Installing Monitoring Module on IBM Cloud Pak for MultiCloud Management V1.3.0-Part 2: Install Steps
In the previous post we saw how to complete the prerequisite readiness for Installing the Monitoring Module. In this post we will go through the steps involved in Installing the Monitoring Module.
To Install the Monitoring Module, complete the following steps as an administrator on your hub cluster on the infrastructure node of your IBM Cloud Pak for Multicloud Management environment. In my case the ose-ose-4 node is the infrastructure node and I have performed all the steps in this node.
The first step in the Installation process is to load the installation archive file to the registry. The steps involved in the process is as given below:
- Create a directory in the node to save the installation and configuration files. I have created a directory called install_dir. Change to this directory after you create it. Execute the below commands to complete the step
mkdir install_dir/
cd install_dir/
2. Transfer the icam_ppa_2020.1.0_prod.tar.gz file to the install_dir created in the above step.
3. Log in to the OpenShift Container Platform:
oc login -u admin
4. As an administrator, log in to the management console (this is the hub management console).
cloudctl login -a <my_cluster_URL> -n kube-system — skip-ssl-validation -u admin
In my environment, I used the below command to login
cloudctl login -a https://icp-console.apps.ose-ose-4.fyre.ibm.com:443 — skip-ssl-validation -u admin
5. Log in to the docker registry:
docker login -u <username> -p <password>
docker login -u $(oc whoami) -p $(oc whoami -t) docker-registry.default.svc:5000
6. Load the Monitoring Module installation image file into IBM’s Docker registry using the below commad:
cloudctl catalog load-archive — archive ./<installation_image_file> — registry $(oc registry info)/kube-system
Where installation_image_file is the compressed Monitoring installation image file that you downloaded in step 5. The load command is run as shown below:
7. Once the archive load is completed, the confirmation screen is shown as below:
The next step involved in the Installation process is to deploy the Helm Chart. For this complete the steps as given below:
- Log in to the IBM Cloud Pak console of your target cluster.
2. Click Catalog as shown below:
3. Select the Repositories and Select local-charts as shown below:
4. Select the chart and click Configure
5. Under the Configuration area, configure the required parameters:
Helm release name: ibmcloudappmgmt
Target namespace: kube-system
Target cluster: local-cluster
License Accepted: Yes
The values are filled up as shown below:
6. Under the Parameters area, expand Quick Start and configure the Global Parameters
Cloud Console FQDN
Run the below command to find the value for your cluster:
kubectl get configmap ibmcloud-cluster-info -n kube-public -o=jsonpath=’{.data.cluster_address}’
Cloud Console Port
Run the below command to find the value for your cluster:
kubectl get configmap ibmcloud-cluster-info -n kube-public -o=jsonpath=’{.data.cluster_router_https_port}’
Cloud Proxy FQDN
Run the below command to find the value for your cluster:
kubectl get configmap ibmcloud-cluster-info -n kube-public -o=jsonpath=’{.data.proxy_address}’
Cluster Master FQDN
Run the below command to find the value for your cluster:
kubectl get configmap ibmcloud-cluster-info -n kube-public -o=jsonpath=’{.data.cluster_address}’
Cluster Master Port
Run the below command to find the value for your cluster:
kubectl get configmap ibmcloud-cluster-info -n kube-public -o=jsonpath=’{.data.cluster_router_https_port}’
I have created a script called Global.sh in my environment to find the values of the above parameters. The script and its output is as shown below:
The parameters are filled up as shown below:
7. Under the Parameters area, expand All parameters and configure the required parameters
Resource monitoring: Selected
Image Repository: docker-registry.default.svc:5000/kube-system
ICAM Default Storage Class: glusterfs-storage
8. Rest of the parameters that I have used in my environment is given in the below screen shots:
For more information on the Other Values to be filled, kindly refer to the link below:
9. When you are finished configuring all the required parameters, click Install as shown below to deploy the ibm-cloud-appmgmt-prod Helm chart.
10. After approximately 10–30 minutes, the Monitoring pods are in either a Running or Completed status. You can check by running the oc get pods -l release=my_release_name -n kube-system command.
Next, complete the post-deployment steps to register with OIDC and register the service policy:
- The command to register with OIDC is as shown below:
kubectl exec -n kube-system -t `kubectl get pods -l release=ibmcloudappmgmt -n kube-system | grep “ibmcloudappmgmt-ibm-cem-cem-users” | grep “Running” | head -n 1 | awk ‘{print $1}’` bash — “/etc/oidc/oidc_reg.sh” “`echo $(kubectl get secret platform-oidc-credentials -o yaml -n kube-system | grep OAUTH2_CLIENT_REGISTRATION_SECRET: | awk ‘{print $2}’)`”
2. The command that adds two policy registrations is as given below:
kubectl exec -n kube-system -t `kubectl get pods -l release=ibmcloudappmgmt -n kube-system | grep “ibmcloudappmgmt-ibm-cem-cem-users” | grep “Running” | head -n 1 | awk ‘{print $1}’` bash — “/etc/oidc/registerServicePolicy.sh” “`echo $(kubectl get secretibmcloudappmgmt-cem-service-secret -o yaml -n kube-system | grep cem-service-id: | awk ‘{print $2}’)`” “`cloudctl tokens — access`”
The Monitoring Module is successfully installed with IBM Cloud Pak® for Multicloud Management. If you want to verify that your installation is successful, you can run the helm test my_release_name --cleanup --tls command. If the Monitoring installation is successful, all the test pods that are generated are passed.
This completes the Installation of the Monitoring Module.
