kwatch runs inside your Kubernetes cluster, traces every failure to its root cause, and sends one message: what broke, why, who changed it and what to do next.
-
Updated
Nov 14, 2026 - Go
kwatch runs inside your Kubernetes cluster, traces every failure to its root cause, and sends one message: what broke, why, who changed it and what to do next.
Kubernetes operator that scales down workloads stuck in CrashLoopBackOff and similar terminal failure states
Captures the heap, threads and core of a Kubernetes container before it dies
Kubernetes operator that detects CrashLoopBackOff, alerts Slack with root cause, and auto-rolls-back bad deployments.
kubectl plugin that explains why a container or pod died or restarted, with evidence and confidence labels
This repo contains files related to k8s troubleshooting scenarios
Explains why a Kubernetes pod crashed: a kubectl plugin and in-cluster controller that turns crashes into 18 stable cause codes for Prometheus, Loki and Grafana.
Read-only Kubernetes pod triage — crash/OOM/saturation root-cause as a token-thrifty markdown digest. CLI + MCP server.
Troubleshooting Kubernetes CrashLoopBackOff with practical examples, root causes, kubectl debugging techniques, and fixes for container crashes, probes, and resource issues.
Kubernetes lab demonstrating emptyDir storage behavior, blue-green deployment, service selector switching, and CrashLoopBackOff troubleshooting.
This repository simulates a one-time database migration/backup: Pod stuck on CrashLoopBackoff and no body cares.
Deterministic root-cause analysis engine and read-only diagnostic collector for Kubernetes CrashLoopBackOff, OOMKilled, and probe failures.
Rank pods by restart count with the last termination reason and exit code, turning a wall of CrashLoopBackOff into an ordered list of what to look at first.
CLI that summarizes CrashLoopBackOff pods across a cluster with their most recent logs.
Common Kubernetes troubleshooting scenarios and resolution steps.
Check that a program will start in a container image before you deploy, without Docker: missing ELF loader (exec no such file or directory), exec format error, missing shared libraries, GLIBC version not found, unknown image user, and missing CA certificates or time zone data for Go programs. Claude Code skill and Go CLI.
Claude Code skill and Go CLI that checks the flags Kubernetes manifests pass to a Go program against the flags its source defines, with the program's real flag, pflag or cobra parser. Finds args that crash it at startup, such as unknown flags and bad values, and silent mistakes such as --debug false.
Check Kubernetes liveness, readiness and startup probes against the servers a Go program really starts. Finds probe paths that return 404, 401 or 405, ports nothing listens on, servers bound to localhost and HTTP/HTTPS mismatches, by running the program with the container's args and env without opening a port.
To associate your repository with the crashloopbackoff topic, visit your repo's landing page and select "manage topics."