Comment 7 for bug 1327218

In one case, we're attaching the encrypted luks volume to the instance here:

http://logs.openstack.org/93/156693/7/check/check-tempest-dsvm-postgres-full/d3b26e8/logs/screen-n-cpu.txt.gz#_2015-03-12_16_38_09_061

We initialize the connection and get the connection_info back here:

http://logs.openstack.org/93/156693/7/check/check-tempest-dsvm-postgres-full/d3b26e8/logs/screen-n-cpu.txt.gz#_2015-03-12_16_38_11_064

I see an os-attach call here:

http://logs.openstack.org/93/156693/7/check/check-tempest-dsvm-postgres-full/d3b26e8/logs/screen-n-cpu.txt.gz#_2015-03-12_16_38_15_223

We start detaching the volume here:

http://logs.openstack.org/93/156693/7/check/check-tempest-dsvm-postgres-full/d3b26e8/logs/screen-n-cpu.txt.gz#_2015-03-12_16_38_16_902

We're failing to detach the volume here:

http://logs.openstack.org/93/156693/7/check/check-tempest-dsvm-postgres-full/d3b26e8/logs/screen-n-cpu.txt.gz#_2015-03-12_16_38_17_567

And six minutes later we're terminating the bdm for that volume here:

http://logs.openstack.org/93/156693/7/check/check-tempest-dsvm-postgres-full/d3b26e8/logs/screen-n-cpu.txt.gz#_2015-03-12_16_44_54_876

After failing to detach, I'm also seeing the same volume_id showing up in the logs in other test runs:

VolumesV1SnapshotTestJSON:

http://logs.openstack.org/93/156693/7/check/check-tempest-dsvm-postgres-full/d3b26e8/logs/screen-n-cpu.txt.gz#_2015-03-12_16_42_03_507

TestMinimumBasicScenario:

http://logs.openstack.org/93/156693/7/check/check-tempest-dsvm-postgres-full/d3b26e8/logs/screen-n-cpu.txt.gz#_2015-03-12_16_44_40_119