[ 
https://issues.apache.org/jira/browse/KARAF-5962?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16717339#comment-16717339
 ] 

Jean-Baptiste Onofré commented on KARAF-5962:
---------------------------------------------

I'm testing a fix on Felix Utils.

> Regression when installing some features containing fragments 4.2.0 -> 4.2.1
> ----------------------------------------------------------------------------
>
>                 Key: KARAF-5962
>                 URL: https://issues.apache.org/jira/browse/KARAF-5962
>             Project: Karaf
>          Issue Type: Bug
>          Components: karaf
>    Affects Versions: 4.2.1
>         Environment: java.specification.name Java Platform API Specification
> java.specification.vendor Oracle Corporation
> java.specification.version 1.8
> java.runtime.name OpenJDK Runtime Environment
> java.runtime.version 1.8.0_181-8u181-b13-1~deb9u1-b13
>            Reporter: rg chi
>            Assignee: Jean-Baptiste Onofré
>            Priority: Major
>             Fix For: 4.2.2
>
>
> Bug reproduction uses Apache Ignite karaf feature with JDK 8. In this case 
> org.apache.ignite.ignite-core plays host to fragment 
> org.apache.ignite.ignite-indexing.
> 1) Expose
> {noformat}
> com.sun.management, \
> sun.misc, \
> sun.nio.ch
> {noformat}
> in system bundle
> (for some reason ignite uses Import-Package to wire these so boot delegation 
> is not enough; we need to append packages to system bundle)
> 2)
> {noformat}
> karaf@root()> feature:repo-add 
> mvn:org.apache.ignite/ignite-osgi-karaf/2.6.0/xml/features
> karaf@root()> feature:install ignite-core
> karaf@root()> feature:install ignite-indexing
> {noformat}
> Works fine in 4.2.0 but in 4.2.1 this generates exception:
> {noformat}
> java.lang.ArrayIndexOutOfBoundsException: -1
>       at java.util.ArrayList.elementData(ArrayList.java:422)
>       at java.util.ArrayList.set(ArrayList.java:450)
>       at org.apache.felix.resolver.util.ShadowList.replace(ShadowList.java:74)
>       at org.apache.felix.resolver.Candidates.prepare(Candidates.java:941)
>       at 
> org.apache.felix.resolver.ResolverImpl.getInitialCandidates(ResolverImpl.java:505)
>       at 
> org.apache.felix.resolver.ResolverImpl.doResolve(ResolverImpl.java:390)
>       at org.apache.felix.resolver.ResolverImpl.resolve(ResolverImpl.java:378)
>       at org.apache.felix.resolver.ResolverImpl.resolve(ResolverImpl.java:332)
>       at 
> org.apache.karaf.features.internal.region.SubsystemResolver.resolve(SubsystemResolver.java:257)
>       at 
> org.apache.karaf.features.internal.service.Deployer.deploy(Deployer.java:388)
>       at 
> org.apache.karaf.features.internal.service.FeaturesServiceImpl.doProvision(FeaturesServiceImpl.java:1025)
>       at 
> org.apache.karaf.features.internal.service.FeaturesServiceImpl.lambda$doProvisionInThread$13(FeaturesServiceImpl.java:964)
>       at java.util.concurrent.FutureTask.run(FutureTask.java:266)
>       at 
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
>       at 
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
>       at java.lang.Thread.run(Thread.java:748)
> Error executing command: -1
> {noformat}
> Now this does not seem to happen with all fragment-host combinations. But the 
> above example is a reliable way to reproduce the faulty behavior in Karaf 
> 4.2.1. 
> This might be related to changes introduced in KARAF-5706 but as of yet 
> undetermined.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to