Skydive Documentation

To access the Skydive UI, use kubectl port-forward to map the skydive-analyzer service port 8082 to your local machine.

kubectl port-forward service/skydive-analyzer 8082:8082

Use Docker Compose to set up a complete environment including Elasticsearch (for history support), a Skydive analyzer, and a Skydive agent.

curl -o docker-compose.yml https://raw.githubusercontent.com/skydive-project/skydive/master/contrib/docker/docker-compose.yml
docker-compose up

Run Skydive in allinone mode using a Docker container. This requires --privileged mode, host PID and network access, and mounting the Docker socket and network namespaces to allow the agent to capture topology information.

docker run -d --privileged --pid=host --net=host -p 8082:8082 -p 8081:8081 \
    -e SKYDIVE_ANALYZER_LISTEN=0.0.0.0:8082 \
    -v /var/run/docker.sock:/var/run/docker.sock -v /run/netns:/var/run/netns \
    skydive/skydive allinone

To remove the Skydive deployment, delete the resources defined in skydive.yaml.

kubectl delete -f skydive.yaml

To use Graffiti as a dependency in your Go project, install it using go get.

go get github.com/skydive-project/skydive/graffiti

To install Skydive on a Kubernetes cluster, apply the provided skydive.yaml manifest using kubectl.

kubectl apply -f skydive.yaml

You can use the Skydive automation script to resolve inconsistencies in Kubernetes when a secret is updated but dependent pods still hold old configuration values. The script identifies graph edges between secrets and pods and triggers a pod recreation to ensure they pick up the updated secret data.

To trigger the automatic recreation of pods dependent on a specific secret, run the script with the -c flag pointing to your Skydive configuration and specify the secret resource name.

$ ./skydive.sh -c <skydive-config-yaml> secret default/mysql-pass

The fastest way to run Skydive is using the allinone mode via a standalone binary. This mode combines the necessary components into a single process.

Set the SKYDIVE_ETCD_DATA_DIR and SKYDIVE_ANALYZER_LISTEN environment variables before running the command. After starting, access the Web UI at http://localhost:8082.

curl -Lo - https://github.com/skydive-project/skydive-binaries/raw/jenkins-builds/skydive-latest.gz | gzip -d > skydive && chmod +x skydive && sudo mv skydive /usr/local/bin/

SKYDIVE_ETCD_DATA_DIR=/tmp SKYDIVE_ANALYZER_LISTEN=0.0.0.0:8082 sudo -E /usr/local/bin/skydive allinone

Follow these steps to access the Skydive web interface:

1. Find the UI endpoint: Retrieve the host from the skydive-ui route: oc get route skydive-ui -o jsonpath='http://{.spec.host}'
2. Navigate: Point your browser to the retrieved URL.
3. Logout: If already logged in, click the small person icon in the top right to logout.
4. Configure Analyzer: Specify the analyzer endpoint by retrieving its route: oc get route skydive-analyzer -o jsonpath='http://{.spec.host}'
5. Sign in: Use the following credentials:
   • Username: admin
   • Password: password

To deploy Skydive in a Kubernetes cluster, use Helm. You will need the helm-git plugin to install directly from the Git repository. The deployment involves installing both the skydive-analyzer and the skydive-agent.

helm plugin install https://github.com/aslafy-z/helm-git --version 0.10.0
helm repo add skydive git+https://github.com/skydive-project/skydive@contrib/charts
helm repo update
helm install skydive-analyzer skydive/skydive-analyzer
helm install skydive-agent skydive/skydive-agent
kubectl port-forward service/skydive-analyzer 8082:8082