I have a java application which is implemented to find providers of an interface using java.util.serviceLoader. When I a add a new provider jar at runtime to the classpath of the application the application is not finding it. If I restart the application the application finds the provider jar. Are there any options for java application to find the provider jar at runtime as I think that is what java serviceLoader is supposed to do.
I am following the instructions in this page.
http://cr.openjdk.java.net/~mr/jigsaw/spec/api/java/util/ServiceLoader.html
Deploying service providers on the class path
Explanation can be found here Creating Extensible Applications (at the end of the tutorial):
Limitations of the ServiceLoader API
The ServiceLoader API is useful, but it has limitations. For example, it is impossible to derive a class from the ServiceLoader class, so you cannot modify its behavior. You can use custom ClassLoader subclasses to change how classes are found, but ServiceLoader itself cannot be extended. Also, the current ServiceLoader class cannot tell your application when new providers are available at runtime. Additionally, you cannot add change-listeners to the loader to find out whether a new provider was placed into an application-specific extension directory.
Seems to apply to Java 8. Don't know if this has changed in the later editions.
Related
I have an ear file with a web module and a ejb module(just used for message driven beans). The ejb module has dependency on web module and it's classes. I would need to load the web module first and then ejb module. But the liberty always loading the ejb module first causing com.ibm.ws.container.service.state.StateChangeException: java.lang.NoClassDefFoundError:
How to control the order of modules loading within the same ear file? On traditional webshpere there is an option called 'Starting weight'. Whichever module has lowest value takes precedence and loads it first. so the application works good on tradition Websphere. However, this property seems missing on Liberty. I already looked at this. It only talks about deploying the multiple war files and their order.
If your EJB module depends on Web, that is bad design. It should be the other way around.
If you have such situation , proper way would be to extract shared classes in to a common jar file, let say mycommon.jar and then put that into ear\lib folder. In this way they will be visible by both modules ejb and web.
If your EJB module depends on the javax web api (e.g. servletRequest) that is even worse, and you should redesign such classes to POJO DTOs.
As a last resort you could try what is described here and add <initialize-in-order>true</initialize-in-order> in your application.xml.
FRowe's solution will not work, as classes are not shared between the applications, so changing load order of apps will not help. Each Java™ EE application has its own class loader in a running Liberty server.
Consider using the ability to control app start order as described here: https://www.openliberty.io/blog/2020/06/05/graphql-open-liberty-20006.html?_ga=2.4728563.17466047.1620833568-1423690488.1614284842&cm_mc_uid=99965752544816136653536&cm_mc_sid_50200000=61078141620909829332#ORDER
You'll have to deploy the war module as an app instead of packaging it within the ear, but you should be able to achieve the proper ordering.
I have a spring-boot app that acts as a small framework for other apps. It provides a couple of JMS queues and a DAO layer to retrieve and store data from a common set of data stores. The problem is that the original developer of this framework app is scanning all the package "com.mycompany" (rather than com.mycompany.framework) so that it can load the beans of the specific app that may be declared under com.mycompany.myapp1 or com.mycompany.myapp2 an which JARs are bundled together with the JARs of the framework.
We only load a single app in the JVM (app1 or app2), but these apps may share other libraries and sometimes we end up with beans in the context that we don't need. (these may be needed in app1 but not in app2)
So, what would be your advice ?
My problem is similar to what was described here:
https://github.com/spring-projects/spring-boot/issues/3300
I am debating if each app should be aware of the framework and load it. Or if the framework should instantiate a class loader and create a new Spring context loading the app specific code as suggested in the link above.
Perhaps you should consider leveraging some of Spring Boot's Auto Configuration capabilities such as #ConditionalOnProperty or #ConditionalOnClass in your framework. That way, you can only actually enable certain beans if and when the application using your framework takes some specific action (e.g. has a given jar on the classpath, or sets a configuration value). For reference check out: http://docs.spring.io/spring-boot/docs/current/reference/htmlsingle/#boot-features-developing-auto-configuration
My project has a set of custom defined annotations that could be present in any bundle deployed in the OSGi 4.3 framework. I want to find any class with these annotations in the classpath. I tried using BundleWiring.listResources(...) and Bundle.loadClass(...) for each class found. I have done some tests with an small set of bundles and it needs almost 200MB of Permanent Generation JVM memory space because all classes are loaded.
Is there a way to free loaded classes PermGen memory space when the program realizes that they does not have these annotations?
Is there a better way to look for annotated classes in an OSGi framework?
I think you should not do annotation scanning as it slows down startup and needs a lot of memory. JEE application servers do annotation scanning at startup to make lazy programmers happy and the result is very annoying (e.g. scan for JPA or EJB annotations).
I guess you are implementing a technology where you can define the rules. I suggest that you should define rules that are similar to these:
Annotate your class
Have a MANIFEST header where the annotated class must be listed.
An even better solution can be to use a custom capability namespace with specified attributes. E.g.:
Provide-Capability: myNamespace;classes=com.foo.myClass1,com.foo.myClass2
In your technology, you should write a BundleTracker that calls:
BundleWiring.getCapabilities("myNamespace");
If the namespace is present, you can find the classes that should be processed.
If you implemented the technology, you can consider an extension to Bnd to fill that MANIFEST header automatically. That extension can be used than when bnd is started from the command line or from build tools like maven.
Btw.: You can use ASM to parse the class bytecode or use the built in possibility of Java to build up AST. Although those could work to solve the memory issue, I still think that you should define the list of classes directly in the MANIFEST header as it makes things much more clear. You can read the MANIFEST headers, you can check the capabilities on webconsole but you cannot do the same with bytecode.
Usually, classpath scanning for annotations is a bad idea in an OSGi context, as the classpath is more like a graph. However, there are situations where this can be useful. Hence, OSGi encourages the usage of the Whiteboard Pattern.
What you could possibly do is register each of these classes as services in the OSGi registry. Then, create a separate bundle that just tracks these services and transforms/manipulates them in some way. For example, this project scans for all classes annotated with #Path and #Provider annotations, and transforms them into Jersey REST APIs.
Does parent_last affect other applications ?
Is my application using parent_last affected by other applications?
The server is set to "Multiple Classloaders" (one for each application IIUC).
In my understanding, this would mean that all classes are loaded from the application, correct?
What happens if another application loads a class present in websphere extensions, and lateron my application uses that class. Will it be reloaded, or will the one from parent be used? (it might be incompatible to the classes in my application).
The websphere manual says:
"By specifying Classes loaded with local class loader first (parent last), your application can override classes contained in the parent class loader, but this action can potentially result in ClassCastException or LinkageErrors if you have mixed use of overridden classes and non-overridden classes."
But I do not understand this potential risk. Can somone please provide examples for this?
Classes loaded by application class loader cannot interfere in any way with those loaded by other applications since application classloaders are isolated from each other.
The quote that you provided warns about potential problems that might occur when libraries have transitive dependencies. For example, your application uses libraryA that depends on libraryB. LibraryA is some sort of "standard" library provided by application server, so you don't package it with the application. Then, if for some reason you need to have libraryB in the application and the version of it is different from the one expected by libraryA, you might have problems with incompatibility.
We run Restlet 2.1 in an OSGi environment (Equinox) as bundle (ie. not as library within a bundle). The problem is that the Restlet Engine does not detect helpers (like converters) that are provided by Restlet extensions. Specifically, the EngineClassLoader#getResources() call does not return any result. The extensions are also deployed as OSGi bundles in the target platform.
Is automatic converter registration actually supposed to work within OSGi environments?
In fact, Restlet supports such feature thanks to a dedicated activator (see the Activator class in the package org.restlet.engine.internal).
This activator introspects bundles to find out the following things:
servers corresponding to registered servers
servers corresponding to registered clients
authenticators corresponding to registered clients
converters
Be aware that to use this feature, we must use the OSGi edition of Restlet since it's the only that has the MANIFEST file of the org.restlet bundle with the activator class specified. Otherwise you don't have to care about the bundle loading order...
Hope it helps you.
Thierry
Unless the Restlet-bundle explicitly imports the packages that contain the extensions (and I doubt it does, and it shouldn't), it wouldn't be able to load them, because bundles have isolated class-spaces.
A possible solution would be to provide the extensions as fragments attached to the Restlet-bundle. Thus, if you make it use the bundle-classloader (the documentation says this can be done by setting the Engines classloader), it would be able to load classes from the fragments.
Indeed it doesn't quite work for OSGi, as it depends on the ability to see the entire class space.
The way to do this in OSGi would be to use the service registry for the extensions, but that only works for OSGi aware libraries.
There is some help on the way: In the recently released OSGi 5 (Service Loader Mediator) there will be support to 'bridge' META-INF/services (I don't know if Restlet uses those, though) onto OSGi services, so 'legacy' libraries should work well within OSGi.
There is an implementation in Apache Aries called Spi-Fly. I looked at it briefly a while back. It might do the trick for you, it might not.