Open amarneris opened 8 years ago
Hello,
there is a fix for this made by Tim Lawrence, with a pull request here: https://github.com/cloudfoundry-attic/bosh_vcloud_cpi/pull/18
Is it possible to merge it in and close this off please?
Regards, Andreas
will take a look at the pr today.
Sent from my iPhone
On May 12, 2016, at 8:27 AM, amarneris notifications@github.com wrote:
Hello,
there is a fix for this made by Tim Lawrence, with a pull request here: cloudfoundry-attic/bosh_vcloud_cpi#18
Is it possible to merge it in and close this off please?
Regards, Andreas
— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub
Hello!
We recently had a failed bosh deployment on a vCloud director, which resulted in the vCloud Director returning failed tasks for days after the original deployment failure.
The following code snippet in the vcd_client.rb file (lines 165 to 175) has a check, which is not time-bound, so it checks through all failed tasks against each entity (VM/vApp).
# verify all tasks succeeded
unless entity.tasks.nil? || entity.tasks.empty?
unless failed_tasks.empty?
@logger.debug "Failed tasks: #{failed_tasks}"
unless accept_failure
failed_tasks_info = failed_tasks.map { |t| "Task #{t.urn} #{t.operation}, Details:#{t.details}
raise "Some tasks failed: #{failed_tasks_info.join('; ')}"
end
end
end
Is there a reason why this is not time bound? Would it be an acceptable pull request to add some logic that only checks tasks with an endTime of less than 24 hours?
The debug log snippet, with tasks being returned from previous failed deployments is below: