WebSphere Application Server base Cloud Pak

IBM Cloud Paks are a set of prepackaged enterprise capabilities for deployment, lifecycle management, and production cases. Cloud Paks ensure that you can rely on a trusted and tested release with each of the key components that are verified as a single set.

With the WebSphere Application Server Cloud Pak, you can build a custom container application image that extends the new WebSphere Application Server version 9.0 image available from Docker Hub. This base image was designed to allow customers to reuse their Jython scripts and properties file while an application image builds, a pattern that takes advantage of existing WebSphere administration skills and assets.

The default log format from the base container image is in JSON, which allows you to process the log data efficiently by using tools like ELK. If you want to change the default JSON log format, you can follow the steps that are documented in the Logging configuration section on GitHub.

You can also find the Dockerfile and all documentation for the container application image on GitHub.

In addition to the container application image, a production grade Helm chart is available on GitHub. This Helm chart is loaded into the IBM Cloud Private catalog by default. This chart is secure by design and does not require any elevated privileges. The Helm chart is able to run with IBM's most secured policy  and contains a rich set of configurable parameters that use standard Helm mechanisms to ensure consistent enterprise deployments, such as Ingress setup and Horizontal Pod Autoscaling.

You can inject dynamic startup properties with the Helm chart. You can create a Kubernetes ConfigMap that contains a set of traditional WebSphere Application Server properties and bind them to the application container during startup by specifying the name of the ConfigMap.

The log messages that are produced by the Helm chart can be viewed and analyzed by using the Kibana dashboard, which is available on IBM Cloud Private. The Analyzing WebSphere Application Server messages section on GitHub documents details of using Kibana to analyze Helm chart messages.

You can also take advantage of persisted logging. With the Helm chart, the logs from WebSphere Application Server are written to the containers/logs, which can be written to a persistent volume by enabling a single Boolean attribute.