 Okay, can you hear me? Can you hear me? Yeah. Okay. Good morning everybody. Thanks for coming, even if it's lunch time Today on stage with me Federico Minzoni lead developer at enter. My name is Mariano Cunietti I am the CTO at enter. Enter is a company based in Italy. We run a public cloud name enter cloud suite. It's Multi-region OpenStack based cloud it's based entirely on an open source tech We have regions in Milan, Frankfurt and Amsterdam and on the last June enter was acquired by a big hedge fund An Italian hedge fund that funded the company name Iridios Who is acquiring a lot of infrastructure companies in Italy. So big plans for tomorrow. So What we have done is basically we have done a lot of automation during the last years and We have developed a set of tools that we wanted to share With the people with the with the community Over github. So if you go on github.com it's the last enter class cloud suite you will find the Terraform modules we have developed and the Ansible modules to run our infrastructures and the infrastructures we run for the customers, but over the years Basically, we have discovered that Customers want to handle Customers are software developers mainly and they want to handle and to manage their own CICD at home on-prem So we decided to address another problem We wanted to solve that in order to interact with an infrastructure You need to deploy the build you receive from the developers. So we developed a tool By using basically the the automation scripts We have developed on Terraform and Ansible to provide a cloud independent Automation tools for managing develop the deployments. How does it work? This tool is called the Automium. We are Europeans we strive for a cloud operators based in Europe for Europe so Automium is the symbol of Europe and Plus automation comes to Auto Automium. So basically Automium is a dashboard or an API set You can use both that interacts with gith hence the term githops which was Copyright here invented by we've works and basically is when you define your infrastructures and you store everything on gith so you are basically doing governance and disaster recovery, etc. All at once using gith. So Automium interacts with these scripts generates them on the fly just getting the information from the user the Configurations and just generating them and storing them into the gith into gith which can be our gith or customers gith on-premise So what does it do next? Okay It basically calls an agent which is set inside the customers tenant on the chosen provider In this case, it's an open stock base, but we have plans to roll out integration with Terraform with VMware and AWS Can be more in the future but basically it startups a VM inside the tenant and Starts to deploy or the infrastructure you need by infrastructure. I mean balancers web servers caching servers database clusters file system clusters Monitoring tools logging tools or whatever it takes to have a proper infrastructure running and monitor all the time. So We will switch to the demo we are very short on time. So you will excuse us if we will Move very quickly around. So basically, this is the interface. We in Automium You have your infrastructure basically basically you have your desktop where you can choose The components you want to run you just drag and drop them into this tool The the web UI is intended for non-experienced users You will see afterwards how you can interact directly with the code and configurations. So basically You just drag and drop them here. We have a mysql cluster with the orchestrator to run automation of master slave replica and Federico now is Showing some components, but the first thing we want to do is the most time-consuming one which is starting centralized logging and monitoring services logging is based on elastic search and Plus we will provide out-of-the-box Kibana and the monitoring is done using Prometheus You can also choose influx DB or whatever you are used to plus Grafana So Federico started the machine now In later on we will show you the open stack horizon interface so you we can see the machines start in booting and and start in the services when you launch a service you are basically defined in the cluster and you You can define cluster-wise that you have you want to have the logging and the monitoring defined for that service by doing this Regardless of the number of nodes you are running in the cluster You will get all these nodes configure configured with an agent specifically for that kind of centralized service to just push all the logs all the metrics Towards your centralized system and you will see them in the dashboard on this view You can see that there is a very simple code for defining what you're doing in the configuration Web form you will be able to Cut and paste your own configurations. You already are running on your premise. So Let's let's move on and add a very simple docker cluster So we can run an application the demo application excuse us will be the usual WordPress because for sake of time we had to To be very easy on the demo Also here we added the monitoring and logging and so now you can apply all the changes you can deploy them and This creates a restore point. So in the future whatever changes you may Apply you can revert to the last stable situation. So you can see the button restore Which means there you have a restore point. You will have more in the future Once you apply more changes This view shows you that these files are creating to git and they just Collect all the information you have a Inserted in the forms or in as code or as design and basically they are stored into git and you can Safely review them or change them. You can also change them in this view You can see the not only the logical architecture, but also the physical architecture. We are waiting here for the private IPs to Show up Please note that you won't see from the UI a console node a console cluster Maybe you can decide whether it's a single node or a cluster of three console is made to provide Auto-discovery so basically every service you bring up Regardless of the number of nodes is registered into console and provides information to the load balancer and to the any kind of balancer you can use so Auto-scaling is basically seamless the balancer knows Automatically when you are adding or removing nodes so you can see console here and information. So you have my SQL orchestrator This is the horizon view as I told you before so you can see the topology. We are bringing up and The instance is Okay Let's move on. Okay. We are very short on time. So you you see the IPs popping up So the main goal for Automium was that we have studied a lot about DevOps Movement principles best practices tools and we thought that the average customer in Italy and Europe is not Ready to just jump on it Alone and they need someone just to bring them take them by the hand and bring them to do something different Very easily so only then they can choose whether they want to study it to investigate and to go more in depth So now you can see from Kibana We are accessing the data for logging stored in Elasticsearch. You can mangle the data You can build your own dashboard some of the dashboards we provide them already baked in So basically you can Instantly see what's going on on your systems this kind of activity usually takes weeks for an experienced system administrators or even experienced system administrators that don't have a complete knowledge of all the domains into cloud native infrastructures so So back here now we are ready to deploy our balancer which is of course The best load balancer round which is HAProxy We define the password the secrets are not stored into Git, of course They are stored into vault. So basically they are referenced from the code from vault This is another machine. You won't see your one managed from here, but you will see from the OpenStack interface So now we are deploying the load balancer and And in a few seconds we will be able so this is the actual Complexity of what we have just deployed so basically the operations take a lot of time to study understand fix and iterate over Manual tasks that can bring to a lot of defects what we want the problem We wanted to solve is to provide an automation tool to do things very quickly very easily and Almost defect-free. I hope so Just from in 20 minutes. Okay, so today the demo is challenging, but it's a very good Opportunity to show what we can do So now it's up and running. Yeah, it's coming Okay You want to show We are waiting for the load balancer the balancer is not coming up. Okay. Let's go to the to the Grafana part in meantime, okay So it's that okay, it's here So the god of demo has looked up on us So basically here you have your WordPress and you can start configuring it and you have my SQL cluster already running So everything is working. So I still have a job is he still has a job and so very It was very I think it was very clear what we can do here the So this is Grafana Basically, you can see instantly from from day zero you can see what's going on on your system So you can monitor them because whenever you have a problem if you don't have these you basically are blind I know it may be Odd for many of you, but for the customers we are seeing they usually have problems They don't even know where the problem is what's going on. How much is that big how much how big is the problem? Etc. So here you can start and have everything in place and do whatever you want with your infrastructure so I Think this this is also more almost it how much time? Really? So if you have any question or something Anybody has a question? Okay Yeah, okay, so The question is if the project is open source the plans are to open source it basically all the tools we are using are terraform Ansible a console vault Docker etc. So the components are all open source The tool itself is not yet open source We are discussing with the legal team in order to provide it open source We want it by the end of the year. I can be pretty sure it will be open source The the basic things about this is are we want to enable multi cloud because these tools usually come as a service inside the provider and we want it to be out of the provider and The other thing is we have to deploy stuff into the tenant because once you centralize it. It's not more Provider independent. So yes being open source is one part of that. So users can Just install it that on their premise, especially the the ones that in the future. We want to run it on against the VMware VMware hypervisors we don't think that cloud depends very much on the hypervisors the cloud cloud is about this kind of operation automation and distribution basically So we think that with this tool can help people jump on the cloud with cloud native architectures easily Any other question? Okay. Thank you very much. Thank you for coming if you want to know more about Automium we have a booth right over there the last one in the dark corner in front of the table. Thank you very much Thank you