Skip to content

Latest commit

 

History

History
118 lines (71 loc) · 3.87 KB

20-fabric.md

File metadata and controls

118 lines (71 loc) · 3.87 KB
Raw

Deploy a Fabric Network

PREV: Deploy a Kube <==> NEXT: Install Chaincode

Fabric-operator extends the core Kubernetes API with a set of custom resources suitable for describing the nodes of a Hyperledger Fabric Network. With the operator, a set of CA, peer, and orderer resources are applied to the Kube API controller. In turn, the operator reflects the network as a series of Pod, Deployment, Service, and Ingress resources in the target namespace.

After the nodes in the Fabric network have been started, the fabric peer and CLI binaries are used to connect to the network via Ingress, preparing a channel for smart contracts and application development.

Fabric Operator

Ready?

just check-kube

Sample Network

kubectl apply -k https://github.com/hyperledger-labs/fabric-operator.git/config/crd
  • Apply a series of CA, peer, and orderer resources directly to the Kube API controller. In turn, fabric-operator will reconcile a network of Kubernetes Pods, Deployments, Services, and Ingress to reflect the target network structure.
just cloud-network
  • Create a Fabric channel:
just cloud-channel
  • Set the location for the network's TLS certificates, channel MSP, and user enrollments:
export WORKSHOP_CRYPTO=$WORKSHOP_PATH/infrastructure/sample-network/temp

Post Checks

curl \
  -s \
  --cacert $WORKSHOP_CRYPTO/cas/org1-ca/tls-cert.pem \
  https://$WORKSHOP_NAMESPACE-org1-ca-ca.$WORKSHOP_INGRESS_DOMAIN/cainfo \
  | jq

Kube Navigation and Peer Logs

To watch the peer logs throughout the workshop, we'll need to identify the pod name for one of the peers, let's find the pod name for org1-peer1 by using kubectl. Set the default namespace to test-network so that we don't have to pass the namespace (-n) to each kubectl command:

kubectl config set-context --current --namespace=test-network
kubectl get pods

You'll see the org1-peer1 pod with a name like org1-peer1-79df64f8d8-7m9mt, your pod name will be different!

We can then tail the org1-peer1 log in a terminal window so that we can see proof that chaincodes get deployed, blocks get created, etc:

kubectl logs -f org1-peer1-79df64f8d8-7m9mt peer

Now that you know how to use kubectl, let's learn the shortcut! You can easily monitor all of the the pods in the k9s utility. If you haven't started k9s yet, start it in a new terminal:

k9s -n test-network

You'll see the fabric-operator, peer, orderer, and CA pods. Navigate around by hitting ENTER on one of the pods, ENTER again on one of the containers, and then hit 0 to tail the container's log. Go back up by hitting ESCAPE. More tips are available at the top of the k9s user interface.

Troubleshooting

# While running "just cloud-network and/or just cloud-channel":
tail -f infrastructure/sample-network/network-debug.log

Take it Further:

PREV: Deploy a Kube <==> NEXT: Install Chaincode