[
https://issues.apache.org/jira/browse/JCLOUDS-171?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13704406#comment-13704406
]
ASF subversion and git services commented on JCLOUDS-171:
---------------------------------------------------------
Commit 7eb9fb0372c161ebffb0f3671b0c817c6da59c9d in branch refs/heads/master
from [~nacx]
[
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-labs.git;h=7eb9fb0 ]
JCLOUDS-171: Fixed NPE when destroying nodes in Abiquo
There was a race condition when destroying a node in Abiquo
using the ComputeService: jclouds performed the active wait
until the node was in state TERMINATED, and used the transformation
functions to convert an Abiquo VirtualMachine to a NodeMetadata. If the
node was effectively deleted during this transformation, a NPE
was raised. This is because there are a couple of API calls that are
executed in that transformation to retrieve all information required
to build the NodeMetadata.
To prevent this, the AbiquoComputeServiceAdapter will block until the
node has been actually deleted instead of delegating that to jclouds.
This way there is no need to run the transformation functions that had
the race condition.
> NPE in call to destroyNode for Abiquo
> -------------------------------------
>
> Key: JCLOUDS-171
> URL: https://issues.apache.org/jira/browse/JCLOUDS-171
> Project: jclouds
> Issue Type: Bug
> Components: jclouds-labs
> Affects Versions: 1.6.0
> Environment: Abiquo "London" environment
> Reporter: Alex Heneveld
> Assignee: Ignasi Barrera
> Labels: abiquo
> Fix For: 1.7.0, 1.6.2
>
> Attachments: JCLOUDS-171.patch
>
>
> caused by a race condition, as suggested in the stack trace below
> Caused by: java.lang.NullPointerException
> at org.jclouds.http.functions.ParseXMLWithJAXB.apply(ParseXMLWithJAXB.java:69)
> at org.jclouds.abiquo.domain.DomainWrapper.refresh(DomainWrapper.java:101)
> at
> org.jclouds.abiquo.strategy.cloud.ListAttachedNics.execute(ListAttachedNics.java:67)
> at
> org.jclouds.abiquo.domain.cloud.VirtualMachine.listAttachedNics(VirtualMachine.java:316)
> at
> org.jclouds.abiquo.compute.functions.VirtualMachineToNodeMetadata.apply(VirtualMachineToNodeMetadata.java:120)
> at
> org.jclouds.abiquo.compute.functions.VirtualMachineToNodeMetadata.apply(VirtualMachineToNodeMetadata.java:57)
> at
> com.google.common.base.Functions$FunctionComposition.apply(Functions.java:210)
> at
> org.jclouds.compute.strategy.impl.AdaptingComputeServiceStrategies.getNode(AdaptingComputeServiceStrategies.java:142)
> at
> org.jclouds.compute.predicates.AtomicNodeTerminated.refreshOrNull(AtomicNodeTerminated.java:49)
> at
> org.jclouds.compute.predicates.AtomicNodeTerminated.refreshOrNull(AtomicNodeTerminated.java:35)
> at
> org.jclouds.compute.predicates.internal.TrueIfNullOrDeletedRefreshAndDoubleCheckOnFalse.apply(TrueIfNullOrDeletedRefreshAndDoubleCheckOnFalse.java:53)
> at
> org.jclouds.compute.predicates.internal.TrueIfNullOrDeletedRefreshAndDoubleCheckOnFalse.apply(TrueIfNullOrDeletedRefreshAndDoubleCheckOnFalse.java:38)
> at org.jclouds.util.Predicates2$RetryablePredicate.apply(Predicates2.java:113)
> at
> org.jclouds.compute.internal.BaseComputeService.doDestroyNode(BaseComputeService.java:306)
> at
> org.jclouds.compute.internal.BaseComputeService.destroyNode(BaseComputeService.java:247)
> at brooklyn.location.jclouds.JcloudsLocation.release(JcloudsLocation.java:817)
> at brooklyn.location.jclouds.JcloudsLocation.release(JcloudsLocation.java:1)
> at
> brooklyn.entity.basic.SoftwareProcessImpl.stopInLocation(SoftwareProcessImpl.java:505)
> at
> brooklyn.entity.basic.SoftwareProcessImpl.stop(SoftwareProcessImpl.java:469)
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
