Testimonial Story: IT System House of the Federal Employment Agency

Keyfacts:

Field of activity

As an IT service provider of the Federal Employment Agency (BA), the IT System House is entrusted with the development and operation of IT processes and the creation of IT and communication solutions.

Headquarters

Nuremberg, Germany

Company Information

The BA Information Technology Division is made up of the IT Control Department, which is assigned to the BA Head Office in organisational terms, and the IT System House. In total, BA information technology employs about 2,000 people – of which about 170 work at the head office and about 1,300 in the IT System House.

About the IT System House of the BA

The IT System House serves users on 170,000 networked PC workstations, for example in the employment agencies (AA) and job centers. This also includes an efficient infrastructure, a communication network, and computer centers. The BA Information Technology thus operates one of the largest IT landscapes in Germany and continues to develop it.

The BA Information Technology Division operates two highly efficient central computer centers at the Nuremberg site. A total of 10,000 servers are located there, around 6,000 of which are virtual.

Conversation partner:

Tobias Wagner from ATIX AG asks two employees of the Federal Employment Agency how they use orcharhino and work with it every day. Stefan Bock from the Federal Employment Agency (BA) and an external employee working for the BA provide the answers.

Stefan Bock (BA), has been working in the IT System House of the BA for a good 20 years. Since then he has been working there in the Linux environment. He is mainly responsible for release management and the provision of software in the area of repositories.

Interview:

Tobias Wagner (TW): “What systems do you use?”

Stefan Bock: “Originally, we were only running SUSE Linux Enterprise Server. But now, systems have been added in Red Hat.”

TW: “What made you look for an automation solution?”

Stefan Bock: “We previously had Foreman But now, a second operating system was just added, which we had to map. We, therefore, needed a unified solution so that we didn’t have to synchronize the entire software across two different systems.”

External employee: “The problem, as Stefan mentioned, is the number of systems we have to support. This starts with SLES12 over SLES15 and also RHEL7. The high number of repositories, some with several version strings at the same time, generates a lot of staging effort. For this purpose automation is essential.”

TW: “How and why did you become aware of the orcharhino?”

Stefan Bock:“We already had Puppet and Foreman. On this basis, a colleague started an evaluation with Katello and came across the orcharhino. Since we need to synchronize SUSE repositories, the orcharhino was the product of choice. One of the crucial points was that the included SSC module was written by ATIX themselves. Another important aspect is that ATIX provides support for this. This is particularly important in a government environment.”

External employee: “As Stefan mentioned, we use the upstream product Foreman. We have been using it for Puppet for five years now. I also had the Katello in the back of my mind for a long time, only this one did not support SUSE for quite some time. Meanwhile, the ATIX is very active in the Open Source Community, so that there is now a solution for it. The orcharhino, for which support is available, was the right product for our requirements”

TW: “What is the orcharhino currently used for?”

External employee:“So far, we’ve only used it to manage repos.”

TW: “What has changed since you used the orcharhino?”

External employee: “I would have to take away the fact that we are not yet in production – which is not due to orcharhino, but to internal processes. But for us, the big advantage is that with the orcharhino we can cover several operating systems with one tool and also simplify things like staging, etc.”.

Stefan Bock: “The repository management has clearly improved by using orcharhino. When rolling out a major release, we used to take a script controlled approach. In other words, we had to replace the entire repos here. In doing so, we always had systems that fell out of line. Now we can control the whole thing more effectively by activating keys and assigning different repositories.”

TW: “How many servers do you currently manage with the orcharhino on your test system?

Stefan Bock: “The use of orcharhino on about 3,500 SLES systems with different versions of SLES11 over SLES12 and meanwhile also SLES15 is definitely planned. In the Red Hat section, we are then at about 250 systems with currently Red Hat 7.6., which are then upgraded with the orcharhino to Red Hat 7.7. “

External employee: “We can already say that they will all be connected in the productive environment.

TW: “Are these servers physical or virtual?”

Stefan Bock: “There are physical servers, but we currently assume that there will be at least 2/3 virtual servers.”

TW: “What are you currently planning or which feature do you wish for the orcharhino?”

Stefan Bock: ” Overall, we are very satisfied at the moment. We are simultaneously in the process of building our own cloud solution. Therefore we try to use the Ansible function. There are a few small challenges here and there, but the interface for software management works very well.”

External employee: “If you ask what else we would like to have, these would be very specific things. One thing would be, for example, that there would be some sort of environment function like there is for Puppet, also available for Ansible. But like I said, this is very specific.”

TW: “Many thanks for your time and the pleasant conversation!”

This post is also available in: German