clusters for problems and explains them in plain language using an LLM backend of your choice. CNCF sandbox project. -
Komodor Commercial platform for Kubernetes troubleshooting and operations that tracks changes and health across clusters and uses AI agents to investigate and fix issues. -
Kubeshark Captures and dissects API traffic between Kubernetes pods (HTTP, gRPC, Kafka, DNS and more) so you can inspect requests cluster-wide, like Wireshark for Kubernetes. -
Lens Desktop Kubernetes IDE for browsing clusters, resource metrics and logs across many clusters. Free for individuals and small organizations; paid plans add team and enterprise features.