Aim of project

The aim of the project is to allow clients- global logistics companies- automised scaling of OpenShift clusters with new nodes for increasing the performance of the clusters. Due to the fact that the client uses virtual servers from the VMware vSphere environment set up on OS RHEL 7.x as nodes, it was not possible to use the properties of OpenShift clusters like when using RHEL CoreOS nodes.

Implementation

During implementation of the project a range of Ansible playbooks and roles were created, which after their activation led to the connection of new nodes to the cluster in agreement with the defined parameters. Implementation, successful testing and handover to the client took place in a short time.

Implementation steps

An analysis of the current manual scale-up process was carried out on the basis of the client’s requirements. As part of the evaluation we recommended optimalisation and connection of individual steps in the process.

The necessary playbooks and roles were created and tested. The set of parameters describing the function of the node in the cluster were also defined, along with the inventory. The role can be used collectively as part of the complete scale-up workflow, like for carrying out independent operations, for instance registration of operating systems of virtual servers for the subscription system. Activation of the playbook representing scale-up workflow will lead to a widening of the cluster to include new nodes, defined in the Ansible inventory.

Part of the project was documentation of individual roles and whole playbooks, with all being saved to the client’s GIT environment.

Further phases of the project

The subject of the following phase was optimalisation of several of the already existing client-created Ansible playbooks so as together with the introduction of parameterisation, repeatedly executable (idempotent) Ansible modules were used instead of ‘cmd’ and ‘shell’ modules, which are not effective for normal use.

Project result

Instead of the previous manual process, the client now defines the inventories of new nodes and activates automised action leading to expansion of OpenShift to include these nodes. Its benefit is quick implementation of changes, flexibility and minimalisation of the error rate. These are the main advantages of the use of automisation in complex IT environments.