Open hickst opened 7 years ago
paper: PMC1977042
error:
java.lang.NullPointerException
stack trace:
org.clulab.reach.context.RuleBasedContextEngine.query(RuleBasedEngine.scala:126)
org.clulab.reach.context.RuleBasedContextEngine$$anonfun$assign$1.apply(RuleBasedEngine.scala:102)
org.clulab.reach.context.RuleBasedContextEngine$$anonfun$assign$1.apply(RuleBasedEngine.scala:93)
scala.collection.Iterator$class.foreach(Iterator.scala:893)
scala.collection.AbstractIterator.foreach(Iterator.scala:1336)
scala.collection.IterableLike$class.foreach(IterableLike.scala:72)
scala.collection.AbstractIterable.foreach(Iterable.scala:54)
org.clulab.reach.context.RuleBasedContextEngine.assign(RuleBasedEngine.scala:92)
org.clulab.reach.ReachSystem.extractFrom(ReachSystem.scala:83)
org.clulab.reach.ReachSystem.extractFrom(ReachSystem.scala:146)
org.clulab.reach.ReachSystem.extractFrom(ReachSystem.scala:140)
org.clulab.reach.ReachSystem.extractFrom(ReachSystem.scala:66)
org.clulab.reach.PaperReader$.getMentionsFromEntry(PaperReader.scala:133)
org.clulab.reach.ReachCLI.processPaper(ReachCLI.scala:104)
org.clulab.reach.ReachCLI$$anonfun$2.apply(ReachCLI.scala:53)
org.clulab.reach.ReachCLI$$anonfun$2.apply(ReachCLI.scala:49)
This error looks like the previous one, but in a different paper:
paper: PMC1567125
error:
java.lang.NullPointerException
stack trace:
org.clulab.reach.context.RuleBasedContextEngine.query(RuleBasedEngine.scala:126)
org.clulab.reach.context.RuleBasedContextEngine$$anonfun$assign$1.apply(RuleBasedEngine.scala:102)
org.clulab.reach.context.RuleBasedContextEngine$$anonfun$assign$1.apply(RuleBasedEngine.scala:93)
scala.collection.Iterator$class.foreach(Iterator.scala:893)
scala.collection.AbstractIterator.foreach(Iterator.scala:1336)
scala.collection.IterableLike$class.foreach(IterableLike.scala:72)
scala.collection.AbstractIterable.foreach(Iterable.scala:54)
org.clulab.reach.context.RuleBasedContextEngine.assign(RuleBasedEngine.scala:92)
org.clulab.reach.ReachSystem.extractFrom(ReachSystem.scala:83)
org.clulab.reach.ReachSystem.extractFrom(ReachSystem.scala:146)
org.clulab.reach.ReachSystem.extractFrom(ReachSystem.scala:140)
org.clulab.reach.ReachSystem.extractFrom(ReachSystem.scala:66)
org.clulab.reach.PaperReader$.getMentionsFromEntry(PaperReader.scala:133)
org.clulab.reach.ReachCLI.processPaper(ReachCLI.scala:104)
org.clulab.reach.ReachCLI$$anonfun$2.apply(ReachCLI.scala:53)
org.clulab.reach.ReachCLI$$anonfun$2.apply(ReachCLI.scala:49)
Update: just looked at the 7 errors so far in processing: 6 are the RulesBasedEngine NullPointer error and 1 is the tail of empty list error (example of both shown above).
The NullPointer error is fixed now
@hickst, can provide the commit hash you used for this run? I'm trying to track the first error (tail of empty list).
Thanks!
The fix is in branch Issue375_context_engine
The commit hash is 6785d0971fb2eebae2bbd03fe01d9d1f1c1df2d8
@enoriega, I'm guessing that is the hash for your fix, right? Can you put in a PR for the context engine fix, or is there still work that needs to be done?
A somewhat disturbing error occurred right at the very end of one of the BigRun batches. Disturbing because this looks like the kind of error one might get chasing a cycle in a data structure (Gus just questioned this possibility at our last weekly meeting):
[error] (run-main-0) java.lang.StackOverflowError
java.lang.StackOverflowError
at edu.stanford.nlp.graph.DirectedMultiGraph$EdgeIterator.primeIterator(DirectedMultiGraph.java:542)
at edu.stanford.nlp.graph.DirectedMultiGraph$EdgeIterator.primeIterator(DirectedMultiGraph.java:542)
at edu.stanford.nlp.graph.DirectedMultiGraph$EdgeIterator.primeIterator(DirectedMultiGraph.java:542)
at edu.stanford.nlp.graph.DirectedMultiGraph$EdgeIterator.primeIterator(DirectedMultiGraph.java:542)
.....repeat this error 1024 times.... then the JVM died, leaving 4 files unprocessed from that batch.
By comparing start messages to finish messages in the log file, I believe the culprit is PMC4370879
but since a total of five files were left unfinished (out of 60685) it could also be one of PMC4014127 PMC4045150 PMC4230053 PMC4372909
Hitting a few internal errors during the big run. All referenced papers are available on River at:
/net/kate/storage/data/nlp/corpora/pmc_openaccess/pmc_aug2016_explorer/nxml