Version: 1.15.0

Application Replication

This topic describes how to use a replication slice to replicate applications from one cluster to another.

Overview

The replication slice type is a new type of slice within the KubeSlice platform. It is designed specifically for efficiently transferring applications/resources from one cloud or data center cluster to another cloud or data center cluster. After the replication of namespaces is complete, they can be integrated into an application slice that manages them. Namespaces can be integrated into an application slice if it is beneficial for them to communicate with remote applications in the source cluster or any other.

A replication slice can also be used to back up or replicate namespaces by specifying the same cluster as the source and the destination during slice creation. The replicated namespaces can be restored later.

Replicate Applications Between Two Worker Clusters

Let us install the KubeSlice controller, register worker clusters, create a slice, and connect worker clusters to the slice. On the worker clusters, let us deploy the Boutique application microservices on a worker cluster.

You must clone the examples repo as it contains all the required configuration files in the examples/replication-demo directory.

Use the following command to clone the examples repo:

git clone https://github.com/kubeslice/examples.git

After cloning the repo, use the files from the examples/replication-demo directory.

The following tree structure describes all the folders and files of the examples/burst-demo directory.

├── boutique-app-manifests           # Contains the Boutique application's frontend and backend services 
│   ├── boutique-app.yaml            # Contains the boutique app
├── kubeslice-cli-topology-template     # Contains the topology template for installing the KubeSlice Controller and register two worker clusters
│   └── kubeslice-cli-topology-template.yaml
├── replication-job-config               # Contains the replication job YAML configuration     
│   └── replication-job-config.yaml
├── replication-slice-config               # Contains the replication-slice YAML configuration     
│   └── replication-slice-config.yaml

You must install the KubeSlice Controller. Identify a controller cluster to install the KubeSlice Controller. Identify two worker clusters, worker-1 and worker-2 with the KubeSlice Controller. Use the following template to install the KubeSlice Controller and register the two worker clusters:
```
examples/replication-demo/kubeslice-cli-topology-template/kubeslice-cli-topology-template.yaml
```
1. Modify the values as per your requirement. Refer to topology parameters for more details. Add the project name as kubeslice-avesha in which you would create a replication slice. If you want to provide a different project name, then be sure to add that project while creating the replication slice.
  
  info
  For more information, see kubeslice-cli. Refer to installing the KubeSlice Controller to install the KubeSlice Controller and KubeSlice Manager using YAML.
  
  note
  KubeSlice Manager simplifies cluster registration and slice operations. To know more, see cluster operations.
2. Install KubeSlice using the following command on the controller cluster:
```
kubeslice-cli --config examples/replication-demo/kubeslice-cli-topology-template/kubeslice-cli-topology-template.yaml install
```
  Running this command installs the KubeSlice Controller and registers worker-1 and worker-2 clusters with it.
3. On the controller cluster, create a namespace called boutique using the following command:
```
kubectl create ns boutique
```
4. (Optional) To retrieve the endpoint/URL for accessing the KubeSlice Manager:
  1. Run the following command on the KubeSlice Controller:
    kubectl get service kubeslice-ui-proxy -n kubeslice-controller -o jsonpath='{.status.loadBalancer.ingress[0].ip}'
  2. The output should display the endpoint/URL in the following format:
    https://<LoadBalancer-IP>
  3. Copy the endpoint/URL from the output and paste into your browser window to access the KubeSlice Manager.
  4. You must create a service-account token to log in to the KubeSlice Manager. Create a service-account token using the following command:
    
    note
    This command works for an admin with admin-ui as the username.
    kubectl get secret kubeslice-rbac-rw-admin-ui -o jsonpath="{.data.token}" -n kubeslice-avesha | base64 --decode
    Running this command returns a secret token. Use this token to log in to the KubeSlice Manager.

Deploy the Boutique application on the worker-1 cluster using the following command:

kubectl apply -f examples/burst-demo/boutique-app-manifests/boutique-app.yaml -n boutique

Validate the Boutique application pods on the worker-1 cluster using the following command:

kubectl get pods -n boutique

Expected Output

NAME                                     READY   STATUS    RESTARTS   AGE
dservice-5c6f8b77fb-7t6tf               1/1     Running   0          12s
cartservice-5445649469-fzc6l             1/1     Running   0          19s
checkoutservice-5cc44f775c-kkgn2         1/1     Running   0          8s
currencyservice-66f5c9f698-dgw2v         1/1     Running   0          17s
emailservice-59c44668d7-jrk5k            1/1     Running   0          9s
frontend-77464b4775-mzj5z                1/1     Running   0          5s
paymentservice-6c5465c467-q2fhj          1/1     Running   0          2s
productcatalogservice-c57fbfdf-vxh7g     1/1     Running   0          1s
recommendationservice-5677d665d6-9tgs8   1/1     Running   0          6s
redis-cart-bbf4477f4-gs6cn               1/1     Running   0          13s
shippingservice-5d94d4496f-njbdj         1/1     Running   0          16s

Validate the Boutique application based on the following service type:
- For NodePort, use http://externalip:nodeport/.
- For LoadBalancer, use https://<external-ip-of-LoadBalancer>/.

Validate the services using the following command:

kubectl get svc -n boutique

Expected Output

NAME                    TYPE           CLUSTER-IP       EXTERNAL-IP                                                               PORT(S)          AGE
adservice               ClusterIP      172.20.112.204   <none>                                                                    9555/TCP         1s
cartservice             ClusterIP      172.20.249.101   <none>                                                                    7070/TCP         8s
checkoutservice         ClusterIP      172.20.90.84     <none>                                                                    5050/TCP         17s
currencyservice         ClusterIP      172.20.96.78     <none>                                                                    7000/TCP         7s
emailservice            ClusterIP      172.20.117.63    <none>                                                                    5000/TCP         19s
frontend                ClusterIP      172.20.28.197    <none>                                                                    80/TCP           14s
frontend-external       LoadBalancer   172.20.190.246   <Load-Balancer>                                                           80:31305/TCP     13s
paymentservice          ClusterIP      172.20.60.15     <none>                                                                    50051/TCP        11s
productcatalogservice   ClusterIP      172.20.114.176   <none>                                                                    3550/TCP         10s
recommendationservice   NodePort       172.20.23.96     <none>                                                                    8080:32200/TCP   15s
redis-cart              ClusterIP      172.20.227.236   <none>                                                                    6379/TCP         3s
shippingservice         ClusterIP      172.20.213.101   <none>                                                                    50051/TCP        5s

On the controller cluster, create a replication slice called boutique-migration using the following command:

note
In the command below, kubeslice-avesha is the project name that you pass in the kubeslice-cli topology template. If you have passed a different project name, then add that project name in this command.
```
kubectl apply -f examples/replication-demo/replication-slice-config/replication-slice-config.yaml -n kubeslice-avesha
```
1. Use the following command to validate the slice configuration:
```
kubectl get migrationslice -n kubeslice-avesha
```
Expected Output
```
NAME                 AGE
boutique-migration   14s
```
1. Use the following command to validate the source and destination worker clusters connected to the replication slice:
```
kubectl get workermigrationslice  -n kubeslice-avesha
```
  Expected Output
```
NAME                          AGE
boutique-migration-worker-1   20s
boutique-migration-worker-2   19s
```
2. Switch to the worker 1 and worker 2 clusters, and use the following commands to verify the replication slice from the source or the destination worker cluster:
```
kubectl get deploy -n migration  
```
  Expected Output
```
NAME     READY   UP-TO-DATE   AVAILABLE   AGE
velero   1/1     1            1           35s
```
```
kubectl get bsl -n migration
```
  Expected Output
```
NAME                           PHASE       LAST VALIDATED   AGE   DEFAULT
kubeslice-boutique-migration   Available   9s               83s
```
After validating the replication slice, switch to the controller cluster and deploy the replication job using the following command:
```
kubectl apply -f examples/replication-demo/replication-job-config/replication-job-config.yaml -n kubeslice-avesha
```
To validate the replication job on the controller cluster:
1. Use the following command to validate the replication job configuration:
```
kubectl get migrationjobconfig -n kubeslice-avesha
```
  Expected Output
```
NAME                 AGE
boutique-migration   43s
```
2. Use the following command to validate the worker clusters connected to the replication slice:
```
kubectl get workermigrationjobconfig -n kubeslice-avesha
```
  Expected Output
```
NAME                          AGE
boutique-migration-worker-1   49s
boutique-migration-worker-2   49s
```
3. Switch to worker-1 cluster to validate the replication job on in it (the source worker cluster) using the following command:
```
kubectl get backup -n migration
```
  Expected Output
```
NAME                           AGE
kubeslice-boutique-migration   9s
```
4. On the worker-1 cluster, use the following command to check whether the backup was successful or not:
```
kubectl describe backup kubeslice-boutique-migration -n migration
```
5. Switch to the worker-2 cluster and validate the replication job using the following command:
```
kubectl get restore -n migration
```
  Expected Output
```
NAME                 AGE
kubeslice-boutique   9s
```
6. On the worker-2 cluster, use the following the command to check whether the restore was successful or not:
```
kubectl describe restore kubeslice-boutique -n migration
```

After applying the replication job, check if the Boutique pods are running using the following command:

kubectl get pods -n boutique

Expected Output

NAME                                     READY   STATUS    RESTARTS   AGE
dservice-5c6f8b77fb-7t6tf                1/1     Running   0          12s
cartservice-5445649469-fzc6l             1/1     Running   0          19s
checkoutservice-5cc44f775c-kkgn2         1/1     Running   0          8s
currencyservice-66f5c9f698-dgw2v         1/1     Running   0          17s
emailservice-59c44668d7-jrk5k            1/1     Running   0          9s
frontend-77464b4775-mzj5z                1/1     Running   0          5s
paymentservice-6c5465c467-q2fhj          1/1     Running   0          2s
productcatalogservice-c57fbfdf-vxh7g     1/1     Running   0          1s
recommendationservice-5677d665d6-9tgs8   1/1     Running   0          6s
redis-cart-bbf4477f4-gs6cn               1/1     Running   0          13s
shippingservice-5d94d4496f-njbdj         1/1     Running   0          16s

After applying the replication job, check if the Boutique services are running using the following command:

kubectl get svc -n boutique

Expected Output

NAME                    TYPE           CLUSTER-IP       EXTERNAL-IP                                                               PORT(S)          AGE
adservice               ClusterIP      172.20.112.204   <none>                                                                    9555/TCP         1s
cartservice             ClusterIP      172.20.249.101   <none>                                                                    7070/TCP         8s
checkoutservice         ClusterIP      172.20.90.84     <none>                                                                    5050/TCP         17s
currencyservice         ClusterIP      172.20.96.78     <none>                                                                    7000/TCP         7s
emailservice            ClusterIP      172.20.117.63    <none>                                                                    5000/TCP         19s
frontend                ClusterIP      172.20.28.197    <none>                                                                    80/TCP           14s
frontend-external       LoadBalancer   172.20.190.246   <Load-Balancer>                                                           80:31305/TCP     13s
paymentservice          ClusterIP      172.20.60.15     <none>                                                                    50051/TCP        11s
productcatalogservice   ClusterIP      172.20.114.176   <none>                                                                    3550/TCP         10s
recommendationservice   NodePort       172.20.23.96     <none>                                                                    8080:32200/TCP   15s
redis-cart              ClusterIP      172.20.227.236   <none>                                                                    6379/TCP         3s
shippingservice         ClusterIP      172.20.213.101   <none>                                                                    50051/TCP        5s

Application Replication

Overview​

Replicate Applications Between Two Worker Clusters​

Overview

Replicate Applications Between Two Worker Clusters