/ KUBERNETES

How to Create/Update/Scale/List/Delete a DeploymentConfig using Openshift Golang SDK

This article will walk the user through the process of Creating, Updating, Scaling, Listing, and Deleting Openshift DeploymetConfig in an Openshift cluster with the help of Openshift golang client sdk.

Red Hat OpenShift is an enterprise-ready Kubernetes container platform. Openshift is built on top of Kubernetes and has a user-friendly UI to interact with the cluster. It has all the features of Kubernetes, along with a few elements of its own.

Openshift provides us with a deployment object called the DeploymentConfig(DC), which enables us to manage the desired state for a group of Pods/ReplicationControllers. The DeploymentConfig is similar in functionality to the Kubernetes Deployment object. The only difference is that DeploymentConfig is available only in Openshift.


Table of Contents:


Creating a client to communicate with the Openshift API Server

The Openshift container platform is built on top of Kubernetes and uses much of the underlying functionality. We can learn more about communicating with the Kubernetes API Server using go-client SDK in this article.

To perform the operations on Openshift DeploymentConfigs, we need to create a client that will interact with the API Server. Openshift uses the Kubernetes SDK to interact with the kubeconfig file and exposes functionality to create a client. Let’s dive right into the code. We will explain the stuff as we go ahead.

package main

import (
    "fmt"
    "os"
    "log"
    "path/filepath"
    v1 "github.com/openshift/client-go/apps/clientset/versioned/typed/apps/v1"
    "k8s.io/client-go/tools/clientcmd"
)

func main() {
    userHomeDir, err := os.UserHomeDir()
    if err != nil {
        fmt.Printf("error getting user home dir: %v\n", err)
        os.Exit(1)
    }
    kubeConfigPath := filepath.Join(userHomeDir, ".kube", "config")
    fmt.Printf("Using kubeconfig: %s\n", kubeConfigPath)

    kubeConfig, err := clientcmd.BuildConfigFromFlags("", kubeConfigPath)
    if err != nil {
        err := fmt.Errorf("Error getting kubernetes config: %v\n", err)
        log.Fatal(err.Error)
    }
    client, err := v1.NewForConfig(kubeConfig)
    fmt.Printf("%T\n", client)

    if err != nil {
        err := fmt.Errorf("error getting kubernetes config: %v\n", err)
        log.Fatal(err.Error)
    }
}

In the code snippet, we get the user’s home directory using the function os.UserHomeDir() and then acquire the path to the ~/.kube/config file. We then pass the kubeConfigPath to the clientcmd.BuildConfigFromFlags() function, which returns a kubeconfig object. Finally, to get the openshift appsv1 client, we invoke the v1.NewForConfig() function of the Openshift client-go SDK, passing it to the kubeConfig object.

Note: The v1 in the v1.NewForConfig() returns a *v1.AppsV1Client; this is part of the appsv1, but since we are using the DeploymentConfig object further in this tutorial which is also a part of the appsv1 group, we ended up using the appsv1 for the later. While both the v1 and appsv1 referred to in this code fall under the Openshift appsv1.

Creating DeploymentConfigs

In the previous section, we created an Openshift client. Now we will be using it to perform actions on the Openshift resources. For example, this section will use the client to create Openshift DeploymentConfig. But first, let’s review the DeploymentConfig object.

Walk through the DeploymentConfig object

The Openshift go client SDK provides a Create function to create DeploymentConfigs. The procedure takes multiple parameters, one of which we will explain in this section: the DeploymentConfig object. First, let’s walk through the DeploymentConfig object.

...
    dc := &appsv1.DeploymentConfig{
        ObjectMeta: metav1.ObjectMeta{
            Name: name,
        },
        Spec: appsv1.DeploymentConfigSpec{
            Replicas: replicas,
            Selector: map[string]string{
                "app": name,
            },
            Template: &corev1.PodTemplateSpec{
                ObjectMeta: metav1.ObjectMeta{
                    Labels: map[string]string{
                        "app": name,
                    },
                },
                Spec: corev1.PodSpec{
                    Containers: []corev1.Container{
                        {
                            Name:  name,
                            Image: image,
                            Ports: []corev1.ContainerPort{
                                {
                                    ContainerPort: 8080,
                                    Protocol:      corev1.ProtocolTCP,
                                },
                            },
                        },
                    },
                },
            },
            Triggers: []appsv1.DeploymentTriggerPolicy{
                {
                    Type: appsv1.DeploymentTriggerOnConfigChange,
                },
            },
        },
    }
..

Here in the &appsv1.DeploymentConfig{...} struct, we pass the intended name of the DeploymentConfig in the metav1.ObjectMeta struct. Under the appsv1.DeploymentConfigSpec struct, we provide the Replicas count and Selector map. The Template uses the &corev1.PodTemplateSpec struct defined in the Kubernetes k8s.io/api/core/v1. It has a Label map which should be the same as the Selector mentioned earlier. The corev1.PodSpec struct contains the []corev1.Container struct which is an array as we can have multiple containers running inside a Pod. Further, we have the Name, Image and Ports related to a container. The []corev1.ContainerPort struct is an array as we can expose multiple ports in our Pod.

We can define multiple Triggers using the []appsv1.DeploymentTriggerPolicy struct. In this case, we are using the trigger type appsv1.DeploymentTriggerOnConfigChange, which refers to the ConfigChange trigger. The ConfigChange trigger creates a new replication controller whenever the control loop detects changes in the pod template of the deployment configuration.

Using the DeploymentConfig object to create a Deployment

Now that we have deconstructed the DeploymentConfig struct let’s go through the CreateDeploymentConfig() function.

func CreateDeploymentConfig(name, namespace, image string, replicas int32, client *v1.AppsV1Client) error {
    fmt.Printf("Creating new DeploymentConfig `%s` in namespace `%s`\n", name, namespace)
    dc := &appsv1.DeploymentConfig{
        ObjectMeta: metav1.ObjectMeta{
            Name: name,
        },
        Spec: appsv1.DeploymentConfigSpec{
            Replicas: replicas,
            Selector: map[string]string{
                "app": name,
            },
            Template: &corev1.PodTemplateSpec{
                ObjectMeta: metav1.ObjectMeta{
                    Labels: map[string]string{
                        "app": name,
                    },
                },
                Spec: corev1.PodSpec{
                    Containers: []corev1.Container{
                        {
                            Name:  name,
                            Image: image,
                            Ports: []corev1.ContainerPort{
                                {
                                    ContainerPort: 8080,
                                    Protocol:      corev1.ProtocolTCP,
                                },
                            },
                        },
                    },
                },
            },
            Triggers: []appsv1.DeploymentTriggerPolicy{
                {
                    Type: appsv1.DeploymentTriggerOnConfigChange,
                },
            },
        },
    }
    dcObj, err := client.DeploymentConfigs(namespace).Create(context.Background(), dc, metav1.CreateOptions{})
    if err != nil {
        err := fmt.Errorf("[x] Error creating DC: %v\n", err)
        return err
    }
    fmt.Printf("Successfully created deploymentconfig to `%s` in namespace `%s`\n", dcObj.ObjectMeta.Name, namespace)
    return nil
}

The CreateDeploymentConfig() function receives parameters like the name of the DeploymentConfig, and the namespace we need to create the DeploymentConfig. The container image to be used for the Container, the number of replicas that the DeploymentConfig should create and finally, the Openshift client for interacting with the Openshift cluster. The function returns an error which is nil in case the DeploymentConfig gets created.

The Create() function takes in a context, the *v1.DeploymentConfig struct and the metav1.CreateOptions{}. It returns an error and a *v1.DeploymentConfig struct for the created DeploymentConfig. We will return with an error if there is any error while creating the DeploymentConfig. Else, we return with a nil.

Using the code below, we can call the CreateDeploymentConfig() function from the main().

...
    deploymentName := "my-deployment"
    namespace := "my-namespace"

    // Creating a new DeploymentConfig
    image := "docker.io/httpd:latest"
    err = CreateDeploymentConfig(deploymentName, namespace, image, 1, client)
    if err != nil {
        log.Fatal(err)
    }
...

Listing DeploymentConfigs

Now that we can create a DeploymentConfig using the SDK let’s try to list them for a given namespace. This section will use the Openshift client to list DeploymentConfigs running in a namespace. So first, let’s go through the function ListDeploymentConfigs().

func ListDeploymentConfigs(namespace string, client *v1.AppsV1Client) (*appsv1.DeploymentConfigList, error) {
    fmt.Printf("Listing DeploymentConifgs in namespace `%s`\n", namespace)
    deploymentConfigs, err := client.DeploymentConfigs(namespace).List(context.Background(), metav1.ListOptions{})
    if err != nil {
        err := fmt.Errorf("[x] error listing DeploymentConfig: %v\n", err)
        return nil, err
    }
    return deploymentConfigs, nil
}

The ListDeploymentConfigs() receives two arguments, the namespace and the Openshift client. The function returns two parameters *appsv1.DeploymentConfigList and error.

We can call the ListDeploymentConfigs() function from the main() and iterate over the *appsv1.DeploymentConfigList array using the following code.

...
    // Listing deployment Configs
    deploymentConfigs, err := ListDeploymentConfigs(namespace, client)
    if err != nil {
        log.Fatal(err)
    }
    for _, d := range deploymentConfigs.Items {
        fmt.Printf("%s\n", d.ObjectMeta.Name)
    }
...

Using a for loop, we can iterate over the deploymentConfigs variable returned from the ListDeploymentCofigs() function. In the above example, we are printing out the name of the DeploymentConfig using the d.ObjectMeta.Name. The item has all the information related to DeploymentConfig.

Updating the Image of a Container in DeploymentConfig

We saw that using the Openshift SDK; we can easily create and list DeploymentConfigs. Next up, we will update the image of a container in DeploymentConfig. Previously we created a DeploymentConfig with the image docker.io/httpd:latest. We will change that to docker.io/nginx:latest, an open-source web server. Finally, we will update the image with the UpdateDeploymentConfigImage() function. Let’s dive right into the code.

func UpdateDeploymentConfigImage(name, namespace, image string, client *v1.AppsV1Client) {
    fmt.Printf("Updating DeploymentConfig `%s` in namespace `%s`\n", name, namespace)
    payload := []stringPatch{
        Op:    "replace",
        Path:  "/spec/template/spec/containers/0/image",
        Value: image,
    }
    payloadBytes, _ := json.Marshal(payload)

    _, err := client.DeploymentConfigs(namespace).Patch(context.TODO(), name, types.JSONPatchType, payloadBytes, metav1.PatchOptions{})
    if err != nil {
        err := fmt.Errorf("[x] Error Update DC Image: %v\n", err)
        panic(err)
    }

    fmt.Printf("Successfully update image for deploymentconfig to %s\n", name)
}

The UpdateDeploymentConfigImage() function receives the DeploymentConfig name, namespace, and image to be updated and the client. The integerPatch struct defines the payload. It consists of an Op referring to the Operation, which in this case is replace. The Path in the JSON body is where we need to perform the replace operation. The Value is the image name the image variable provides.

The payload is then converted into an array of bytes payloadBytes using the json.Marshall() function. Finally, the Patch() function uses the patch payload payloadBytes along with the types.JSONPatchType, which signifies our patch operation. The UpdateDeploymentConfigImage() is a simple function call without any return value. We can call the above function from the main().

...
    // Updating the image for DeploymentConfig
    UpdateDeploymentConfigImage(deploymentName, namespace, "docker.io/nginx:latest", client)
...

Scaling DeploymentConfigs

Scaling the DeploymentConfig is done through a Patch() function call. Although the SDK has a GetScale() and UpdateScale() function supposedly for getting and updating the scale of the DeploymentConfig. Unfortunately, the UpdateScale() function returned errors and did not work for me. So let’s walk through the ScaleDeploymentConfig() function.

func ScaleDeploymentConfig(name, namespace string, scale int, client *v1.AppsV1Client) {
    fmt.Printf("Scaling DeploymentConfig `%s` in namespace `%s`\n", name, namespace)
    replicas := uint32(scale)

    payload := []integerPatch{
        Op:    "replace",
        Path:  "/spec/replicas",
        Value: replicas,
    }
    payloadBytes, _ := json.Marshal(payload)

    _, err := client.DeploymentConfigs(namespace).Patch(context.TODO(), name, types.JSONPatchType, payloadBytes, metav1.PatchOptions{})
    if err != nil {
        err := fmt.Errorf("[x] Error Scaling DC Image: %v\n", err)
        panic(err)
    }

    fmt.Printf("Successfully scaled deploymentconfig to %d replicas\n", replicas)
}

This function receives a DeploymentConfig name, namespace, number of scale and the Openshift client. Similar to the UpdateDeploymentConfigImage() function, we create a payload to replace the /spec/replicas value in the DeploymentConfig. The function does not return any value. We can call the above function from the main().

...
    // Scaling DeploymentConfig
    ScaleDeploymentConfig(deploymentName, namespace, 1, client)
...

Deleting DeploymentConfigs

Now that we have performed all the operations in the scope of this article, we can go ahead and delete the DeploymentConfig from using the Openshift SDK. We have created a function DeleteDeploymentConfig() to perform the cleanup task. Let’s review the code for this function.

func DeleteDeploymentConfig(name, namespace string, client *v1.AppsV1Client) error {
    fmt.Printf("Deleting DeploymentConfig `%s` in namespace `%s`\n", name, namespace)
    err := client.DeploymentConfigs(namespace).Delete(context.Background(), name, metav1.DeleteOptions{})
    if err != nil {
        err := fmt.Errorf("[x] error deleting DC: %v\n", err)
        return err
    }
    fmt.Printf("Successfully deleted deploymentconfig %s\n", name)
    return nil
}

The DeleteDeploymentConfig() function receives a DeploymentConfig name, namespace and Openshift client and returns an error. The Delete() function call of the Openshift SDK does the work of deleting the DeploymentConfig. We can call the DeleteDeploymentConfig() function from the main().

...
    // Deleting DeploymentConfig
    err = DeleteDeploymentConfig(deploymentName, namespace, client)
    if err != nil {
        log.Fatal(err)
    }
...

Assembling the Pieces

After putting all the pieces together, we have a main.go file containing all the code we discussed previously. Secondly, we have a go.mod file containing the dependencies.

Installing the Dependencies and Testing

We are using the go version 1.19. To install the dependencies, we need to run the following command:

$ go mod tidy

We can run the code using the below command:

$ go run ./main.go

Before running the code, ensure you can access the target Kubernetes cluster using kubectl. The kubeconfig file should be in the user’s home .kube folder for the setup to work as expected.

Note: Instead of using the Openshift golang SDK, which pulls an outdated version, uses the code in the master branch, which has the latest code. There is a GitHub issue in the Openshift client-go repo with more information on this issue.

Conclusion

The Openshift golang SDK is similar in structure and nomenclature to the Kubernetes golang SDK. The code is open-source on GitHub. Therefore, we can go through the code and check out the functionality available in the SDK.

faizan

Faizan Bashir

Principal Engineer | Architecting and building distributed applications in the Cloud | Adventurer

Read More