 Hello, dear audience. My name is Michel Stilkmaier. I'm from Amsterdam, the Netherlands. I work with Quality and I'm presenting for you monitoring your SOA stack with Prometheus and Grafane. I'm sorry I can't attend the event in person, so I will do it virtually remote behind my desktop. I hope you enjoy this session. So in nowadays monitoring, there are a lot of tools around commercial tools which are licensed to be used, but also some open source tools and well I will discuss open source used by commercial off the shelf license product which Oracle has also. It's the SOA stack the SOA suite in fact. And when you look at the different tooling around open source there are lots more of course but we're on Promcom so I will talk about that. You have Grafana for dashboard visualization and Prometheus to scrape and gather diagnostics for your monitoring. And same for your log file analytics. You have your cabana to visualize all your log files and log stack to gather all the log entries and do something with it. And if you look a bit closer into how Oracle has come more into the open source and the open container and open cloud movement you can see they made their traditional application server WebLogic container ready actually. So it's certified to run at a docker and Kubernetes and all the different flavors of Kubernetes open shift and all the others. But it's also certified for more the open container initiative like Cryo and ContainerD and Cata containers. And looking at the integration part so integrate all this technology Oracle also creates or has some projects around the around an operator for WebLogic and also in terms of monitoring it has written an exporter for Prometheus and also for ElasticStack. So you can see that the platform is ready to be monitored open source. Now on top of WebLogic you can run all kinds of other products from Oracle you can build your own Java applications but you can also use commercial off the shelf products like WebLogic and Java application server but also build their own software and this software is the Oracle SOA suite and this consists of certain kind of products about the main products running our Oracle service bus which is an enterprise service bus and the Oracle SOA BIPL execution language which is an enrichment language for getting more data into your transaction. This is a schematic overview of how the SOA architecture looks like with all kinds of open standards like the service component architecture composites which consists of some of components like an interface and an autonomy process within it. And the infrastructure is built on WebLogic and on top of that it's all Java based, Java connection architecture well I already told you BIPL some internal routing with mediator and some database related stuff like a dehydration store to dehydrate processes which are running maybe for some long time. And of course the web based stuff which is for the administration. And the important part for your WebLogic and for your monitoring of course is that WebLogic enables to control over REST. So the REST management API within WebLogic is also good for your monitoring stack for your monitoring tool. It's a REST API and you can do a lot of stuff with it. You can configure it, you can start, stop it, deploy applications, monitoring and also the operator makes use for it. And you don't need any client for it, it's just over HTTP. And this is important for your Caravana and Prometheus because it also uses this interface. So let's focus a little bit about monitoring diagnostics and everything within WebLogic SOA. So WebLogic provides a lot of frameworks within which can be used by Prometheus to extract all the metrics out of it. You have the WebLogic diagnostic framework and some additional frameworks are added to that. And also because SOA is a bit of another product which lands on WebLogic, you need some additional tooling or called diagnostics, logging, dynamic monitoring service and selective tracing. And this all can be used by Prometheus to make an extract and to share it within Caravana. So specific diagnostic tools for SOA can be extracted out of the WebLogic diagnostic framework within with a combination of other tooling DMS metrics dynamic monitoring system and have some dumps of logs, metrics and images which can be used for Prometheus to make a logical understanding of how the system is doing at that moment. Now I'm going to show you an overview of how you can set up Prometheus and Caravana for your SOA stack. So you have a few components in here of your traditional Prometheus and Caravana in your Kubernetes environment using an operator for your Prometheus perhaps to install everything and to control it and also to publish your metrics somewhere. But specific for WebLogic you have some additional stuff which is needed. You need a war file to install in your WebLogic domain. You have to repackage or rebuild it with a proper configuration and you can use some JSON dashboard imports which are already configured and done from the Oracle side or you can configure them yourself. And you can also use an SQL agent part for querying the SOA repository. I already mentioned the Hydration Store and the Aurora Hospital to get all the diagnostics metrics out of that. Now this is a schematic overview of how Prometheus in general works but also how it works within a typical Java so environment. So an important part for that is the WLS WebLogic JMEX exporter which exports all the MB metrics out of the Java application or the SOA stack. Now here you have a flow of how these metrics end up in Prometheus and finally of course in Krofana. I already told the WebLogic exporter war file, the Web archive, you deploy it into your WebLogic domain and with this war file you can export all the MB and JMEX metrics into you can scrape them from the exporter into Prometheus. And if you have set that up you can already see all the metrics in the Prometheus console and then you have to create a data source for Krofana to visualize all the metrics into your dashboards. And looking into Prometheus you can look at the endpoints of the metrics and there you can see all the metrics which are collected with this endpoint. The scraping which is done from the WebLogic domain into Prometheus. And here you can see that the WebLogic admin server is scraped but also one of the managed servers which is running in a pod. When it comes to customization, Krofana is the tool to be used within Prometheus. And within WebLogic you can use your customized dashboards or using templates with the JSON format to import them into your Krofana. But you can also write your own queries and create your own dashboards. So here's an example of how I created a query around the WebLogic server that can be running or failed or some other stilt and you can build this into your Krofana dashboard. And with some value mappings you can give them values, integer values and correspond some text which also matches with the WebLogic running and starting and shutdown states. Okay, that's about it. This was my session, my lightning session. I hope you enjoyed it. It was a short session and a short overview of what's possible. But if you want to know more, please contact me on my channels. Thank you very much. Have a wonderful day.