[ 
https://issues.apache.org/jira/browse/FINERACT-1203?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17214187#comment-17214187
 ] 

Petri Tuomola commented on FINERACT-1203:
-----------------------------------------

Hi [~vorburger] - this is the issue I mentioned when I was getting Spring Boot 
to work at all in May, from my original email: 

"bootWar creates a WAR that can be deployed to an external Tomcat and works as 
expected. However the WAR file can’t be started directly from command line, as 
OpenJPA doesn’t seem to be able to find persistence-enhanced classes from 
within a WAR (looking at the source, it seems to be hardcoded to only handle 
classes from filesystem or JAR)."

So assuming my analysis back then was correct, some clever magic would be 
required to get this working - I suppose an option would be to raise a PR 
against OpenJPA, or alternatively complete the migration to EclipseLink...

> Fix broken Executable WAR (and unify & ditch Executable JAR?)
> -------------------------------------------------------------
>
>                 Key: FINERACT-1203
>                 URL: https://issues.apache.org/jira/browse/FINERACT-1203
>             Project: Apache Fineract
>          Issue Type: Improvement
>            Reporter: Michael Vorburger
>            Priority: Major
>         Attachments: war.log
>
>
> I've noticed that the WAR we build is not a traditional Tomcat only WAR, but 
> a Spring Boot Executable WAR - but a broken one! :( Whereas our Executable 
> JAR works of course, our WAR with {{java -jar 
> fineract-provider/build/libs/fineract-provider.war}} fails to start, see 
> attached {{war.log}}.
> [~ptuomola] and [~aleks] perhaps this is something one of you would like to 
> look into?
> If you can fix this, then it poses the question why we would build and 
> document and distribute and support two different artifacts. My vote, if this 
> can be fixed, could be to unify on only having a (both "traditional AND 
> Executable") WAR and no JAR. Unless there are reasons why that may be stupid, 
> and both are useful, even if the problem above was fixed. Thoughts?



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to