dockerctl restore error: MD5 CHECK FAILED

Several SWARM tests failed today on 8.0-518, one of scenarios:

Scenario:
1. Revert snapshot "deploy_ha_one_controller_flat"
2. Backup master
3. Check backup
4. Run OSTF
5. Add 1 node with compute role
6. Restore master

Actual:

'
', 'Restore failed!
'], 'stdout': ['Stopping containers...
', 'Stopping nginx...
', 'Stopping rabbitmq...
', 'Stopping astute...
', 'Stopping rsync...
', 'Stopping keystone...
', 'Stopping postgres...
', 'Stopping rsyslog...
', 'Stopping nailgun...
', 'Stopping cobbler...
', 'Stopping ostf...
', 'Stopping mcollective...
', 'Output filename is: /var/backup/fuel/restore-2016-02-04_0438//fuel_backup.tar
', 'Decompressing...
', ' 10% 743.42 / 7433.99 MB
 20% 1486.82 / 7433.99 MB
 25% 1898.91 / 7433.99 MB
 30% 2230.22 / 7433.99 MB
 40% 2973.60 / 7433.99 MB
 50% 3717.02 / 7433.99 MB
 51% 3797.70 / 7433.99 MB
 60% 4460.42 / 7433.99 MB
 70% 5203.85 / 7433.99 MB
 76% 5696.61 / 7433.99 MB
 80% 5947.23 / 7433.99 MB
 90% 6690.65 / 7433.99 MB
100% 7433.99 / 7433.99 MB

', 'Average DeCompression Speed: 45.323MB/s
', 'MD5 CHECK FAILED.
', 'Stored:cc4b50cab33e62354bb312e514d7f585
', 'Output file:2f2895074ec5b268a300275008f537f9
Fatal error - exiting
'

Failed tests:
ha_one_controller_backup_restore https://mirantis.testrail.com/index.php?/tests/view/2655145
backup_restore_master_base https://mirantis.testrail.com/index.php?/tests/view/2655146&group_by=cases:section_id&group_order=asc&group_id=6678
neutron_tun_ha_backup_restore https://mirantis.testrail.com/index.php?/tests/view/2655147&group_by=cases:section_id&group_order=asc&group_id=6678
create_backup_reset_restore_and_deploy_via_cli https://mirantis.testrail.com/index.php?/tests/view/2655240&group_by=cases:section_id&group_order=asc&group_id=6678