Skip to content
Docs

Collecting Kedify Config

When encountering issues with Kedify, collecting comprehensive debug information is essential for effective troubleshooting. This guide explains how to quickly gather all necessary configuration and log data.

Using kubectl kedify Plugin

Section titled “Using kubectl kedify Plugin”

Prerequisites

Section titled “Prerequisites”

To use the kubectl kedify commands, install the plugin first:

Terminal window
kubectl krew install --manifest-url=https://github.com/kedify/kubectl-kedify/raw/main/.krew.yaml

Collect debug information with:

Section titled “Collect debug information with:”
Terminal window
kubectl kedify dump --all-namespaces --archive

Common options:

  • -A, --all-namespaces - Collect from all namespaces
  • --archive - Create tar.gz archive
  • -n, --namespace NS - Specific namespace only
  • -q, --quiet - Suppress status output (useful for automation)

Using Standalone Script

Section titled “Using Standalone Script”

If you cannot install the plugin, run the dump script directly:

To collect information from all namespaces

Section titled “To collect information from all namespaces”
Terminal window
bash <(curl -s https://raw.githubusercontent.com/kedify/kubectl-kedify/refs/heads/main/dump.sh) --archive -A

To collect information from a specific namespace

Section titled “To collect information from a specific namespace”
Terminal window
bash <(curl -s https://raw.githubusercontent.com/kedify/kubectl-kedify/refs/heads/main/dump.sh) --archive -n $NAMESPACE

You can also download and inspect the script first:

Terminal window
curl -s https://raw.githubusercontent.com/kedify/kubectl-kedify/refs/heads/main/dump.sh > dump.sh
chmod +x dump.sh
./dump.sh -A --archive

What Information is Collected

Section titled “What Information is Collected”

The dump command collects comprehensive debug information including:

Cluster-wide Information:

  • Node status, resource usage, and detailed descriptions
  • Resource allocation summary across nodes
  • Cluster autoscaler events, deployments, and configuration
  • Karpenter resources (NodePools, NodeClaims, events) if installed
  • Pending pods that may indicate resource constraints

KEDA/Kedify Components:

  • ScaledObject and ScaledJob resources
  • HTTPScaledObject resources and referenced services
  • HorizontalPodAutoscaler (HPA) resources
  • Kedify-specific resources (PodResourceProfiles, ScalingGroups, ScalingPolicies)
  • kedify-proxy pod manifests, logs, and Envoy configuration dumps
  • kedify-proxy Prometheus metrics and cluster information

Installation Data:

  • Kedify agent and KEDA component logs
  • Helm installation values and version information
  • KedifyConfiguration resources
  • HTTP Add-on queue data and interceptor metrics
  • OTel scaler metrics and memstore data (if installed)

Namespace-specific Data:

  • Kubernetes events
  • Pod manifests, descriptions, and resource usage
  • Service and endpoint configurations

Requirements

Section titled “Requirements”
  • kubectl (configured with cluster access)
  • curl, jq, yq
  • tar and gzip (for archive mode)

Security Note

Section titled “Security Note”

The dump script collects metadata and configurations but does not extract sensitive data like secret values, private keys, or passwords. However, review the collected data before sharing with support teams.

Next Steps

Section titled “Next Steps”

When contacting Kedify support or filing issues, include the generated archive.