Install Everest¶
Before you start¶
Before running the commands in the Installation section, note that Everest will search for the kubeconfig file in the ~/.kube/config path. If your file is located elsewhere, use the export command below to set the KUBECONFIG environment variable:
export KUBECONFIG=~/.kube/config
Installation¶
To install and provision Percona Everest to Kubernetes:
-
Download the latest release of everestctl to provision Percona Everest. For detailed installation instructions, see CLI installation documentation.
-
Install Everest and provision the Kubernetes cluster using one of the following commands:
everestctl installThis will install all the required components in a namespace called
percona-everest.Example
? Namespace to deploy Everest to percona-everest ? Do you want to enable monitoring? Yes ? Select monitoring instance: Add new monitoring instance ? PMM URL Endpoint http://127.0.0.1 ? Username admin ? Password ***** ? Name for the new monitoring instance my-pmm ? What operators do you want to install? MySQL, MongoDB, PostgreSQLAlternatively, you can install Everest and provision the Kubernetes cluster by running the installation in headless mode:
everestctl install --operator.mongodb=true --operator.postgresql=true --operator.xtradb-cluster=true --monitoring.enable=true --monitoring.type=pmm --monitoring.new-instance-name=my-pmm --monitoring.pmm.endpoint=http://127.0.0.1 --monitoring.pmm.username=admin --monitoring.pmm.password=admin --skip-wizardLimitations¶
-
If the Everest CLI fails to install the operators, do the following:
-
Install Percona Everest, starting with the second step.
-
If you install an operator after the initial provisioning and want to create a database cluster, restart the
everest-operator podby running the following commands:kubectl -n percona-everest get po kubectl -n percona-everest delete pod everest-operator-pod-nameThis will fix the database cluster creation.
Note
The issue will be fixed in the upcoming releases.
-
It is crucial to enable monitoring during the provisioning step. If you skip this step, you won’t be able to enable it later from the user interface. Ensure to fill in the monitoring details in the wizard, ensuring that the name of the monitoring instance does not exceed 22 characters, starts and ends with an alphanumeric character, and only uses alphanumeric characters or ‘-‘.
-
If using a PMM server instance with a self-signed certificate, you cannot use HTTPS in the PMM URL endpoint.
-
-
(Optional) Verify that the services have started properly:
kubectl get pods -n percona-everestExpected output
NAME READY STATUS RESTARTS AGE percona-xtradb-cluster-operator-75c9b976cf-jl9j6 1/1 Running 0 10s percona-server-mongodb-operator-79b9668cd5-d7srk 1/1 Running 0 10s percona-postgresql-operator-84947f45cc-2rb2w 1/1 Running 0 10s percona-everest-5c896598d9-99jnr 1/1 Running 0 10s everest-operator-controller-manager-75b59869c-8jx47 2/2 Running 0 10s -
Access the Everest UI/API using one of the following options for exposing it, as Everest is not exposed with an external IP by default:
Run the following command to use
Kubectl port-forwardingfor connecting to Everest without exposing the service:kubectl port-forward svc/everest 8080:8080 -n percona-everest- Use the following command to change the Everest service type to
LoadBalancer:
kubectl patch svc/everest -n percona-everest -p '{"spec": {"type": "LoadBalancer"}}'- Retrieve the external IP address for the Everest service. This is the address where you can then launch Everest at the end of the installation procedure. In this example, the external IP address used is the default
127.0.0.1:
kubectl get svc/everest -n percona-everestExpected output
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE everest LoadBalancer 10.43.172.194 127.0.0.1 8080:31611/TCP 10s - Use the following command to change the Everest service type to
-
To launch the Everest UI and create your first database cluster, go to the IP address configured for the Everest service in step 4. In this example, this is http://127.0.0.1:8080.
Get expert help¶
If you need assistance, visit the community forum for comprehensive and free database knowledge, or contact our Percona Database Experts for professional support and services.