[
https://issues.apache.org/jira/browse/JCLOUDS-1331?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16153743#comment-16153743
]
Duncan Grant commented on JCLOUDS-1331:
---------------------------------------
[~andreaturli] I don't think this is the same issue but I'll be happy to retest.
I noticed this problem when I'd already modified the azure provider to work
round the issue above. I think the problem is more to do with server side
caching on the azure api server. Which means that we have a safety check (see
link above) that fails unnecessarily.
> Azure ARM orphaned resource groups are not cleaned up due to api
> caching/timing issue
> -------------------------------------------------------------------------------------
>
> Key: JCLOUDS-1331
> URL: https://issues.apache.org/jira/browse/JCLOUDS-1331
> Project: jclouds
> Issue Type: Bug
> Components: jclouds-compute
> Affects Versions: 2.0.2
> Reporter: Duncan Grant
> Priority: Minor
> Labels: azurecompute-arm
>
> When an azure arm node, and all of its resources, are deleted the resource
> group is not deleted. This is because there is a short period of time when
> the ARM api call here:
> https://github.com/duncangrant/jclouds-labs/blob/dbadb279f14848f21879f7eb6c7136e1a5f11192/azurecompute-arm/src/main/java/org/jclouds/azurecompute/arm/compute/strategy/CleanupResources.java#L207
> will respond with a list of resources even though they have been deleted.
> Presumably this could be fixed with a) a retry step or b) by deleting the
> resource group anyway if we know that we have already deleted all the
> resources in it.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
