The EAP txn recovery template shows a way how the containers should be configured to make possible running the automatic recovery processing.
(Example of templates for 7.3.x: https://github.com/jboss-container-images/jboss-eap-7-openshift-image/blob/EAP_733_CR1/templates/eap73-tx-recovery-s2i.json and for CD: https://github.com/jboss-container-images/jboss-eap-7-openshift-image/blob/CD21_CR1/templates/eap-cd-tx-recovery-s2i.json)

The templates defines the shared persistent volume for all EAP instances and a separate deployment which finishes XA transactions (XA transaction records stored in Narayana object store) which are left on the persistent volume when some EAP instance crashes or is scaled-down. This processing was used for OpenShift 3 and was deprecated for OpenShift 4. The OpenShift 4 uses StatefulSet instead of multi-write shared persistent volume and WildFly Operator as the process which manages to finished unfinished transaction XA records.

So txn recovery templates is connected with deprecated process on OpenShift 3. Still the template is useful as works as a showcase how to configure the processing.
The template sets up AMQ instance which is then used by quickstart application - it's one of the resources to start an XA transaction.

The trouble is that the current template definition does not work with AMQ LTS broker image (JBEAP-16036) - the LTS one is `amq7/amq-broker-lts-rhel7:7.4` and should be verified if it follows the recommendation how the broker should be installed (https://access.redhat.com/documentation/en-us/red_hat_amq/).