Description of problem: As a user i want to easily restore a removed backuped by cinder backup to a different AZ than the original volume. Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: Currently possible with the cinder create command $ BACKUP_ID=$(openstack volume backup list --name vol-dcn1 -f value -c ID) & OS_VOLUME_API_VERSION=3.51 cinder create --availability-zone az-dcn2 --name vol-dcn2 --backup-id $BACKUP_ID 1 Goal is to integrate this with cinder backup tooling or API.
The patch merged on master, and is on target for OSP-17.
not a GA must have, moving to 17.0.2
Doc updates for this BZ will be limited to updating the command to restore a backup (step 2) in [1]. [1] https://access.redhat.com/documentation/en-us/red_hat_openstack_platform/17.1-beta/html-single/deploying_a_distributed_compute_node_dcn_architecture/index#proc_backup-and-restore-across-edge-sites_appendix However, it turns out cross-az backups themselves are broken in OSP-17 (bug #2229750), which makes it difficult to verify the new command for restoring a cross-az backup. I'm clearing the needinfo now, and will revisit the doc issue when the blocking bug is cleared.
According to our records, this should be resolved by python-cinderclient-7.4.1-1.20220727131518.4f72e6f.el9ost. This build is available now.
Tested on python-cinderclient-7.4.1-0.20210904062152.4f72e6f # Get availability zones (central) [stack@site-undercloud-0 ~]$ cinder availability-zone-list +------------+-----------+ | Name | Status | +------------+-----------+ | az-central | available | | az-dcn1 | available | | az-dcn2 | available | +------------+-----------+ # Check where cinder-backup service is running (central) [stack@site-undercloud-0 ~]$ openstack volume service list cinder-backup service is running on "az-central" # Create a volume on availability zones "az-dcn1" (central) [stack@site-undercloud-0 ~]$ cinder --os-volume-api-version 3.51 create --name test_volume --availability-zone az-dcn1 1 # Create a backup of that volume on the availability zone "az-central" where the cinder-backup service is running $ cinder --os-volume-api-version 3.51 backup-create --name volume_backup --availability-zone az-central test_volume * The backup created successfully in "available" status on "az-central" AZ # Restore the backup to a new availability zone ("new_volume_dcn2") (central) [stack@site-undercloud-0 ~]$ cinder --os-volume-api-version 3.51 backup-restore --availability-zone az-dcn2 --name new_volume_dcn2 volume_backup * The volume created successfully and both volume & backup are available Moving to VERIFIED