ovb-fake-caserver job failing, timing out creating the Compute resource

[Juan Antonio Osorio Robles]

As of April 19, the job started failing with timeouts on the Compute resource, such as the following:

2017-04-20 09:31:12.140949 | 2017-04-20 09:31:08Z [overcloud.Compute.0]: CREATE_FAILED CREATE aborted
2017-04-20 09:31:12.141019 | 2017-04-20 09:31:08Z [overcloud.Compute]: UPDATE_FAILED Operation cancelled
2017-04-20 09:31:12.141039 |
2017-04-20 09:31:12.141069 | Stack overcloud CREATE_FAILED
2017-04-20 09:31:12.141117 |
2017-04-20 09:31:44.305580 | overcloud.Compute.0:
2017-04-20 09:31:44.305945 | resource_type: OS::TripleO::Compute
2017-04-20 09:31:44.306033 | physical_resource_id: 4bffbdbb-0dfe-4683-95a5-066e3eac5e89
2017-04-20 09:31:44.306116 | status: CREATE_FAILED
2017-04-20 09:31:44.306216 | status_reason: |
2017-04-20 09:31:44.306290 | CREATE aborted

It seems that the NetworkDeployment stalled. And there are no logs for the compute node (only for the controller).

[Juan Antonio Osorio Robles (juan-osorio-robles)]

This is an instance of a failed deployment: http://logs.openstack.org/26/458026/1/experimental-tripleo/gate-tripleo-ci-centos-7-ovb-fakeha-caserver/967ec0f/console.html

[Juan Antonio Osorio Robles (juan-osorio-robles)]

This was fixed by this commit https://review.openstack.org/#/c/458596/

The issue was that the net-iso configuration had changed on CI, when changing that, the original commit didn't fix it for the fakeha-caserver job.