I am trying to run GraphDB with multiple rulesets. GraphDB offers the possibility to add rulesets via SPARQL. However, I realized that no ruleset added via SPARQL is ever activated. Only a ruleset uploaded at repository creation is activated. Later one can even replace the rulesets config. But the rule engine only processes this ruleset (obviously because it is somehow hooked to the ruleset's name) and never one later added. Is this an implementation issue, a configuration issue or maybe not yet implemented?
Thank you in advance for any hint!
Related
Everytime I build my Service Fabric solution in Visual Studio, the ApplicationManifest and ServiceManifest are modified and Default Services entries are added for all Reliable Actors in my solution.
How can I prevent that?
Note: Default Services are not created for the Reliable Services which is what I am expecting.
Ultimately, our goal is to create an "empty" Named Application instance every time a new tenant/customer signs up through our Web Portal (one instance of the application type per tenant).
Then the tenant can enable/disable features through our Web Portal which translates to creating/removing an instance of an Actor/Service types inside the tenant's application instance.
Thanks for providing guidance if this is not the right way to implement isolation in multi-tenancy scenario in Service Fabric.
Thanks Kiryl... I finally found an article that explains the way to disable the generation of in the application manifest while building the Service Fabric application in Visual Studio.
Open the .csproj of the Actor project and modify the following property to "false"
<UpdateServiceFabricManifestEnabled>false</UpdateServiceFabricManifestEnabled>
You can find the article here: https://github.com/Azure/service-fabric-issues/issues/271
Cheers!
I'd say you can't... There was a time I was trying to set my own settings for partitioning scheme(switch the default one to Named partitioning or change low/high key for UniformInt64) but VS was overriding them no matter what I did. Turned out it was an expected behavior as I 'misunderstood' how partitioning is applied with actors, so VS gladly was 'fixing' my mistakes. See Can't change partitioning scheme for actors for the reference. The bottom line is, seems like VS adjusts Actors Service manifest to preserve the holy purpose of actors as a pattern.
As for multi-tenant application, consider deploying a separate app type with its own settings per each customer/tenant.
P.S. I was able to get what I want by deploying Actors Service manually rather than via VS, but that's another story.
We are planning to use Drools/JBoss BRMS 6 for business rules management. Our plan is to write rules using the workbench, deploy the rules package in multiple Execution Servers and allow applications to access the Rules package by making calls to the REST API. We do not have any Java wrappers or custom classes in between the calling applications and the rules package.
I am trying to incorporate some logging into the rules engine. I understand that there are EventListener interfaces that can be implemented.
Please would you provide some information/guidance on how to implement Listeners in our kind of set up? Where will I create and store the Java classes/methods that would implement Event Listeners?
How can a calling application insert an Event Listener into the session? Will it be part of the xml/json payload?
Thanks
1. Where to implement the listeners?
The listeners must be obviously implemented in Java. One simple place I found to put those implementation is in a separate maven project. After all, a project in the kie-workbench is a maven project itself. So you can create a separate project (outside the kie-workbench) implement the listeners you want to and then add this new project as a dependency in your kie-workbench's project (check the documentation on how to do that).
The only problem I found with this approach is that once you defined the dependency between your projects, the kie-workbench will scan every single class of it and of any other dependency it has. Check this link for more information.
So, if your listener project doesn't have too many dependencies, you should be good to go. Please note that, in theory, you could add any kie/drools dependency you have in your listener project as <scope>provided</scope>.
2. How can I configure these listeners?
A trick that I always use is to have what I call a "configuration" rule to do this kind of job.
A "Configuration" rule is a rule without LHS (and, if you are distrustful, a high salience). This kind of rules are guaranteed to be executed only once. Just make sure that you call a fireAllRules() before the first interaction with the kie-server, or that the first interaction always starts with a fireAllRules command.
Your configuration rule could look like this:
/**
Configures the session's listeners.
**/
rule "[SUB-CONFIG] Listeners Configuration"
salience 1000
when
then
((org.drools.impl.StatefulKnowledgeSessionImpl)kcontext.getKnowledgeRuntime()).addEventListener(new MyWorkingMemoryEventListener());
((org.drools.impl.StatefulKnowledgeSessionImpl)kcontext.getKnowledgeRuntime()).addEventListener(new MyAgendaEventListener());
end
You can place this rule in your kie-server project.
Hope it helps,
I am interested in creating rules for Drools Planner. I want that a user can create his own rules in a java app before starting the Drools Planner. Maybe a Drools-rule-file could be generated after the user has added his rules. Would this be possible or do I have to create the rule-file while developing the whole java application?
Many thanks...
Yes it's possible.
The trick is to build your own RuleBase and set it in the Planner config.
See section "5.3.4.2.2. A RuleBase (possibly defined by Guvnor)" in the manual.
You can construct a RuleBase by several, depending on how you want your user to edit his/her rules:
From a DRL file. This presumes the user knows DRL. See Drools Expert manual.
From a DSL file. This allows you to use natural language.
From the guvnor webapp. This allows you to use the tooling Guvnor, such as a guided rule editor, a decision tables spreadsheet, ... You can even use a changeset.
From guvnor in eclipse or a standalone app (under development and experimental). There's some work being in this area, but it's still young.
I find no documentation on how to update objects vaadin asynchronously. Can anyone help me? What I need is to render a table and then update the values of a column with a call rather slow, and so I want to make it asynchronous ..
This has been discussed a lot on this thread on the Vaadin forum. You might want to read it, it contains a lot of useful information.
Just do the updates in another thread. UI modifications from background threads must be synchronized to application object. Add icepush, refresher or proggresbar to get changes from server to client.
As far as I know Vaadin provides two add-ons for solving this problem: ServerPush and DontPush. Both add-ons can be imported via maven and both support WebSockets as well as fallback solutions for browsers without WebSocket support. Although ServerPush provides seemingly more features than DontPush, it is lower rated than DontPush, probably because it is more complicated.
For pushing updates to the client DontPush provides a very simple solution that does not require any changes to the web application. Only the servlet-class in web.xml needs to be replaced by org.vaadin.dontpush.server.impl.jetty.DontPushServlet and the widget set has to be updated afterwards via mvn vaadin:update-widgetset. That's all. Any changes on the server will be automatically pushed to the client. I successfully tested this add-on with Chrome 14. Unfortunately, I could not get it working with Firefox 7.
According to the web page of ServerPush the ServerPush add-on should provide this functionality, too. However, I could not figure out how to setup ServerPush to be working with jetty. Moreover, it seems to be more complicated in use. It requires several changes to the web.xml as well as additional configuration files for the atmosphere server.
In contrast to DontPush ServerPush provides also an explicit pushing mechanism which allows to update the GUI manually by calling the push() method of a certain pusher component which needs to be added to the main window beforehand. However, I also failed to get this working.
Hey guys. We're using OSGi services in an Eclipse RCP application. To track them, we're using the org.osgi.util.tracker.ServiceTracker class. A sample code from the application looks like
mailServiceTracker = new ServiceTracker(context, MailService.class.getName(), null);
mailServiceTracker.open();
MailService service = (MailService) mailServiceTracker.getService();
Now my problem is that the getService() method frequently returns null when I created a new service. The code works very well for services that are existing for a long time in the application, but each time I create a new service, I have to do many things until the service is finally found and tracked. I regularly try for example
'Clean...' in Eclipse
'Refresh' all projects in Eclipse
Rebuild the project on the command line
Sometimes those things help, and sometimes they don't. Does anyone have experiences with those trackers and can tell me how to avoid this behavior and how to get the services tracked immediately upon creation?
Thanks
The problem is that the services you want may not have been created yet (especially in an bundle activator, as some bundles may not yet have started). If you still want to use the service tracker, you will need to provide a ServiceTrackerCustomizer, and keep track (sorry, no pun intended) of the services as they come and go.
Or, you could just switch over to Declarative Services that handle this for you.
There is nothing wrong with using ServiceTrackers other than the fact that it's a fairly low-level way of tracking services. Whilst I agree that declarative services are a nice mechanism, simply dismissing ServiceTrackers because of "all sorts of issues" sounds like bad advice.
Back to the question.
As soon as a service tracker is created and opened, it gives you access to all services that match the filter condition you specified upon creation. There is no delay there. The only thing I can think about is that somehow your bundles are not correctly resolved, so services that are registered from a bundle A are simply not visible to a bundle B using a ServiceTracker. To check this, first locate the bundle that exports the package containing the service interface, and then make sure both A and B are actually wired to it.
Explaining the update/refresh mechanism in OSGi a bit more:
Whenever you update something in OSGi, it's a two step process.
Let's assume you update a bundle that contains a new version of an exported package. Let's also assume there is some consumer that imports it. As long as you only update the bundle but not explicitly refresh the wiring (of which import links to which export) the consumer will still be wired to the old version of the package. As soon as you do a package refresh (something you can do in OSGi via the PackageAdmin service) your consumer will be resolved again and will be wired to the new version.
The reason this is decoupled is that you might want to do updates of several bundles and not "refresh" after each one but instead defer such a refresh until all of them are updated.
It's quite possible that this is the effect you're seeing. Initially you only do an update, and only after the refresh will the tracker actually see the new version of the service.
Not being flippant at all, don't use service trackers. They appear to make your life simple, but there are all sort of issues with them. I'd recommend that you look into using Declarative Services instead. The support for DS in Eclipse has been very good from 3.5 onward.
You might want to check out this book and the associated presentations for more information on why using Service Trackers is a bad idea.
http://equinoxosgi.org/