cinder incremental backup fails with ceph "dict' object has no attribute 'name'""
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Cinder |
New
|
Undecided
|
Unassigned |
Bug Description
Hi helen,
Here my setup, devstack+ceph, where enable cinder backup services did below things
ubuntu@
+----+-
| ID | Volume ID | Status | Name | Size | Object Count | Container |
+----+-
+----+-
ubuntu@
+------
| ID | Status | Name | Size | Volume Type | Bootable | Attached to |
+------
| 2688b075-
+------
ubuntu@
+------
| Property | Value |
+------
| id | 7a984b15-
| name | None |
| volume_id | 2688b075-
+------
ubuntu@
+------
| ID | Volume ID | Status | Name | Size | Object Count | Container |
+------
| 7a984b15-
+------
ubuntu@
+------
| Property | Value |
+------
| id | d0181918-
| name | None |
| volume_id | 2688b075-
+------
ubuntu@
+------
| ID | Volume ID | Status | Name | Size | Object Count | Container |
+------
| 7a984b15-
| d0181918-
+------
ubuntu@
+------
| Property | Value |
+------
| availability_zone | nova |
| container | backups |
| created_at | 2018-02-
| data_timestamp | 2018-02-
| description | None |
| fail_reason | None |
| has_dependent_
| id | 7a984b15-
| is_incremental | False |
| name | None |
| object_count | 0 |
| size | 1 |
| snapshot_id | None |
| status | available |
| updated_at | 2018-02-
| volume_id | 2688b075-
+------
ubuntu@
+------
| Property | Value |
+------
| availability_zone | nova |
| container | backups |
| created_at | 2018-02-
| data_timestamp | 2018-02-
| description | None |
| fail_reason | 'dict' object has no attribute 'name' |
| has_dependent_
| id | d0181918-
| is_incremental | True |
| name | None |
| object_count | 0 |
| size | 1 |
| snapshot_id | None |
| status | error |
| updated_at | 2018-02-
| volume_id | 2688b075-
+------
While taking incremental backup it got failed with "'dict' object has no attribute 'name'" and also tried with --parent ( as like trove), still no luck.
Where fullbackup and restore work fine. Only with incremental getting isssue as above.
summary: |
- cinder incremental backup fails with ceph, where normal backup and - restore works succesfully + cinder incremental backup fails with ceph "dict' object has no attribute + 'name'"" |
Are the volumes stored in a Ceph backup as well? Because the Ceph backup driver doesn't need the "--incremental" parameter to do incremental backups if the source volume is also stored in Ceph, and it doesn't do incremental for non Ceph volumes.
Where did you get that error? On the client? The API? Scheduler? Backup? Volume?
Could you include the traceback? Because I have just tested RBD incremental to incremental with devstack+ceph and it worked just fine with the CLI commands you mentioned.