Internal Server Error when run on TorqueBox

[jasonjackson]

trafficstars

I have tried different versions of gems, as well as pulling from git. All configurations seem to result in the same error:

18:14:50,461 SEVERE com.sun.jersey.server.impl.application.WebApplicationImpl The RuntimeException could not be mapped to a response, re-throwing to the HTTP container: java.lang.NullPointerException

Full log message: http://pastie.org/pastes/4829082/text

Gemfile: http://pastie.org/pastes/4829091/text

Neo4j itself seems to work fine. The application does actually start, I can sign-up and login without issue. Torquebox seems to be working as expected, the JBoss console and everything looks OK. As far as I can tell, the issue is scoped to neo4j-admin.

Thoughts?

[andreasronge]

Hmm, I don't know. Have not seen this problem before. I guess there is a conflict in JARs used by TorqueBox and Neo4j or JRuby. Btw, did you test with the new 0.2.0 version of neo4j-admin I just released today ? There was some problem with JRuby 1.7.X and the Cypher Console using the neo4j-admin console

[jasonjackson]

I just tested the latest gem, same result. I did have this working with webrick using the same gem and jruby versions. Neo4j gems and jruby 1.6.7 are happy with torquebox, issue is scoped only to the neo4j-admin gem. Any idea what jar is bottled up with the admin gem that could be conflicting, or a direction to get this to play nice?

[andreasronge]

Sorry but I have no clue. I guess you have to do a binary search and try to remove things and see when it starts working again. For example try without JRuby and start the webadmin in JBoss. I know - not that fun ... The other alternative is to start another process and use neo4j enterprise HA cluster. I've got some utilities for doing that. See the HA Cluster section at https://github.com/andreasronge/neo4j/wiki/Neo4j%3A%3ARails-Config I also know that the neo4j team are trying to simplify the HA setup by avoiding zookeeper