RBD incremental backup creation raise exception

[wanghao]

reproduction steps:

Step1: create an RBD volume
Step2: create an RBD backup（full backup）
Step3: create an incremental backup

Cinder backup raises an exception:

DEBUG cinder.backup.drivers.ceph [req-9b1bc514-779c-4855-a082-29c680fcaf73 req-92e696e1-b516-457f-ae53-bcb44d3ca1e2 admin None] Volume file is RBD: attempting incremental backup. {{(pid=1677774) backup /opt/stack/cinder/cinder/backup/drivers/ceph.py:940}}
/build/ceph-13.2.2/src/common/Mutex.cc: In function 'void Mutex::Lock(bool)' thread 7f6f55de4740 time 2019-03-02 03:51:13.727069
/build/ceph-13.2.2/src/common/Mutex.cc: 110: FAILED assert(r == 0)
 ceph version 13.2.2 (02899bfda814146b021136e9d8e80eba494e1126) mimic (stable)
 1: (ceph::__ceph_assert_fail(char const*, char const*, int, char const*)+0x102) [0x7f6f415605c2]
 2: (()+0x26c787) [0x7f6f41560787]
 3: (Mutex::Lock(bool)+0x1de) [0x7f6f415354ee]
 4: (()+0x92a02) [0x7f6f3ef5ba02]
 5: (()+0x115dbd) [0x7f6f3efdedbd]
 6: (rbd_snap_get_namespace_type()+0x29) [0x7f6f3ef2c9d9]
 7: (()+0x26fac) [0x7f6f3f478fac]
 8: (()+0x1521d) [0x7f6f3f46721d]
 9: (()+0x67878) [0x7f6f3f4b9878]
 10: (()+0x1d21f) [0x7f6f4a82521f]
 11: (PyEval_EvalFrameEx()+0xab0) [0x56214edfaa30]
 12: (PyEval_EvalFrameEx()+0x52b2) [0x56214edff232]
 13: (PyEval_EvalFrameEx()+0x52b2) [0x56214edff232]
 14: (PyEval_EvalFrameEx()+0x52b2) [0x56214edff232]
 15: (PyEval_EvalCodeEx()+0x6da) [0x56214edf7d0a]
 16: (PyEval_EvalFrameEx()+0x567e) [0x56214edff5fe]
 17: (PyEval_EvalFrameEx()+0x52b2) [0x56214edff232]
 18: (PyEval_EvalCodeEx()+0x6da) [0x56214edf7d0a]
 19: (()+0x10f8bc) [0x56214ee138bc]
 20: (PyObject_Call()+0x3e) [0x56214ede377e]
 21: (PyEval_EvalFrameEx()+0x2aa1) [0x56214edfca21]
 22: (PyEval_EvalFrameEx()+0x52b2) [0x56214edff232]
 23: (PyEval_EvalFrameEx()+0x52b2) [0x56214edff232]
 24: (PyEval_EvalCodeEx()+0x6da) [0x56214edf7d0a]
 25: (()+0x10f8bc) [0x56214ee138bc]
 26: (PyObject_Call()+0x3e) [0x56214ede377e]
 27: (PyEval_EvalFrameEx()+0x2aa1) [0x56214edfca21]
 28: (PyEval_EvalFrameEx()+0x52b2) [0x56214edff232]
 29: (PyEval_EvalCodeEx()+0x6da) [0x56214edf7d0a]
 30: (()+0x10f8bc) [0x56214ee138bc]
 31: (()+0x1280de) [0x56214ee2c0de]
 32: (PyObject_Call()+0x3e) [0x56214ede377e]
 33: (()+0x1845f7) [0x56214ee885f7]
 34: (PyObject_Call()+0x3e) [0x56214ede377e]
 35: (PyEval_EvalFrameEx()+0x2aa1) [0x56214edfca21]
 36: (PyEval_EvalCodeEx()+0x6da) [0x56214edf7d0a]
 37: (()+0x10f8bc) [0x56214ee138bc]
 38: (()+0x1280de) [0x56214ee2c0de]
 39: (PyObject_Call()+0x3e) [0x56214ede377e]
 40: (PyEval_CallObjectWithKeywords()+0x6d) [0x56214ee0383d]
 41: (()+0x36fc) [0x7f6f547e16fc]
 42: (()+0x3056) [0x7f6f547e1056]
 43: [0x7f6f380af910]
 NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.

The backup status is error:

f7fe4f5e-f294-4a8b-bca5-51a5d69fe564 | 02a4fdbb-b68f-4c5d-9b6a-aefffb8a03c5 | error | - | 1 | 0 | backups

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/640564

[OpenStack Infra (hudson-openstack)]

Change abandoned by wanghao (<email address hidden>) on branch: master
Review: https://review.openstack.org/640564
Reason: same patch as https://review.openstack.org/#/c/625396/