Description of problem:
The complete message in the UI:

ERROR: deployment failed with status: CREATE_FAILED and reason: Resource CREATE failed: ResourceInError: resources.Compute.resources[0].resources.NovaCompute: Went to status ERROR due to "Message: Unknown, Code: Unknown"

This happened in 26.0% of the RHELOSP deployment. Content synced successfully. This is my third re-deploy of RHELOSP, I have previously un-deployed the overcloud from the Director webUI, so it could be created again. I did not delete nodes, I'm using nodes discovered during the first RHELOSP deployment.

From the nova-compute.log:

Command: qemu-img convert -O raw /var/lib/ironic/master_images/tmpHrdQaf/ed45a705-e8cd-4b7f-84fd-62d50de7c1f8.part /var/lib/ironic/master_images/tmpHrdQaf/ed45a705-e8cd-4b7f-84fd-62d50de7c1f8.converted
Exit code: 1
Stdout: u''
Stderr: u'qemu-img: error while reading sector 715008: Input/output error\n'

and then:

NeutronClientException: 404 Not Found
The resource could not be found.

Please see attached production.log and nova-compute.log excerpts.

I tried to resume the failing tasks and got:

There was a conflict when trying to complete your request. The Stack (overcloud) already exists.

Version-Release number of selected component (if applicable):
RHCI-6.0-RHEL-7-20151201.t.0
RHCIOOO-7-RHEL-7-20151201.t.1

How reproducible:
Happened to me once

Steps to Reproduce:
1. Do a RHELOSP re-deployment with previously discovered nodes and previously undeployed Overcloud
2. Deployment fails at 26.0% with above error
3. 

Actual results:
Deployment fails; upon resuming the failing task, another error appears; unable to continue deployment

Expected results:
Deployment completes successfully

Additional info: