snapshot keep in saving state backend with ceph

[Zeeshan Ali Shah]

I have glance backend with ceph , nova images also run in ceph . Both vm creation adn glance image create works fine. However when we create snap shot both via horizon and via nova image-create it keeps in saving state even when size of pools showed correct image size as of parent vm.

 dpkg -l glance
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name Version Architecture Description
+++-=================================-=====================-=====================-========================================================================
ii glance 1:2014.2.1-0ubuntu1~c all OpenStack Image Registry and Delivery Service - Daemons

dpkg -l ceph
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name Version Architecture Description
+++-=================================-=====================-=====================-========================================================================
ii ceph 0.87-1trusty amd64 distributed storage and file system
r

nova image-create a6f18d3e-6b5b-48a1-ab04-afc50665f4f1 18snap2
root@cloud21:/etc/glance# nova image-list
+--------------------------------------+---------------------------------+--------+--------------------------------------+
| ID | Name | Status | Server |
+--------------------------------------+---------------------------------+--------+--------------------------------------+
| 996726a5-7976-428f-8d1c-44fbc20b3b5f | 18snap2 | SAVING | a6f18d3e-6b5b-48a1-ab04-afc50665f4f1 |

nova list
+--------------------------------------+---------------------------+---------+-----------------+-------------+-------------------------------------+
| ID | Name | Status | Task State | Power State | Networks |
+--------------------------------------+---------------------------+---------+-----------------+-------------+-------------------------------------+
|
| a6f18d3e-6b5b-48a1-ab04-afc50665f4f1 | admin-instancefedcloud182 | ACTIVE | image_uploading | Running | admin-net=10.1.2.10 |

Now if we see rbd info it shows the correct size

rbd info images/996726a5-7976-428f-8d1c-44fbc20b3b5f

rbd image '996726a5-7976-428f-8d1c-44fbc20b3b5f':
 size 20480 MB in 2560 objects
 order 23 (8192 kB objects)
 block_name_prefix: rbd_data.780581989a2f9
 format: 2
 features: layering

Compare it with VM rbd

rbd info vms/a6f18d3e-6b5b-48a1-ab04-afc50665f4f1_disk
rbd image 'a6f18d3e-6b5b-48a1-ab04-afc50665f4f1_disk':
 size 20480 MB in 5120 objects
 order 22 (4096 kB objects)
 block_name_prefix: rbd_data.77fe06b8b4567
 format: 2
 features: layering

but after hour vm is keep in Image Uploading state . via nova list
and nova image-list shows Saving state

[Feilong Wang (flwang)]

We're also using rbd as the backend, but we never see this issue. Can you pls provide your glance-api.conf and nova.conf?

[sampsonhuo (13261149612-5)]

 Zeeshan Ali Shah, We encountered the same problem.

As the glance source code glance/api/v1/images.py delete function says, when the image's location attribute is None, it will not delete image data form backend storage.

[Kairat Kushaev (kkushaev)]

Please provide an additional info about you configuration:
1) glance-api.conf
2) log files

[Ian Cordasco (icordasc)]

This has been marked incomplete for well over a year with no response from the reporter. Closing as invalid.