Some Errors

[diego delavega]

cumulus 1.1.0
cassandra 2.0.7  singleNode, 1 replication factor ..

19:02:42,265 INFO  <CIRRUS-00005> : Using Triple Store layout.
19:02:43,802 INFO  <CIRRUS-00021> : SELECT answer:
19:07:59,347 ERROR <CIRRUS-00026> : Not-well specified system internal failure has been detected. See below for further details.
com.datastax.driver.core.exceptions.NoHostAvailableException: All host(s) tried for query failed (tried: localhost/127.0.0.1 (com.datastax.driver.core.exceptions.DriverException: Timeout during read))
    at com.datastax.driver.core.exceptions.NoHostAvailableException.copy(NoHostAvailableException.java:64)
    at com.datastax.driver.core.DefaultResultSetFuture.extractCauseFromExecutionException(DefaultResultSetFuture.java:256)
    at com.datastax.driver.core.ArrayBackedResultSet.fetchMoreResultsBlocking(ArrayBackedResultSet.java:197)
    at com.datastax.driver.core.ArrayBackedResultSet.isExhausted(ArrayBackedResultSet.java:135)
    at com.datastax.driver.core.ArrayBackedResultSet$1.hasNext(ArrayBackedResultSet.java:170)
    at edu.kit.aifb.cumulus.datasource.impl.AbstractResultIterator.hasNext(AbstractResultIterator.java:44)
    at edu.kit.aifb.cumulus.store.sesame.CumulusRDFIterator.getNextElement(CumulusRDFIterator.java:33)
    at edu.kit.aifb.cumulus.store.sesame.CumulusRDFIterator.getNextElement(CumulusRDFIterator.java:15)
    at info.aduna.iteration.LookAheadIteration.lookAhead(LookAheadIteration.java:91)
    at info.aduna.iteration.LookAheadIteration.hasNext(LookAheadIteration.java:60)
    at info.aduna.iteration.IterationWrapper.hasNext(IterationWrapper.java:68)
    at info.aduna.iteration.FilterIteration.findNextElement(FilterIteration.java:79)
    at info.aduna.iteration.FilterIteration.hasNext(FilterIteration.java:54)
    at info.aduna.iteration.ConvertingIteration.hasNext(ConvertingIteration.java:73)
    at org.openrdf.query.algebra.evaluation.iterator.JoinIterator.getNextElement(JoinIterator.java:69)
    at org.openrdf.query.algebra.evaluation.iterator.JoinIterator.getNextElement(JoinIterator.java:30)
    at info.aduna.iteration.LookAheadIteration.lookAhead(LookAheadIteration.java:91)
    at info.aduna.iteration.LookAheadIteration.hasNext(LookAheadIteration.java:60)
    at info.aduna.iteration.ConvertingIteration.hasNext(ConvertingIteration.java:73)
    at info.aduna.iteration.IterationWrapper.hasNext(IterationWrapper.java:68)
    at info.aduna.iteration.FilterIteration.findNextElement(FilterIteration.java:79)
    at info.aduna.iteration.FilterIteration.hasNext(FilterIteration.java:54)
    at info.aduna.iteration.IterationWrapper.hasNext(IterationWrapper.java:68)
    at org.openrdf.sail.helpers.SailBaseIteration.hasNext(SailBaseIteration.java:61)
    at info.aduna.iteration.IterationWrapper.hasNext(IterationWrapper.java:68)
    at edu.kit.aifb.cumulus.cli.Query.doExecute(Query.java:72)
    at edu.kit.aifb.cumulus.cli.Command.execute(Command.java:57)
    at edu.kit.aifb.cumulus.cli.Cirrus.main(Cirrus.java:57)
Caused by: com.datastax.driver.core.exceptions.NoHostAvailableException: All host(s) tried for query failed (tried: localhost/127.0.0.1 (com.datastax.driver.core.exceptions.DriverException: Timeout during read))
    at com.datastax.driver.core.RequestHandler.sendRequest(RequestHandler.java:100)
    at com.datastax.driver.core.RequestHandler$1.run(RequestHandler.java:171)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
    at java.lang.Thread.run(Thread.java:722)


any ideas ??

Thank you!!

[Andrea Gazzarini]

Hi Diego,
I was working at this issue just a moment ago. Let me guess: you're running CumulusRDF 1.1.0 with Cassandra 2.x.

If so, instructions on the wiki (Getting Started) are not valid for this scenario...they refer to CumulusRDF 1.0.1 (the current version that you can download)

To be more precise: you can run the mentioned mvn command *only* if you are using Cassandra 1.2.x. This because maven cassandra plugin 2.0 starts an embedded cassandra which is not detected by datastax driver (the cassandra client we are using in 1.1.0).

So, briefly I removed the cassandra maven plugin from the CumulusRDF maven profile associated with Cassandra 2.x. This is what you can do:

If you checked out the 1.1.0 before today (I committed the change above about two hours ago)

• if you want to start CumulusRDF 1.1.0 + Cassandra 1.2.x follow the instructions on the Wiki (Getting started); in this specific case they are valid also for 1.1.0. So
  

mvn clean cassandra:stop cassandra:start tomcat7:run

• if you want to start CumulusRDF 1.1.0 + Cassandra 2.x you must have an external Cassandra 2.x running. After that replace the mvn command in the Getting Started page with
  

mvn clean tomcat7:run

if you checked out the 1.1.0 just few minutes ago:

• if you want to start CumulusRDF 1.1.0 + Cassandra 1.2.x type replace the mvn command in the Getting Started page with

mvn clean cassandra:stop cassandra:start tomcat7:run  -Pcassandra12x-hector-full-tp-index

• if you want to start CumulusRDF 1.1.0 + Cassandra 2.x you must have an external Cassandra 2.x running. After that replace the mvn command in the Getting Started page with

mvn clean tomcat7:run

We are going to release 1.1.0 very soon and after that, the Wiki pages will refer to that version.

Best,
Andrea

--
You received this message because you are subscribed to the Google Groups "cumulusrdf" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cumulusrdf-li...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[diego delavega]

Hi Andrea!

I am using cumulus 1.1.0 & cassandra 2.0.7 (external),
and i compiled it according to your instructions on the previous post "CulumusRDF benchmark".

I have to say that generally it works fine. I am getting some errors here and there but i suspect it must have something to do with it
some restrictions i might have like the java heap size or the disk space.

Practically i the error i posted above is from the second query of the sp2b in a dataset of 10mil triples.
The same query on 1mil triples was  complied successfully.

I hope this clarify things a little more.

thank you !

[Andrea Gazzarini]

Hi Diego,

mmm...interesting...that has something to do with internal threading & connection usage....a couple of questions: I see there are five minutes between

19:02:43,802 INFO  <CIRRUS-00021> : SELECT answer:

and

19:07:59,347 ERROR <CIRRUS-00026> : Not-well specified system internal failure has been detected. See below for further details.

Could you please:

- detail each step of your scenario?

- post also the other errors you have here and there?

Many thanks,
Andrea

Best,
Andrea

[Andreas Wagner]

Hi Diego,

On 06/11/2014 11:22 AM, diego delavega wrote:
> Practically i the error i posted above is from the second query of the
> sp2b in a dataset of 10mil triples.
> The same query on 1mil triples was complied successfully.

Does CumulusRDF give you any results for that query on SP2-10M dataset?
In other words, what happens if you run the query with a "LIMIT 1"
attached? Also, could you post the Cassandra logs for that query/time?

Thanks :)
Andreas

P.S.: Do you already have some benchmark results? I would be interested
in an early version of your benchmark results.

[diego delavega]

Sorry i haven't tried the "limit trick yet" and don't have all the errors that hapened.

P.S.: Do you already have some benchmark results? I would be interested
in an early version of your benchmark results.

this are the results what i have right now. 

CUMULUSrdf single node,  10 mil triples ( SP2B)

query

q1

q2

q3a

q3b

q3c

q4

q5a

q5b

q6

q7

q8

q9

q10

q11

q12a

q12b

q12c

time (s)

0.039919

TF

125.3417

118.6042

117.9805

EF

EF

EF

TF

TF

391.6772

625.8533

0.043272

EF

TRUE

TRUE

TRUE

results (#)

1

312446

2128

0

487

4

645

remarks

*TF : Time Fail >30'

*EF : Error Fail

ASK

ASK

ASK