You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@cloudstack.apache.org by "Yiping Zhang (JIRA)" <ji...@apache.org> on 2015/04/10 19:13:12 UTC

[jira] [Created] (CLOUDSTACK-8376) ghost volumes in projects

Yiping Zhang created CLOUDSTACK-8376:
----------------------------------------

             Summary: ghost volumes in projects
                 Key: CLOUDSTACK-8376
                 URL: https://issues.apache.org/jira/browse/CLOUDSTACK-8376
             Project: CloudStack
          Issue Type: Bug
      Security Level: Public (Anyone can view this level - this is the default.)
          Components: Management Server
    Affects Versions: 4.3.2
         Environment: RHEL 6.5 x86_64, XenServer 6.2
            Reporter: Yiping Zhang


I first asked this in users mailing list, but got no answers other than someone thinks it might be a bug. So here I am:

We always create VM instances in project context, so that all resources are assigned to project instead of individual accounts.  I have a script to call listProjects API to show project’s resource limits and current usages.

Recently, I noticed that for a couple of projects, my script reports a different number of volumes from what I can see in web UI for these projects.  Poking around in mysql db, it looks like web UI only reports volumes in Ready state, while listProjects API reports volumes in both Ready and Destroy states.

Following is a project where UI says it has two volumes while listProjects API says it has nine:


mysql> select v.id, v.name, v.state, v.removed, v.path from volumes v, projects p where v.account_id = p.project_account_id and p.name='activityservice' and v.state != 'Expunged';

+------+-----------+---------+---------------------+--------------------------------------+

| id   | name      | state   | removed             | path                                 |

+------+-----------+---------+---------------------+--------------------------------------+

|  961 | ROOT-1053 | Destroy | 2015-03-02 21:36:09 | NULL                                 |

|  965 | ROOT-1061 | Destroy | 2015-03-03 00:25:17 | NULL                                 |

|  966 | ROOT-1063 | Destroy | 2015-03-03 00:28:16 | NULL                                 |

|  967 | ROOT-1066 | Destroy | 2015-03-03 00:41:48 | NULL                                 |

|  968 | ROOT-1067 | Destroy | 2015-03-03 00:49:11 | NULL                                 |

|  969 | ROOT-1069 | Destroy | 2015-03-03 00:53:08 | NULL                                 |

|  970 | ROOT-1070 | Destroy | 2015-03-03 00:53:52 | NULL                                 |

| 1000 | ROOT-1100 | Ready   | NULL                | 2c315f38-3a5f-4b2e-a765-332ddbe7a938 |

| 1217 | ROOT-1314 | Ready   | NULL                | 1d3f6716-d607-4d9d-b71c-93b7a88cd064 |

+------+-----------+---------+---------------------+--------------------------------------+

9 rows in set (0.00 sec)


mysql>


These ghost volumes are counted into project’s total usage of primary storage, so I’d like to get rid of them to free up space for real volumes.


How do I clean them up ? Is this a known issue ?




--
This message was sent by Atlassian JIRA
(v6.3.4#6332)