 Hi, this is Yoosaflip Haltia and welcome to another episode of Mainframe Matters. And today we have two guests from IBM, Kuryoko's Jijo George, lead of the Tessia project at the Open Mainframe Project and Sibel Ekud's manager of the team, which is contributing to this project. Sibel, Jijo, it's great to have you both on the show. Thank you, Sotnil. It's nice to be here. Yeah, thank you. Nice to be here. Let's talk about some of the basics, which is, what is this project all about? Tessia is basically a tool that we are using. It stands for task execution system and software installation assistant. It basically helps a user, you know, like for people who want to install operating system on the Z-Linux servers. It basically makes it easier for them to install the various flavors of these operating systems on L-Pars, ZVM, or the KVM guests on the Z-Systems. So normally people find it a bit challenging, but this tool is used to help ease that part. So for people who are not that used to the internal interfaces, Tessia can help you in installing your operating system on these Z servers. And on top of that, there is also the possibility of helping in task execution. And that is currently handled by using an Ansible environment through which you can automate your tasks on these particular systems that you have installed. For example, we have test teams from all, you know, we are developing Linox on C as operating system. And so we have test teams testing that and those teams use Tessia and use the scheduler to run their test cases on the Z-Systems. So that is very helpful. What is the origin? What was the challenge or problem that the equities saw that you're like, hey, we need to create a project like that? As Sibyla mentioned, that Tessia is a tool that is quite extensively used in our development teams and teams who want to do any forms of tests and implementations on the Z server. And Tessia basically helps in automating those tasks using, you know, like a command line interface and also a web interface. The idea is that, like, so teams who would like to do some regression testing or, you know, daily scans or daily, like, you know, continuous integration testing based on, you know, newer versions of distros which are made available. So those processes can be automated using Tessia. Apart from that, there are currently, you know, multiple projects which are, so Linux development team is just one of the projects. So there are other projects who are using Tessia for automating their installations and then, you know, running their regression tests and performance tests on these Z servers. And Tessia basically is like the tool which sort of helps them in implementing that. Now, if you look at Linux distribution, there is not one distribution even on ZOS, sorry, systems. A lot of distributions are officially supported. Folks can also run whatever they want as Linux. But you also have to manage the whole life cycle because Linux distribution, depending on what, if you go with LTS release, it's good for two or three years, 10 years, but other six months, some, so do you, this project helps only with getting started with that distribution but to also manage the whole life cycle of that distribution. We do provide some level of support in terms of the life cycle, but it is, like I said, more towards our users. So like depending on what our users wants, we do provide them from canonical. So all the, as you mentioned, all the versions of Ubuntu for testing purposes, of course, LTS versions have a little more support and we do support them for the end of life. In some cases, also for, you know, for like backward compatibility and testing purposes, we do support the older installations as well. And the same goes for, you know, Red Hat being, you know, like a core partner of IBM. So we do perform all of these real based tests and, you know, all the beta versions are tested by our development teams using, you know, the Tessia environment as the base. So it's kind of a tool to, you know, help these teams in, you know, simplifying their installation and, you know, configuration for like any new changes that might come with these distros. Are there any official distributions which are officially supported by this service or it really doesn't matter, developers can add whichever they want? Developers can surely add, but, you know, their own distros, but officially we are currently working with the canonical. So like basically Ubuntu systems, Rails systems and Sless. So Susie, yeah. So we do support these three internally, but of course, as you mentioned, you know, this being an open source project, users are free to implement their own versions of, you know, of OSS and distros into the system. How old is the project? When was it like officially kind of, you know, came into this thing? Officially the project was initiated around the time of 2017, 2018. But like I said, you know, at that time it was more of an internal project. So it was only around 2020 when we, you know, when we implemented that, I mean, we joined open main from project and open sourced the implementation for like the community in general. So initially the idea was that, yes, you know, it was more of a tool as I mentioned for the internal use cases and customers like from within IBM to help them in their tasks. But over time, you know, we thought that, yes, it is a like nice strategic tool. So apart from, you know, like a couple of, you know, internal modifications like the whole application has been made open source and available as a, as a, as a, what? I mean, you can execute it as a dockerized container. So yeah, that helps. As you said, the project has been inside IBM for a while that it was released an open source. So which means that there's already a decent user base of this project. So I want to understand, talk a bit about the user base use cases. We are to see, hey, folks are actually leveraging this project to the Linux on their systems. Currently we have around 30, 35 projects within the, the IBM, I mean, the internal version that we are helping our teams to, to install and set up their implementations on these, on their like Linux on the Z-Linux servers. And basically it, like, as I mentioned, it is used, you know, extensively by the Linux development team, the, the test lab teams, the various, you know, open shift teams, like for testing, you know, crypto systems. So any and all, you know, like any and all code bases which might be, you know, helpful to automate for the installation and then run some tests on top of that using, you know, Ansible. Like currently, like internally, we do have, you know, other works as well that, you know, you can create your own container and implement your own sort of state machines, right? So that you can run any application on top of these. So wherein Tessia will help you to configure and set up the, all the, all the information about the Z servers, the LPAS, the ZVMs in the database. And then those can be used, you know, be used to run a particular kind of job. So like, when I say state machines, I mean, like Ansible state machine has the simplest example that we do have it set up. And wherein users can use it to execute their playbooks directly on top of these Z servers. Talk a bit about, you know, if you, when you look back at some of the major achievements that you're like, hey, when you look back at them, you're like, hey, this is what we achieved. And we are really satisfied with the growth adoption and the progress of the project. In terms of the growth, we have, you know, we have seen consistent growth within like the team to, you know, like of adopters who you like, who heard about, you know, Tessia from one or the other teams, as I mentioned, it started off as a development activity, you know, for the development team, but then it got incorporated by other members by word of mouth within IBM that, you know, that Tessia is a tool that can be used to automate some part of your testing implementation or your code changes. So those kind of, that's how, you know, we have observed that there has been like a steady growth in Tessia at the moment. And yeah, I mean, we hope to continue that growth, you know, across the coming year. Well, I mean, like to be fair, I would say that, you know, like the open source part is like yet to kick off. And, you know, I think open mainframe project is like a, I think the key partner in that, you know, to help us reach to a broader audience. Because until then, you know, it is kind of limited to, you know, users within IBM who know about these tools and, you know, who are exploring it as a, you know, as a utility for performing there or automating their tasks. What kind of people should look at this project who would say, hey, you should be using this project to help your Linux deployments there? Could be any customers who use Linux on C. So, I mean, the team changed a little bit half a year ago. So we are now looking for new ideas. We are talking to our customers. And of course, the Linux team is talking to their customers. And there are some ideas coming up to us, which means at the moment we have, we need an installation for Tessia to use it. But there is a new idea. You might develop an Ansible wrapper to use Tessia without installation so that we can install a Linux image on C. So that is something which our Linux product manager brought up to us. So we are looking in that direction. I think that's an important development for us. And we are very open to our customers to bring or to develop Tessia in a direction where it can be used by a broader audience. I also want to hear about some of the roadblocks, some of the challenges that you still see are there for the project that you're like, hey, we still have to either solve these problems for the project, overcome these challenges. If I ask, what will those be? I mean, I wouldn't say it's a roadblock. It's more of like a journey into open source that you have to cater not just to your needs, but you have to cater to the changing needs of, in our example, it would be like the changes in the distros. So for example, when Ubuntu changed their installers from the traditional Debian to sub-equity that kind of required rework from our part as well. So any and all modifications or changes that do happen on the distro end, on the installer end of all of these distros, that has to be modified or that has to be adjusted for our internal purposes as well. And like Tessia basically provides the capability to modify your installation based on a template file. So you can adjust or adjust your requirements or like depending on the team. So you can do a lot of pre and post-processing after the installation and the setup part. And that is basically how I see that, the different teams can utilize it for their own particular needs, but that also means that we have to adjust to the changes, when and how we observe any modifications from the distro end. Now, when you're saying, we have to make these changes, sometime it looks like it's a lot of work. Can you talk about what the community is around this project, how big it is or how diverse it is? Just give us a view about overview of the community around the Tessia project. The large part of its life cycle, it has been an internal project. So it's only like, as I mentioned in 2020, when it started developing, I mean, it became a part of the open source ecosystem. So currently it is highly pushed by our teams activities. As Sabila mentioned, like last year, we had like a lot of changes within the team. So it's currently handled by about five people like from like in total, wherein three of us are located in Bublingen in Germany and two of the team members are working from India. They are like developers who are like really passionate about like Linux development activities and they are helping us in developing it for like, as Sabila mentioned, based on the user's new requests and modifications. So we hope to target that and proceed in that direction right now. Last year, if I'm not wrong, I think during the open mainframe summit, there was a Linux working group was also announced. I also talked to the team there. Do you folks also work with those teams because there are a lot of different distribution which are part of that working group. Or right now you're just working within your own team here. I was made aware of the Linux working group, quite recently, I would say, like just a couple of months back. So because still that time, as Sabila mentioned, we were in kind of a restructuring phase. So currently we were working for our implementation in terms of like creating a new roadmap for Tessia and how we want to proceed with it. And yeah, as you mentioned, like working with the Linux working group would be an optimal way forward as well. And yeah, I hope that we as in the Tessia team can contribute over there as well. We are for the restructuring, we are looking in different directions and we are looking in different projects where we can contribute. So that is what we want to figure out and we did not complete the roadmap right now. So we have to prioritize what we do first because we have a lot of ideas but we just can go one step after the other. So yeah, that's probably something we want to do to work with other teams together to use other open source tools to connect it somehow. And yeah, make installation and configuration of systems more effective. Since you kind of briefly touched on that, if I ask you, what are the things that, not essentially the pipeline, but I use the word pipeline and roadmap but not in the very traditional sense. But what are the things that you're working on? I can see that one would be to work with the Linux working group as well. But what are the things that you will be looking at working in 2023? Currently, as Sabil had mentioned as also that we are working on a kind of an updated version of the architecture as well. So we had like based on the previous developers like the architect of Tessia. So we are working with him. We are working on a modified, based on like the learnings from the existing implementation. We are working on like a more like a robust and like distributed kind of implementation for Tessia. So that we can have multiple instances of it working together in conjunction. So that's one of those things. I mean, I just wanted to, I don't want to like give a delve too much into the details of that. But yeah, that's one thing for sure that is a core part of our development activities for 2023. And apart from that, we are obviously working with the distro teams, for like as and when new versions of these, like OSS come in, we want to make sure that, that those are made available for our internal teams as soon as possible. And there is, as I said, there is a huge user base who are working on that, just to bring in any like new beta versions of rels, less and one to as and when it is released and they are there for testing and like for deployment. So like that's more of like how our users are using it. But yes, we intend to support them for this, like with their activities for this year of 2023 as well. And yeah, I mean, of course in open source, I think bug fixing is like a very normal part of it. And as and when new development activities come in, we do work on those as well. Shijo, Sibill, thank you so much for taking time out today and talk about TestJet project. And of course, as you said, there are a lot of things in the pipeline. So I would love to chat with you folks again, whenever there is some new updates, but I really appreciate your time today. Thank you. Thanks for a minute. It was nice to be here with you guys. Yeah, it's nice to be here. Thank you.