Tempest tests failing with some errors related to cinder backups

[Slawek Kaplonski]

In neutron jobs we are quite often facing issues related to cinder backups. Example of such issues are for example in:

http://logs.openstack.org/87/609587/11/check/tempest-multinode-full/2a5c5a1/testr_results.html.gz
http://logs.openstack.org/64/617364/19/check/tempest-slow/18519dc/testr_results.html.gz

[Slawek Kaplonski (slaweq)]

Another examples of issues with Cinder backup:

http://logs.openstack.org/87/609587/11/check/tempest-multinode-full/d4ec462/testr_results.html.gz

Some error with volume status:
http://logs.openstack.org/36/617736/10/gate/grenade-py3/9c55422/logs/grenade.sh.txt.gz#_2018-11-28_11_40_03_264

[Lisa Li (lisali)]

From http://logs.openstack.org/87/609587/11/check/tempest-multinode-full/d4ec462/testr_results.html.gz
It calls restore_volume, and checks status of backup to finish in 196s.

Details: backup a52fcd04-469a-4995-bba8-15bba439756a failed to reach available status (current restoring) within the required time (196 s).

Actual results: The restore tasks took about 4 minutes.
Nov 28 06:36:39.667634 ubuntu-xenial-ovh-bhs1-0000752678 cinder-backup[1888]: INFO cinder.backup.manager [None req-b6bfaf94-4142-40bb-a0f6-f6ddc850a921 tempest-VolumesBackupsTest-1683650084 None] Restore backup started, backup: a52fcd04-469a-4995-bba8-15bba439756a volume: 814cfd25-5d78-43ba-a987-af392587eb2b.
Nov 28 06:40:31.185456 ubuntu-xenial-ovh-bhs1-0000752678 cinder-backup[1888]: INFO cinder.backup.manager [None req-b6bfaf94-4142-40bb-a0f6-f6ddc850a921 tempest-VolumesBackupsTest-1683650084 None] Finished restoring backup a52fcd04-469a-4995-bba8-15bba439756a to volume 814cfd25-5d78-43ba-a987-af392587eb2b.

[Lisa Li (lisali)]

In summary, backup creation/restore took too much time and lead timeout. Currently, volume size is 1G, this is supported minimum capacity. So may we can update the timeout for backup creation/restore.