Jakarta EE JakartaOne Livestream- September 2019, 27 Apr 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Building Interoperable Microservices w/ Eclipse MicroProfile | JakartaOne Livestream - Sept 2019

Description

Speaker: Ivar Grimstad

Eclipse MicroProfile is a collection of community-driven open source specifications that define an enterprise Java microservices platform. This session gives an introduction to Eclipse MicroProfile and the tools available to get started building portable microservices with a minimum of effort. The features of MicroProfile will be in explained in a down-to-earth and easily understandable way.

///GET SOCIAL!
Subscribe: https://www.youtube.com/user/EclipseFdn
Follow us on Twitter: https://twitter.com/JakartaEE
Follow the conference: https://twitter.com/JakartaOneConf
Like us on Facebook: https://www.facebook.com/JakartaEE/
Become a member of our Jakarta Tech Talks Meetup: https://www.meetup.com/jakartatechtal...
Join us on LinkedIn: https://www.linkedin.com/groups/13597...
Tag us: #JakartaEE #JakartaOneLivestream

A

B

And real-live: oh there.

A

Was no countdown hello, Iver Iver is right or even Eve.

B

Are uppity the way I interpret.

A

So now the most important question: where is your Duke tattoo.

B

Sorry where's my do detective Oh. My detective is a hero on my arm, of course, yeah. Okay,.

A

Hard get yeah, okay,.

B

A

Now, that's you, okay! Now you would.

C

A

To tell us about interoperable micro services with Mike reporter right, yeah I am so let's go then share your screen and I'm really curious about your presentation. Oh thank.

B

You so just start sharing here.

B

So there hope you can see this, this presentation about building interoperable micro services with the clips micro profile, I'm Ivar Grinstead I'm, currently in the u.s. at the California, just north of Sacramento I'm in a pretty budget motel. So but the Wi-Fi seems pretty good, so I hope you'll get through it.

B

So some words about me and the couch champion and an Oracle groundbreaker ambassador I'm, a member of the executive committee in the JTP, as well as in the Jakara II working group I'm, also be easy for JP MC explanation.

B

Besides that, I'm also active in Apache with the Apache NetBeans and I'm running a local Java user group in malmö and Sweden.

B

So I won't be able to see any chests or anything, so so just right um and Adam or Tonya will gather up some questions at the end. So this talk is about micro profile and, as you see this subtitle or the the katakana texture underneath the low type here is optimizing, Enterprise Java and why would I have micro profile here in the Jakara one livestream event? Well, they are kind of related to each other, which we'll see in a minute.

B

So micro profile is officially called Eclipse micro profile, and that is because it is an eclipse project and the first project. You can look at the project page to find the details about it. There also take run in the browser after work, so we can familiar us a little bit with it. But what you see here in the mission is about optimizing Enterprise Java for micro services, so it's about all about micro services and an ecosystem around mic services and be able to create micro services.

B

Easy using the technologists were familiar with from the Java EE or jacala world, so much praval. The latest version available now is microphone. 3 though it came out this summer, there is a freedom one release coming up in a month or so, but this is the latest version out there and, as you see it's a lot of api's and there is the four boxes on the bottom arrow CDI JSONP check, Soros and Jason B. There are familiar technologies from Java EE, so when they create a micro profile, they decided that we need the technologies.

B

That is the bare minimum for creating a reasonable micro service using Java, EE, technologists and CDI gives you everything you need with context. Dependency, injection and kind of the glue gluing everything together.

B

P gives you all the need for adjacent processing. Jax-Rs is the API for restful web services and since we're usually making micro services with REST API. So this is the technology for that and Jason B for Jason binding. So you can bind java, okay Jason, so so these are the Java EE 8 technologies that multiple file are built upon.

B

On top of these, we have api's for open tracing, open api rest client configuration for tolerance, metrics, JSON, web tokens and health checks and I'll go through these API in more detail over this presentation and probably show demos and with over.

B

So since microfauna is a set of API. That wouldn't mean anything if it didn't have any implementations, and there are a lot of implementations about marking profile. There are more than these, but these are some of them that I've gathered together here and you can see most the big corporations and vendors within the Java EE space is represented. Here you have a red ass with their thorn tail or the quirkiest products. You have IBM with open, Liberty or registered Liberty as well. You know, Tommy tribe is attached to tell me pyaara with their player micro.

B

You have hell of them from Oracle and a lot of other implementations out there. So it's free to choose whatever implementations you want down.

B

And also all of these give you some way of running your services from the command line or a single Java process like you would expect when you create micro services- and this has been made popular by by spring boots, that came around with with creating a main java file, where you can execute your services right in Java char, and then your jar file and that's how I'm run it, and this is no different for these. You can also run these applications as a package them lower files and run them in the container.

B

If you want that, or you can just access your container into the jar and run it from the command line. Just as this and the older vendors or product there provide some capabilities as doing this, but as I said microphone file is about the API and the advice are built up with a specification document that specifies how the technology should work there are.

B

There is a API artifact that has some interfaces and classes in it and also some Java Doc's who can figure out how how to use them, and there is also a test suite called a TCK or test compatibility, quick kit. You can verify that the implementation actually does what it's supposed to do, and this is very similar to what's done in the JTP and also in the Jakara ii, where we have this kind of triage of other things.

B

You have the specification, you have the API artifacts, you have an implementation and some tests ways to test it.

B

So I'd like to be to just show you how to get started with eclipse my kin profile and the best way to get started is to navigate to the microphone field of I/o web page. Let's do that so here on in my browser, so in the microphone file you can find information here about the latest version and who's involved and get some some resources, some white papers and stuff. You can find all the projects listed in the project tab.

B

If you want to read up on the details about the projects so they're listing here, you can see there are a lot of them. There are samples there are. Some of them are included in the platform. Some of them are on the way. Yeah. You have some support for reactor self-reactive messaging that hasn't made it into a a kind of a collective version or microphone.

B

Yes, but it's there and you have some resources and the most interesting thing for now, if the starter and free also, you have the use the Springville tooling for for creating your web services, you're familiar with Liz and what you do here. Is you call your application? Give you some maven coordinates. You select a Java version, some of them. You can actually have a support, 1211 or and above now, Justin to have a Java right here. You just which version or micro fall you want to run and and depending on which version you create.

B

You see you get different alternatives, so here you have most of the implementations up there for one or two, as you increase the diversity numbers, the implementation will be a little bit fewer. A lot of them are supporting theater too, and the latest version free, throw is spurred by container and of Liberty. Let's just generate the open Liberty one here and what you can see here also that you can get examples generated for you.

B

So if you're creating a application from scratch- and you want to write this in your way- you just uncheck these boxes and you won't get the samples gaudium. But if you're like like me now- and you just want to look at how it's done just leave them there and you'll get all the the sample code in your project and it's as simple as downloading the zip and unzipping it so and since I didn't trust. My internet here, I've done it earlier today or yesterday.

B

Actually so I openly, the Dukes one which has had have prepared and what get in in this zip file is two services and these service a has, has examples for most of the implementations there and I'll go through them in the context when I go through the API themselves, you see you have work for the rest, client, the configuration health check metrics for the fault, tolerance and security, and also you have a second service which is more of a helper service, just to be able to test the security stuff.

B

So, oh look at this when we go through the API afterwards,.

B

B

So that was the starter. That's that's the best way to get started with Mike profile. So if you're new to my profile and want to take something home from this presentation, this is the slide. You should copy or take screenshots, and do it now and or download it later. But this is just the link to remember to get started with the microphone.

B

Yeah, so just then with that one and when we are talking about all these API for health checks and configuration etcetera, it's very easy to come into these notion of much service tablets, because all of these are in one way or the other, implementing some Mike service pattern and a slide here showing how they are related. So we have a list of a lot of micro service patterns and, as you see via the Mac profile, APRs implements these parents in a it.

B

B

You can see here so I'll go through these patterns, one by one and explain the the technologist environment and the paradigm I'm doing here now is the external configuration they'll check, API the application, metrics circuit-breaker access token and distributed tracing.

B

So external configuration, the proper statement for autonomous cooperation is how to enable a service to run in multiple environments without multiplication, and that means that you don't want to rebuild your jar or your application just because you're configuring it with a different data source for sample when you're moving it through environments yeah, you want it to be built once tested as it is, and just configured through the environment and then finally go into production.

B

So you actually test the stuff that you're putting out there and it's only configuration and it should be retrieved from outside of the service and the microphone configuration supports. This you get for free with this specification into requires that you have a support for for setting and a configuration file which you can build into your your application. So it's suitable for what, when you have properties that don't change or properties that you want set for your local landmark.

B

So you can have this this profile in your IDE, and- and that is how you run this when you run locally well and if you said some ability into a docker container, and you want to set some environment variables there, you can do that and they, these properties will override the things in the properties file and over this again, you have the system properties which you can set in your test environment.

B

We simply in your kubernetes or whatever you're running your your services in set some system properties there and they will override the environment variables and the application configuration files which is packaged in your file. So if they have there's the priorities set, and you can also create your own configuration sources such as, if you have some some sources in, there are really some secure environment or you have them in a database or somewhere else in github or in a text file.

B

You can just create implement a a pretty simple interface and give it a higher priority than system properties, and it will use that one, and there are also vendors that provide these. These comfort sources of the box, for you.

B

So the configuration.

B

Api, for my services is one of the implements the external configuration pattern. So let's look at this so in the flower I downloaded from the the starter you can see, there is a configure and I have a config test controller, and what this one does is that it injects some copper properties and you can see that the comfort properties from the much prefer comfy project. So what you do there is to inject a a company property called objective value.

B

They can have different types and invasive or sprains and strains and names and erase and all kinds of properties that you need. This is your say, a simple string string we use for, for example, and you can use this injective value in your application and run this application a little later, and you can also see that in the the maintain order you have the investiture injective value prop state set here. So this is one that we picked up and last you override it with a a system, property or a and wireless variable.

B

That's not how it's done. You can just inject it. You can also do it programmatically if you want, but since we have CDI, it's not no reason not to use it.

B

Check API and the promisee this API is how to detect that a running service instance is unable or able to handle requests and the microphone health API helps with this, and there we have the health checked, which is updated for Russian, 3-row and I'll show a little bit about what's added there.

B

Let's go back to the code again and as you'd expect in the in the health packets, you can find the a service health checked. That's that just returns a health check response and, as you notice here, the health sanitation is defecated, so this sample is actually a little bit outdated. This is in version 3. Do you got some some more cooler feature here? I'll show how this looks like so just create a it's alive.

B

So what I did in here was to create a health check which is annotated with liveness rather than with health. That means that this one can be used for aliveness probe, and this is a very useful in given a test to to give the environment the message as hey I'm alive. Don't kill me, so this should be a fairly quick service. It's just the up and answering very fast when you know that a and nothing was wrong.

B

I was running, but just because I'm up and running doesn't necessarily mean that I'm able to have a request. Yes, so so maybe you shouldn't send me any traffic yet, and so out of that we have the readiness test so.

B

So the readiness object is to say, give them wrong and say: ok, no I'm ready to access traffic. So if you're in the line, let's probe, you just say that hey I'm up, don't don't kill me yes, and for the redness you can check, hey I have the other services or resource of database connections or whatever you need to be able to accept traffic, or maybe you have to build some caches or something, and this can be a longer running a thing before you give a ready signal.

B

So let's try and run this one and I'll just rebuild it quickly.

B

So wherever you have the you, you can still use the health check and the health check will combine the lightness and the redness probes for you, but wherever you're using healthcare today you should consider moving earth to the Linus probe or the redness probe when you upgrade to too much bro feel free to go. So luckily, I did this test this morning because they just shipped a new of Liberty runtime, and it has to download that one, but I did this morning.

B

So we don't have to wait for me downloading all this kind of stuff, so it should be ready up and running pretty soon.

B

Okay, let's just leave it there revisit after the next pattern,.

B

So the next one we'll look at is application metrics and the problem statement for application metrics is how to understand the behavior of an application and troubleshoot the problems and the microphone. Metrics project helps us with that, and it has been updated for Russian fellow. So, let's look at if it's being fair. Yes, so this and start this one.

B

As you see, I started from command line, just run it here and while it starts up, let's go and look at how matrix is done in my profile.

B

So here here, I have a a unfulfilled metric I'm, injecting a metric here which is of type counter, so I can count things and I'm also using at at time the annotation. So this reckless air will be timed I'm doing some thread sleeves here just to get some statistics in it and I'm doing.

B

If we call the increment endpoint, it will increment the counter. So so we can get data in the macro. Metrics and I. Go ahead, have a gauge here. That shows the the counter.

B

So it looks like it's up and running so let's go and look at it, I think it wasn't 88-81.

B

Yes, it's up and running so.

B

Let's go to the data hello.

B

So it world endpoints that is up and running and let's look it at the health checks that we looked at earlier. So if I go to help, you see that it. It shows me a lioness, checked a resinous checked and only regular health check. So if I only want to have the the aliveness probe, I can go in to the sub resource, live I'm actually give the lioness.

B

So this is kind of the the the posture to give to the environment, to check if your services a lot and to check if it's ready, you go into slash, ready and you'll, get the readiness test.

B

So and the metric standpoint is slash metrics, so here we get the committee s based all the metrics you need for for for monitoring your application and, and most of these are of type space or or they could be vendor. They could be some something else. So if I want to just see the base, metrics I say what's required by the by the yeah certification. If I want to have the application metrics, which is the one I put into the application, you see that there are no metrics yet because I haven't touched.

B

The end point I'll just check what it was in the fridge and timed. So if I go into the.

B

Most creating one in a great.

B

Kind yeah this one was one that had a sweat sleep. So if I go in here and refresh this, one should be showing ya the application. Metric Seneca see the.

B

Time request that all this is what you get out from from just annotating it with at times. If we want the the counter to increment, you see the counter. Endpoint is zero here. What was it increment and I'll just open the increment.

B

So if I go in there, they just say 1. If I hit the refresh a couple of times, you say that it's counting, so this is how it get. You can monitor your application usage using the ADA's metrics.

B

What also is useful here is actually not one of the API so I'm showing today, but it's for a might for a while, and that is you get open, API specification there, so you can just if you don't have to write all these kind of open, API demo files or do any swagger or annotations unless you you need to give some extra information here. This is something you get for free. You don't have to annotate anything to get it with the open Liberty you actually get in UI with it as well.

B

So so yeah you get all all the swagger here for all the endpoints and then you can run it directly from the processor. So it's pretty cool.

B

Like, let's look at the circuit breaker pattern, so the secretary break the pattern. The problem statement is how to prevent a network or service failure from creating to other services, and the circuit. Breaker is something you know from your home that yeah. If a electric circuit is overloaded, it opened up the turkey, so no current is going through it so to prevent from from burning down around so a simple burger in my services works the same way.

B

If a service can't Alex at request, you can have a smooth way of discovering it, so you want to affect the other person service. More than mixture.

B

So and the among other things also supports circuit breakers.

B

Let's look at the circuit breaker.

B

That's under the resilient package and in the resilient controller we have a lot of cool stuff going on. First of all, it's just a regular Jack's, rest or rest service. You have the fallback handler that says that if this one for some reason, this circus times out in this case- or something else happens, usually have this for Blake method. And here it's just a method. Call that a returning for bug answer due to time out and you can see that timer is set to 500.

B

Maybe I only have a thread sleep to 7:00 or adjusted to make sure that it's actually happening, and so, if you're going to hit the endpoint resilience.

B

There's one or two off one: you know.

B

With an S you're going there and it's going to wait for father and a half a second 500 milliseconds, and it says full like answer due to timer, and that is the good times house after half a second but it sleeps for 0.7 seconds, and that's why I never got the s in there if it gets to four for my camera. So this is a very easy way, easy way to showing how I'll be yeah yeah hold. I can tie my circuit breaker turn.

B

This will be that this thread flip is actually you're calling a service that just takes you along.

B

So token, that is how to communicate the identity of the requester to the services that handle the request and in order to let's just go through very quickly some some simple ways of Fanta keishon and we all know the basic authentication mechanism, which is an authorization header, written request, saying basic- and you have your word, your password there yeah it's hashed foot, so it's you just send a password or token in your request coming with the basic prefix and it sends with every request for earth too. It's a pretty similar.

B

It's just that you send something else in the authorization header you send error and a little bit strong, you're interested, but still a password and so for for the basic authentication. It's it's basic, something and for our expiry something it's this strong your encrypted. Then you have all these this way, so refresh tokens and etc, etc, but but the the actual security mechanism is pretty much exactly the same.

B

Let me have the JSON web tokens and JSON web tokens are also exactly the same as with OAuth it's just it's a much longer token, and this is a little bit easier to to read if we color code it. So you have the first purple part, which is a header. They only have the green part, which is the content of the token and the last part, which is a signature, and let's look at how these are built up.

B

So the Heather is a JSON object, it's basically four euro encoded and it gives you information about the algorithm used and the type of token. And then you have the body or payload of the token, which is also Jason, basically for Euro encoded, and you have some standard interests that have to be there.

B

According to the specification- and you have some custom embrace that you can add, if you you want to have some information that your services can make you so and then at the end, you have a binary signature, which is the actual actual signature of the payload and the header. So if somebody tampers with the the entries add a new custom entries or change to role or something the signature was.

B

And this is our: it will look in clear-text. It was just a basic, sporty encoded.

B

So in this case you have the algorithm in they have earth and in the great text you have this type of talk in the discuss of it and you have the exploration etc, and you have some custom attributes such as username here, for example, and the digital signature which is placed by by the issuer of the token and the Jason were talking. Propagation specification gives you only the infrastructure around it and the API to use this very simple. So look at that one.

B

So for the secure service we have a secure controller and.

B

What we have here, it's still just a regular jax-rs resource, just passed out, tourists and stuff, and in this one we have some code to generate a token, and this would usually be done by a token provider.

B

So this is not something you do very often yourself, but sometimes the time you have to generate the token and you use whatever tooling or technologist you you have there to do that, and and this one Crysta header, the actual token, essentially the the required attributes here and can add additional claims, such as the custom values or groups that this particular user is part of and just sign it.

B

Video key and and return a a cheerless token, and then I can use a a regular rest client to call a service, and here I'm calling the the other service that is included in this example and I just set the authorization header with bearer and gets a token string here, another what I'm doing is I'm reading some response out of the d cultivate this service.

B

So look look at the other service.

B

The one I'm calling one called service B and here I- have the protection controller, which is the one I'm calling and here I'm, just injecting a claim with the custom value and.

B

What I'm doing with it is that when I get it a at get request, I check that the role of the caller hash is protected and then I return. The text same protected resource custom claim is custom, gets value so I'm using two elements.

B

Woman token there in the Admiralty loud I check that the role of the caller is protected, and this is a java.security annotation, so it cooks into Java security and also it is pulling out a claim from a from a custom value from the token and returning it as a text and as I see I'm just running it out here and when I built the token the custom value is set to it to a Jesse specific value.

B

So if I have to just start the second service.

A

B

And then I should call the the protected one in the service a.

B

And I think it's called protected.

B

Looks not that, let's see what it's called, we got it. It's.

B

Secured, of course,.

B

I'm generating utopium and then I'm calling the other service. Let's see how its works.

B

B

Well, everything came for everything. Okay, it was nothing running, let's say.

B

There you go claim wet value within the date of utopia. Token was, and this is a text I got from the other services, yeah I'm, creating the the token in my service, a and I'm propagating it to service, be extracting some value and returning that value and displaying in here.

B

The left today is distributed tracing, and that is how to understand the behavior of notation and troubleshoot problems and a pretty common tool using for that kind of tracing his 15. You can also use yeah, ger or other tools supporting that protocol, but 15 is one our demo here, and this is open. Tracing is support by the open tracing specification and, let's look at a more intense example.

B

This is a part of a lab and which are created from microphone and you're free to download it. I'll show the link to it afterwards.

B

I won't be able to run it today, though, because when I tested it earlier when I set these tokens in my browser, I kind of drop out of the crowd cast screaming, and it won't matter happen, but I'll go through the code, so what's happening here is that I have a capitalized service running on PI or a micro, and what it does is to take in a name or a word and capitalize the first letter. So if you handle in lowercase do care, it won't return due to the capital.

B

D and I have a birthday service that calculates a birthday. So if you as a deck parameter, the birthday, Duke I, think it's May, 23rd 1995.

B

It will calculate how many days it is left until oops next next birthday and how many days it was since its birthday, and it will also return its age and it will return his name, which has a capital B. Even though I applied her with a locust D and that's because it calls the capital I service to get the catamites do so service.

B

The British service calls the Kepler service and for the first service, I have a Hello service that, because, with name is Duke and print the text and extract information from what it gets from a birthday service and the virtual service runs turn tail and the hello service turns on Oakland Liberty and the token that is received by the the hello service is propagated to the birthday service and unprotonated further on to the cap, my service.

B

So you can follow it several steps, but, as I said, unfortunately, I won't be able to run it since I'm using a token as input to the hello service, and it will kill the big broadcast. But we'll look at the code.

B

So, let's first the capital, I service and the actual service is super simple. It just takes up get to the first letter, but what's interesting here is that it has the after all, so loud after business says that hey, you need to be protected, so so it won't allow anything in there unless it's protected, and it's also set up that the protected role is secured and it's using the microphone JWT as authentication method.

B

The British service is set up similar, so have the person resource. It also has the Rose allowed and protected, and also has the the microphone called guarantee.

B

Authentication method and the protector which role, but also here, is that it has some configuration where it says to the microphone. Rest, client, propagate headers authorization. So this property here tells the rest client to propagate the JSON web tokens or the authorization header that contains adjustable tokens to the next client and that's the way it is propagated from from one service to the other and and here I'm, using the the.

B

Matter of a rest client to do they call to the the capitalized service, so I'm, injecting here a capitalist service and the capitalized service is just an interface with the generous annotations. It says at pass capitalized and it has the after register risk line, so it will generate the proxy that does the actual call to you and also has the register. Client errors.

B

That says, hey, add whatever headers, that I get from the clients that are mentioned in the propagation configuration, and is that that it's active word in and it has a phobic method, so Arum using the the circuit breaker, also so-called the method, essays and just return the word locusts unless the services and calling it's nothing running and in the company properties. I also say that the endpoint URL for the kepler service is here and that's picked up by the rest light. So it's a pretty cool, easy way of propagating a.

B

Although the security information to the next service and I'm doing the same thing in the hello controller, where I also had some tracing enable I'm doing some some some injections here and I have a birthday service which is set up the same way as the at the kepler service, it's just a interface with the register s climb and climb headers and the patient's.

B

It also has some fallbacks and some rate rising unless a certain services are running so it'll try it 30 times before it gives up and then returns just minus one for for the prophetess, and it also has the configuration for for propagate the authorization headers and they ask the URL for the birthday service.

B

So service I also have sitting running on my local computer, so I run earlier this morning. So if I go in here and say fine phrases, you can see that that's the change.

B

Just me show anything: okay, so if a run the service, it would show the based on the the tracing annotation that I have here. It will trace the yellow service. You can also do that programmatically, which are done in the capitalized service.

B

In the birthdate service, where okay, there we go here, I'm doing some some programmatic tracing, so I see that I'm calling recapitalized. So we can actually trace service calls, even though the the actual class there or the service there isn't hasn't any configuration for tracing at all. As long as I'm tracing this one I can add this code for for tracing the Calphalon service, even though I don't have any tracing there.

B

So some of microphone implements a lot of the important patterns useful for microservices so and it's very easy to get up and running and if you used to using spring boots, for example, for micro-services you'll find this very familiar and probably a little easier because it's less configuration it's just there. It's just super intuitive to to get started with it.

B

Try to actually examples on the starter page just play around with it and you'll see how easy it is to take all the code I'm showing here today of I've been using different vendors implementations and all of it is portable.

B

You can move it across services across the implementation, from the banners in any way you want, and also since micro profile and jakara II is part of the a transformation Kevin mentioned in his talk earlier today that we there will be, and there are synergies there are the same people working in both these communities and I usually stress that it's Chikara, EE and micro profile. It's not vs. or it's not. Two sorts of technologies.

B

Micro profile builds upon Jakarta II and after a while, we will start thinking of. How are we going to make this more clearer and how are these two projects going to co-exist together and how are they going to benefit from each other, because there are so much to gain from from some of these products in microphone fall would be very useful to maybe have in some kind of witness, based economy package, or vice versa.

B

So these the partners on we talk know that today you can find them on Mike services or dial a lots of samples when I get up feel free to contact me on Twitter or or github or anywhere else to go to. If you want to have some question or something microphone does IO is a place to start with my profile, and especially the store, recommend just download the zip file there and look at the code to run it and see how it works and the Chicago key is I, find information, but ricotta II.

B

That was what I had thank you for listening and I'll. Just stop sharing my screen and go back to you and see. If there are any questions or something.

C

Okay, I'm sure there are questions for this.

C

B

You're breaking up a little.

C

B

You're breaking up so I can't hear it, but I see. There are four questions here in the air.

A

B

Just read them up and answer them, so it says without JPA as part of the project. How do you most multiple applications handle persistence? Is there any preference to stay to bc or import your own persistent provider into the application? Well, I, say: I. Think it's!

B

It's probably up to the project insult, I'm a I-I-I-I I'd likely do it as simple as possible, so I usually use JPA or or maybe hibernate, if you're using the Quercus product, for example, a look at apache panache that generates stuff for you, similar to spring data, or you can write your own JDBC, but it's more or less up to the project. If you're, usually, no, no sequel, look at the Jakarta, no secret projects see if something there can be used.

B

So it's it's not specified by micro profiles, but all these products are built on top of Java EE and Jakarta EE technologists. So if you're running, for example, open Liberty and it supports the full Jakari.

B

Profiles so so you have all the technologies they're available for you and that's the same with turntable and under R and the others as well. So it's just whatever is there.

B

So how can you unit s this? The restaurant company himself, I? Guess it's the same way you you test any rest service. Microphone, doesn't write in the infrastructure for testing. Yes, is that it's a question that is the dialogue I've had with the spring community, because spring brings a lot of good support for testing in Java EE Jakari microphone. We chosen to kind of keep it out of it, so so you can use any any environment around it like a Killian or something. So it's basically the way you do testing today.

B

It's running in from and for the out box pattern in mike koval I'm, not sure you just have to check the mainland and.

B

Can you provide a few examples when should I use Takara II instead of microphone pole and vice versa? I'll say: yes, both it's whatever you need for technologists.

B

Do it so you can use Michael broke up with check out AE. So so, if you have a Jakari application and you want to have health checks, just have sex with Akane, is it possible that specifications diverge and a Jakarta, a versa prank of all I would say no, because it's the same people there and and.

B

That match should we will make sure that that don't happens because the and will be taken care of by the by sensible people. I think you know, I, don't think that's gonna be able, and then Reza has a question here.

B

What's your new role as the clitoral nation, so as of October 1st I will start after the pasty Jakari a developer advocate at the Crystal nation, so that will be Mon no role there and I hope to be out there meeting a lot of you at conferences and meetups like this one and talk about all this stuff, but Jacare microphone, Paul and Mike services and Prussia our native stuff. Everything about this, so yeah I, guess I think I got them all.

C

Thank You, Weaver I hope my voice is coming true now loud and clear. If it's not.

B

C

Is okay, fantastic, so do.

B

We have a say: we have more than thousand people listening to us. I know.

C

Look at that, it's amazing thanks. Everyone for joining and registering.

A

C

This amazing session and marking this day, because it's pretty pretty big in Jakarta a community, and at this point we will give you a few minutes of a break and then we'll join you on our next session with Adam bean and I will see you all there. Thank you so much thank.

A