Freeing cinder space from previously errored volumes?

asked 2018-07-09 18:03:01 -0600

KevK gravatar image

After being unable to delete a snapshot (and a volume that used it) from Openstack even after trying --force cmdline options, I deleted the snapshot and the corresponding volume's entry in the mariaDB instead (snapshot was 50 GB and volume was 80). Now, there is a constant 130 GB of space unavailable for allocation of new volumes (from checking vgs). Is there any way to either clear the volumes on iscsi or openstack that I'm missing?

vgs output:

  VG             #PV #LV #SN Attr   VSize    VFree
  *other volume info*
  cinder-volumes   1   6   1 wz--n- <412.00g <172.00g

openstack volume list output:

+-------------------------+--------------+--------+------+---------------------------+
| ID                      | Display Name | Status | Size | Attached to               |
+-------------------------+--------------+--------+------+---------------------------+
| 6f109472-ad79-4320-a7fc |              | in-use |   20 | Attached to ver1.1 on /dev/vda  |
| 546e2b45-cecf-4612-a2f6 |              | in-use |   20 | Attached to car1.1 on /dev/vda  |
| 8cf407f2-804a-450e-bd52 |              | in-use |   20 | Attached to tar1.1 on /dev/vda  |
| 0ec27f89-7cbd-4611-a374 |              | in-use |   50 | Attached to bar1.1 on /dev/vda  |
+-------------------------+--------------+--------+------+---------------------------+
edit retag flag offensive close merge delete