The log aggregation solution in OpenShift 4 is not changed from the ElasticSearch, Kibana and FluentD stack. The installation process, though, has changed.
Take a moment to familiarize yourself with the logging overview.
Deploying the logging solution requires only a few steps. Take a look at the deploying cluster logging documentation. There are some additional notes on the process here.
You can create the required Project with the following command:
oc create -f https://raw.githubusercontent.com/openshift/training/master/assets/logging-project.yaml
This YAML already has the annotations and appropriate details.
If you had followed the instructions for creating infrastructure nodes, you already have some special nodes with special labels. When you get to the section on "Create a cluster logging instance", if you want to deploy the logging solution onto your infrastructure nodes, we have an existing CR that you can use.
https://raw.githubusercontent.com/openshift/training/master/assets/logging-cr.yaml
You can paste the contents of the CR into the web console.
When the Logging solution is finished installing, find the route for Kibana,
click through to it, accept all certificates and login as kubeadmin
. You
will find that Kibana wants you to set a default index pattern.
Click .all
on the left and then click the star icon on the right to set
this as the default index. You will notice that a star icon appears on the
left next to .all
. At this point you can click "Discover" to perform your
searches.