Red Hat Bugzilla – Bug 1312190
heat stack-delete does not clean up CEPH disks
Last modified: 2016-10-18 08:08:40 EDT
Description of problem:
When deleting a deployment that uses individual CEPH storage devices, the devices do not get cleaned up This causes them to retain their CEPH characteristics which prevents them from being reused in future deployments
Version-Release number of selected component (if applicable):
heat packages in 2015.1.2-9.el7ost
Steps to Reproduce:
1.Deploy a cloud which uses CEPH not on the default system disk
2.use `heat stack-delete` to delete the deployment
3.Try to redeploy and observe that the CEPG storage is not used
No CEPH OSDs created
Fully deployed and functional CEPH storage
This bug did not make the OSP 8.0 release. It is being deferred to OSP 10.
Would you please have a look at this bug which was resolved in OSP10?
As per the above, the devices which retain their Ceph characteristics are identified and the deployment fails and indicates why they are unable to be reused in future deployments in the logs. From there the plan is to give the user a new option described in:
so that the user can have OSPd clean the disk during the deployment.
In a situation where one is practicing deployments and rebuilding the overcloud often, this new zap option could simply be left on. If this is your situation, then would this address the problem? If so, then I would update this BZ as a duplicate of BZ 1377867.
Until this option is added, our documentation covers how to use a preboot script  to do the zap.
It's been about a week. So I'm closing this as a duplicate as mentioned in my previous comment.
*** This bug has been marked as a duplicate of bug 1377867 ***