How to use blank node in Datalog rules for GenericRuleReasoner?

I'm using GenericRuleReasoner to infer virtual fact in my ontology. GenericRuleReasoner takes inference rules as Datalog rule in input as explain in jena inference support. here is an example of a DatalogRule attached to a generic reasoner :

String rules = "[r1: (?e1 st:runningTask st:gic_eth0) -> (?e1 rdf:type st:dataFromEthernet2IP)]";
Reasoner reasoner = new GenericRuleReasoner(Rule.parseRules(rules));
reasoner.setDerivationLogging(true);
InfModel inf = ModelFactory.createInfModel(reasoner, rawData);

When i tested this code on my data, it worked fine and infered 2000 virtual facts. However, when I changed the Datalog rule in order to create blank nodes like this

String rules = "[r1: (?e1 st:runningTask st:gic_eth0) -> (_:p rdf:type st:dataFromEthernet2IP)]";

I get only on virtual fact inferred.

Is there a problem with my blank node representation in datalog rule for GenericRuleReasoner ?

Is there a problem with my blank node representation in datalog rule for GenericRuleReasoner ?

Yes. You don't use blank nodes like this in Jena rules. The document that you linked to includes a grammar for rules, and there's nothing in it that would permit something like _:p as a node. The syntax for nodes in Jena rules is:

node      :=   uri-ref               // e.g. http://foo.com/eg
          or   prefix:localname      // e.g. rdf:type
          or   <uri-ref>             // e.g. <myscheme:myuri>
          or   ?varname              // variable
          or   'a literal'           // a plain string literal
          or   'lex'^^typeURI        // a typed literal, xsd:* type names support

If you want to create a new blank node, use the makeTemp(?x) bulitin in the body of the rule to bind ?x to a new blank node. E.g.,

[r1: (?e1 st:runningTask st:gic_eth0), makeTemp(?p)
     ->
     (?p rdf:type st:dataFromEthernet2IP) ]