# Podtrace

Podtrace is a lightweight, eBPF-based observability tool for Kubernetes that continuously reports golden signals, detects anomalies, and identifies likely root causes across workloads — no code changes, sidecars, or. It is open source and self-hosted, installable on Kubernetes (Helm).

Canonical page: https://ourlocalhost.com/app/5006/podtrace

## Description

Podtrace is a lightweight, eBPF-based observability tool for Kubernetes that continuously reports golden signals, detects anomalies, and identifies likely root causes across workloads — no code changes, sidecars, or instrumentation required.

## Platform availability

Available on 1 of 10 platforms: Helm.

## Health score

72/100 (Good), based on 4 of 5 factors.
- Maintained: 100 (last commit 15 days ago)
- Popular: 46 (204 GitHub stars)
- Easy to install: 25 (1 install platform, install notes)
- Security check: 94 (no serious findings, CI, tests, OpenSSF Scorecard 8.4/10)

## Links

- Kubernetes (Helm): https://artifacthub.io/packages/helm/podtrace/podtrace

## Common questions

### What host access is required when running Podtrace as a DaemonSet or container?

Podtrace must see the host's cgroup and process filesystems, and for CRI it must also see the container runtime socket. Without these, resolution can fail or events can be filtered out.

### Does Podtrace require prior instrumentation or configuration?

No. It is activated on demand and works with no prior configuration or instrumentation.

### How do I use declarative tracing in Kubernetes?

Use the Operator docs and choose PodTrace for continuous tracing or the session option. This creates tracing via Kubernetes Custom Resources.

### What should I read first when setting up Podtrace?

Start with Installation, the Usage Guide, and Multi-Pod Tracing.

---
Source: Our Local Host (https://ourlocalhost.com/app/5006/podtrace), the self-hosted and open-source app directory.
