The main benefits are reducing server configuration drift and building new servers or updating software on servers in a very consistent way. My team manages all of the server platforms. We are the middleware team. We support EAP, OpenShift, and the whole gambit. Being able to consistently configure those machines and the software on those machines so that the applications run in a consistent way, good or bad, is key. Even if it is bad, at least it is done consistently. We can pinpoint the problem, make changes from the playbook, and reapply it. We can get it corrected in a very consistent manner, and we do not actively skip a host. If somebody is manually making changes, we know where that goes. I would rate Ansible Automation Platform very highly for helping to reduce the number of steps involved in automating things. After you develop your playbook, that playbook consistently runs, and the playbooks are not difficult at all. Once you understand how Ansible works, it is straightforward. Even on a Windows platform, it is not difficult at all. However, writing new roles is a different story. I have written a couple of those. I have about a dozen or so custom roles written in PowerShell to run on Windows. Even for that, the documentation was not bad. Ansible Automation Platform can help reduce the training required to learn how to automate things. I can give somebody a playbook, and if the author of that playbook follows our standards of documentation, it eases that quite a bit. However, we have some playbooks that are pretty complicated. In those cases, they have to understand what it is doing under the covers. Onboarding to Ansible is pretty straightforward once you get the handle on what tasks are there and what to do in the pretask and the roles. Once you grasp that concept, especially the concept that every task and every playbook will run concurrently on every host which is a hard one for people to grasp, it makes a lot of sense. This reduced training has affected our operations or business. We use the platform's collection of certified content. It is very good. We can find things easily for Red Hat Enterprise Linux. If we find something for Red Hat Enterprise Linux but cannot find it for Windows, we will model the Red Hat Enterprise Linux version to create one for Windows. Ansible Automation Platform has helped to reduce the time we spend on low-value or repetitive tasks. I can give a good example of that. We are also a Dynatrace shop, so we have set up monitors for our certificates on our servers. Dynatrace is alerting us when certificates are 30 days out. We have not yet hooked it up to the automation platform, but the next step is to do that. It is going to run the playbooks that are already there to renew those certificates. After that, we are going to integrate it with ServiceNow to open a ticket when it does that and close the ticket once it succeeds. We are moving down that path, but we are not quite there yet. As an example, we have to renew every single certificate in about three months. I have one playbook that we can run across all of our servers. Regardless of the type of application that is on that server, that playbook will renew the certificate from our certificate server. We will download the certificate to the machine, implement it in Windows, export it from Windows, convert it to the format that the application needs, be it a key store or a PIM file, load it into the application store, and then restart the process of that target application. If you try to do that by hand, you can just forget it.
