Hi Lionel,

You can try to delete that snapshot directly from the database.

In case of using psql [1], once you've logged in to your database, you can run this query:

$ select * from snapshots where vm_id = '<your_vm_id>';

This one would list the snapshots associated with a VM by its id.

In case you don't have you vm_id, you can locate it by querying:
$ select * from vms where vm_name = 'nil';

This one would show you some details about a VM by its name (including the vm's id).

Once you've found the relevant snapshot, you can delete it by running:
$ delete from snapshots where snapshot_id = '<your_snapshot_id>';

This one would delete the desired snapshot from the database.

Since it's a delete operation, I would suggest confirming the ids before executing it.

Hope you've found it useful!

[1] https://www.ovirt.org/documentation/install-guide/appe-Preparing_a_Remote_PostgreSQL_Database_for_Use_with_the_oVirt_Engine/

Regards,

Shani Leviim

On Fri, Feb 23, 2018 at 9:25 AM, Lionel Caignec <caignec@cines.fr> wrote:

Hi,

i've a problem with snapshot. On one VM i've a "snapshot" ghost without name or uuid, only information is size (see attachment). In the snapshot tab there is no trace about this disk.

In database (table images) i found this :
f242cc9a-56c1-4ae4-aef0-f75eb01f74b1 | 2018-01-17 22:02:00+01 | 2748779069440 | 00000000-0000-0000-0000-000000000000 | 00000000-0000-0000-0000-000000000000 | 4 | 2018-01-18 22:01:20.5+01 | 0dd2090c-3491-4fa1-98c3-54ae88be793c
| 2 | 4 | 17e26476-cecb-441d-a5f7-46ab3ef387ee | 2018-01-17 22:01:29.663334+01 | 2018-01-19 08:40:14.345229+01 | f | 1 | 2
1c7650fa-542b-4ec2-83a1-d2c1c31be5fd | 2018-01-17 22:02:03+01 | 5368709120000 | 00000000-0000-0000-0000-000000000000 | 00000000-0000-0000-0000-000000000000 | 4 | 2018-01-18 22:01:20.84+01 | 0dd2090c-3491-4fa1-98c3-54ae88be793c
| 2 | 4 | bf834a91-c69f-4d2c-b639-116ed58296d8 | 2018-01-17 22:01:29.836133+01 | 2018-01-19 08:40:19.083508+01 | f | 1 | 2
8614b21f-c0de-40f2-b4fb-e5cf193b0743 | 2018-02-09 23:00:44+01 | 5368709120000 | 00000000-0000-0000-0000-000000000000 | 00000000-0000-0000-0000-000000000000 | 4 | 2018-02-16 23:00:02.855+01 | 390175dc-baf4-4831-936a-5ea68fa4c969

But i does not know which line is my disk. Is it possible to delete directly into database?
Or is it better to dump my disk to another new and delete the "corrupted one"?

Another thing, when i try to move the disk to another storage domain i always get "uncaght exeption occured ..." and no error in engine.log.

Thank you for helping.

--
Lionel Caignec

_______________________________________________
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users