Scheduling for investigation in 2.2.4.

I assume the 'long runnig FILTER operation' is a sub-query?

Particular cases:

PREFIX foo: <http://foo.bar/someOntology#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

SELECT ?node1 ?node2 ?nid1 WHERE {
 ?circuit1 a foo:Circuit .
 ?circuit2 a foo:Circuit .
 FILTER ( ?circuit1 != ?circuit2) .

 ?circuit1 foo:circuitId ?cid1 .
 ?circuit2 foo:circuitId ?cid2 .
 FILTER ( ?cid1 = ?cid2 )
}

PREFIX foo: <http://foo.bar/someOntology#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

SELECT ?circuit1 ?circuit2 ?cid1 WHERE {
 ?circuit1 a foo:Circuit .
 ?circuit2 a foo:Circuit .
 FILTER ( ?circuit1 != ?circuit2) .

 ?circuit1 foo:circuitId ?cid1 .
 ?circuit2 foo:circuitId ?cid2 .
 FILTER ( ?cid1 = ?cid2 )
}

PREFIX foo: <http://foo.bar/someOntology#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

SELECT ?interface1 ?interface2 ?ifnr ?nodeId WHERE {
 ?interface1 a foo:Interface .
 ?interface2 a foo:Interface .
 FILTER ( ?interface1 != ?interface2) .

 ?interface1 foo:interfaceNumber ?ifnr .
 ?interface2 foo:interfaceNumber ?ifnr .

 ?node1 foo:hasInterface ?interface1 .
 ?node2 foo:hasInterface ?interface2 .
 
 ?node1 foo:nodeId ?nodeId .
 ?node2 foo:nodeId ?nodeId

}


What is hopefully the relevant piece of stack trace from the first query:

 at java.util.HashMap.size(HashMap.java:268)
        at java.util.HashMap.putAll(HashMap.java:498)
        at org.openrdf.query.algebra.evaluation.QueryBindingSet.addAll(QueryBindingSet.java:61)
        at org.openrdf.query.algebra.evaluation.QueryBindingSet.<init>(QueryBindingSet.java:52)
        at org.openrdf.query.algebra.evaluation.impl.EvaluationStrategyImpl$3.convert(EvaluationStrategyImpl.java:299)
        at org.openrdf.query.algebra.evaluation.impl.EvaluationStrategyImpl$3.convert(EvaluationStrategyImpl.java:295)
        at info.aduna.iteration.ConvertingIteration.next(ConvertingIteration.java:77)
        at org.openrdf.query.algebra.evaluation.iterator.JoinIterator.getNextElement(JoinIterator.java:60)
        at org.openrdf.query.algebra.evaluation.iterator.JoinIterator.getNextElement(JoinIterator.java:19)
        at info.aduna.iteration.LookAheadIteration.lookAhead(LookAheadIteration.java:80)
        at info.aduna.iteration.LookAheadIteration.hasNext(LookAheadIteration.java:49)
        at info.aduna.iteration.IterationWrapper.hasNext(IterationWrapper.java:57)
        at info.aduna.iteration.FilterIteration.findNextElement(FilterIteration.java:68)
        at info.aduna.iteration.FilterIteration.hasNext(FilterIteration.java:43)
        at info.aduna.iteration.IterationWrapper.hasNext(IterationWrapper.java:57)
        at info.aduna.iteration.FilterIteration.findNextElement(FilterIteration.java:68)
        at info.aduna.iteration.FilterIteration.hasNext(FilterIteration.java:43)
        at org.openrdf.query.algebra.evaluation.iterator.JoinIterator.getNextElement(JoinIterator.java:58)
        at org.openrdf.query.algebra.evaluation.iterator.JoinIterator.getNextElement(JoinIterator.java:19)
        at info.aduna.iteration.LookAheadIteration.lookAhead(LookAheadIteration.java:80)
        at info.aduna.iteration.LookAheadIteration.hasNext(LookAheadIteration.java:49)
        at info.aduna.iteration.ConvertingIteration.hasNext(ConvertingIteration.java:62)
        at info.aduna.iteration.IterationWrapper.hasNext(IterationWrapper.java:57)
        at info.aduna.iteration.LockingIteration.hasNext(LockingIteration.java:61)
        - locked <0x00007ff5a9ba83e8> (a info.aduna.iteration.LockingIteration)
        at info.aduna.iteration.IterationWrapper.hasNext(IterationWrapper.java:57)
        at org.openrdf.sail.helpers.SailBaseIteration.hasNext(SailBaseIteration.java:50)
        at info.aduna.iteration.IterationWrapper.hasNext(IterationWrapper.java:57)
        at info.aduna.iteration.TimeLimitIteration.hasNext(TimeLimitIteration.java:57)
        - locked <0x00007ff5a9bb81e0> (a org.openrdf.repository.sail.SailQuery$QueryInterruptIteration)
        at org.openrdf.query.impl.TupleQueryResultImpl.hasNext(TupleQueryResultImpl.java:90)

The interrupt thread that goes with that:

"TimeLimitIteration" daemon prio=10 tid=0x00007ff5047cdc00 nid=0x288d waiting for monitor entry [0x0000000042feb000..0x0000000042febc20]
   java.lang.Thread.State: BLOCKED (on object monitor)
        at info.aduna.iteration.TimeLimitIteration.interrupt(TimeLimitIteration.java:103)
        - waiting to lock <0x00007ff5a9bb81e0> (a org.openrdf.repository.sail.SailQuery$QueryInterruptIteration)
        at info.aduna.iteration.TimeLimitIteration.access$000(TimeLimitIteration.java:19)
        at info.aduna.iteration.TimeLimitIteration$1.run(TimeLimitIteration.java:45)
        at java.util.TimerThread.mainLoop(Timer.java:512)
        at java.util.TimerThread.run(Timer.java:462)

TimeLimitIterator, which is responsible for stopping the query evaluation, used a bad synchronization scheme causing the time-out to block until the evaluation returned from referenced iteration.