[
https://issues.apache.org/jira/browse/WHIRR-431?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13149749#comment-13149749
]
Andrei Savu commented on WHIRR-431:
-----------------------------------
After a bit more investigation I think this should be working fine, the spot
instance requests are canceled on destroy[1]. Are you destroying the cluster by
calling ./whirr destroy-cluster or ClusterController.destroyCluster? Can you
provide us with the log files? What version of Whirr are you using?
We've used spot instance for testing in the past and I've never seen this
situation happening.
[1]
https://github.com/jclouds/jclouds/blob/master/providers/aws-ec2/src/main/java/org/jclouds/aws/ec2/compute/strategy/AWSEC2DestroyNodeStrategy.java
> Spot instance requests are not cancelled after cluster "destruction"
> --------------------------------------------------------------------
>
> Key: WHIRR-431
> URL: https://issues.apache.org/jira/browse/WHIRR-431
> Project: Whirr
> Issue Type: Bug
> Reporter: Marco Didonna
> Labels: ec2
>
> I've noticed that when using spot instances, first a request is created and,
> when fulfilled, the associated instances are launched. Standard operating
> procedure on EC2. The problem is when the cluster is destroyed the spot
> request are still there (in an open state) and new instances may be assigned
> to them. I've experienced this behavior three times before starting to
> manually delete spot request when the cluster is no longer needed.
> Anyone can confirm?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
