Automate an Offline Embedded Installation

This topic provides automation steps for setting up an offline, single-node Swimlane installation on embedded clusters.

Installation Preparation and Customization

Before you begin, review the System Requirements for an Embedded Cluster InstallSystem Requirements for an Embedded Cluster Install to confirm your system's compliance. The Swimlane Platform Installer (SPI) performs several checks in the initial phases of the installation to ensure the underlying host is compatible with the application. To bypass any applicable checks, an installer patch YAML file can be applied at installation time to account for several of these settings. Refer to Overriding Installer SettingsOverriding Installer Settings for more options.

Installer Patch File Example

Here is an example installer patch file:

apiVersion: "cluster.kurl.sh/v1beta1" kind: "Installer" metadata: name: "patch" spec: kubernetes: HACluster: true loadBalancerAddress: "<Load Balancer FQDN or IP>:6443" firewalldConfig: disableFirewalld: true selinuxConfig: disableSelinux: true

You also need the following files:

Swimlane Platform Installer license file (ends with .yaml)
Swimlane license file (ends with .lic)
Swimlane tar archive (ends with .tar.gz)
Swimlane airgap bundle (ends with .airgap)

Embedded Cluster Installation

Once you've accounted for any customizations and system checks, install an embedded cluster by running the following command:

tar -zxvf <Swimlane Airgap Bundle>.tar.gz cat install.sh | bash -s installer-spec-file=<Path to Patch File>.yaml

Swimlane Installation and Deployment

After the embedded Kubernetes cluster has been installed, install Swimlane with the following command:

kubectl kots install swimlane-platform \ --namespace default \ --shared-password <SPI Admin Console Password> \ --license-file <SPI License File>.yaml \ --port-forward=false \ --config-values <SPI Config File>.yaml \ --airgap-bundle <Swimlane Airgap Bundle>.airgap

Creating an SPI Config File

The <SPI Config File>.yaml file is a YAML-formatted file that outlines the specifications of a Swimlane deployment. You can get the .yaml file from an previously configured installation of SPI or by creating a new file manually.

To obtain a file from an previously configured SPI install, run the following command from a system with access to the Kubernetes API and the SPI Kubectl Add-on installed:

kubectl kots download swimlane-platform -n <namespace>

This command downloads a copy of the entire SPI deployment specifications to a local folder named swimlane-platform.

The config file is located at ./swimlane-platform/upstream/userdata/config.yaml in the downloaded folder.

To create the file manually, here are all configurable options available in the Embedded Cluster SPIEmbedded Cluster SPI.

Here is a basic version of the config file that you can use for an initial single-node configuration:

apiVersion: kots.io/v1beta1 kind: ConfigValues metadata: creationTimestamp: null name: swimlane spec: values: is_ha: value: "0" mongo_admin_user_password: value: <Base64 encoded string> mongo_admin_user_password_confirm: value: <Base64 encoded string> mongo_swimlane_user_password: value: <Base64 encoded string> mongo_swimlane_user_password_confirm: value: <Base64 encoded string> swimlane_database_encrypt_key: value: <Base64 encoded string> swimlane_database_encrypt_key_confirm: value: <Base64 encoded string> swimlane_enable_ingress: value: "1" swimlane_hostname: value: <Swimlane Hostname>

It takes a few minutes for Swimlane to initialize all of the required components. However, automation can effectively 'wait' for a ready state by waiting for a ready status from different parts of the deployment. There are multiple ways to accomplish this wait period, the following code block is just one example:

kubectl wait -l statefulset.kubernetes.io/pod-name=swimlane-sw-mongo-0 --for=condition=ready pod --namespace <Swimlane Namespace> --timeout=240s export SW_API=$(kubectl get po -n <NameSpace> --no-headers -o custom-columns=":metadata.name" | grep -i api) kubectl wait --for=condition=ready pod $SW_API -n <NameSpace> --timeout=240s

Next, continue with the Swimlane installation by creating the initial administrative user.

Using your Swimlane license .lic file, first generate the necessary license block for the API call with a curl command:

curl -kLv -X POST https://<Swimlane URL>/api/settings/license/upload -H 'Content-Type: multipart/form-data' -F file=@<Swimlane License File>.lic

Then, to create the initial administrative user, insert the returned JSON into the following curl command:

curl -kLv -X POST https://<Swimlane URL>/api/settings/install \ --header "Content-Type: application/json" \ -d '{"mailSettings":{}, "password":"<Admin Password>","confirmPassword":"<Admin Password>", "databaseEncryptionKey":"", "swimlaneUrl":"https://<Swimlane URL>", "license" :<Generated JSON Data From Previous Step>, "adminUserName":"<Admin Username>", "adminEmail":"<Admin Email>"}'