Ovirt Storage not reclaim
I have attached 7TB disk ( pre allocated) for VM ( Redhat 8 ).. Due to some space issue we have unattached disk from VM and deleted it. but free space still has not changed and we can't use that space for other purposes . What would be the issue ? Thanks, Ranesh..
Attaching disk -> how, where ? Best Regards,Strahil Nikolov On Sun, Jul 11, 2021 at 17:02, Ranesh Tharanga Perera<raneshtharanga@gmail.com> wrote: I have attached 7TB disk ( pre allocated) for VM ( Redhat 8 ).. Due to some space issue we have unattached disk from VM and deleted it. but free space still has not changed and we can't use that space for other purposes . What would be the issue ? Thanks,Ranesh.._______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/privacy-policy.html oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/F6NV6ZQCYQCWWL...
On Sun, Jul 11, 2021 at 5:03 PM Ranesh Tharanga Perera <raneshtharanga@gmail.com> wrote:
I have attached 7TB disk ( pre allocated) for VM ( Redhat 8 ).. Due to some space issue we have unattached disk from VM and deleted it.
Did you enable the "discard after delete" option in the storage domain advanced options? If you did not, the logical volume is deleted, but the blocks allocated on the server side from this logical volume are not discarded. Since you deleted the logical volume, it is not possible to discard it now, but you do this: Try this: - Put the storage domain to maintenance - Enable "discard-after delete" - Active the storage domain - Create new raw preallocated disk filling the entire domain - Delete the disk Nir
We have enabled both options discard-after delete and wipe after delete but nothing is working , I have tried to creating small disk ( like 1 GB ). after deleting that disk space has reclaimed succsuuufully. Seems this happens for larger size disk On Mon, Jul 12, 2021 at 1:10 AM Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Jul 11, 2021 at 5:03 PM Ranesh Tharanga Perera <raneshtharanga@gmail.com> wrote:
I have attached 7TB disk ( pre allocated) for VM ( Redhat 8 ).. Due to
some space issue we have unattached disk from VM and deleted it.
Did you enable the "discard after delete" option in the storage domain advanced options?
If you did not, the logical volume is deleted, but the blocks allocated on the server side from this logical volume are not discarded.
Since you deleted the logical volume, it is not possible to discard it now, but you do this:
Try this: - Put the storage domain to maintenance - Enable "discard-after delete" - Active the storage domain - Create new raw preallocated disk filling the entire domain - Delete the disk
Nir
On Mon, Jul 12, 2021 at 12:27 AM Ranesh Tharanga Perera <raneshtharanga@gmail.com> wrote:
We have enabled both options discard-after delete and wipe after delete but nothing is working , I have tried to creating small disk ( like 1 GB ). after deleting that disk space has reclaimed succsuuufully. Seems this happens for larger size disk
If it worked for 1g disk, it should work for a bigger disk. If you can reproduce this again, please file a ovirt/vdsm bug and attach vdsm logs and /var/log/messages (or output of journalcltl). BTW, do you have a real need for wipe-after-delete? This can make deletion much slower, since it may need to allocate the entire disk on the server side for writing zeroes just to deallocate all space right after that when we discard the disk. This option is needed only if you want to ensure that sensitive data cannot leak to another disk on the same storage, maybe used by another tenant.
On Mon, Jul 12, 2021 at 1:10 AM Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Jul 11, 2021 at 5:03 PM Ranesh Tharanga Perera <raneshtharanga@gmail.com> wrote:
I have attached 7TB disk ( pre allocated) for VM ( Redhat 8 ).. Due to some space issue we have unattached disk from VM and deleted it.
Did you enable the "discard after delete" option in the storage domain advanced options?
If you did not, the logical volume is deleted, but the blocks allocated on the server side from this logical volume are not discarded.
Since you deleted the logical volume, it is not possible to discard it now, but you do this:
Try this: - Put the storage domain to maintenance - Enable "discard-after delete" - Active the storage domain - Create new raw preallocated disk filling the entire domain - Delete the disk
Nir
Many thanks for the info. We don't need that option. How we can disable that? just only uncheck that property box ? Is there any downtime? What happens if we delete that VM and recreate it? All space will reclaim? Thanks, Ranesh.. On Mon, Jul 12, 2021 at 3:46 AM Nir Soffer <nsoffer@redhat.com> wrote:
On Mon, Jul 12, 2021 at 12:27 AM Ranesh Tharanga Perera <raneshtharanga@gmail.com> wrote:
We have enabled both options discard-after delete and wipe after delete
but nothing is working , I have tried to creating small disk ( like 1 GB ). after deleting that disk space has reclaimed succsuuufully.
Seems this happens for larger size disk
If it worked for 1g disk, it should work for a bigger disk.
If you can reproduce this again, please file a ovirt/vdsm bug and attach vdsm logs and /var/log/messages (or output of journalcltl).
BTW, do you have a real need for wipe-after-delete? This can make deletion much slower, since it may need to allocate the entire disk on the server side for writing zeroes just to deallocate all space right after that when we discard the disk.
This option is needed only if you want to ensure that sensitive data cannot leak to another disk on the same storage, maybe used by another tenant.
On Mon, Jul 12, 2021 at 1:10 AM Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Jul 11, 2021 at 5:03 PM Ranesh Tharanga Perera <raneshtharanga@gmail.com> wrote:
I have attached 7TB disk ( pre allocated) for VM ( Redhat 8 ).. Due
to some space issue we have unattached disk from VM and deleted it.
Did you enable the "discard after delete" option in the storage domain advanced options?
If you did not, the logical volume is deleted, but the blocks
allocated on the
server side from this logical volume are not discarded.
Since you deleted the logical volume, it is not possible to discard it now, but you do this:
Try this: - Put the storage domain to maintenance - Enable "discard-after delete" - Active the storage domain - Create new raw preallocated disk filling the entire domain - Delete the disk
Nir
On Mon, Jul 12, 2021 at 7:00 AM Ranesh Tharanga Perera <raneshtharanga@gmail.com> wrote:
Many thanks for the info.
We don't need that option. How we can disable that? just only uncheck that property box ? Is there any downtime? What happens if we delete that VM and recreate it? All space will reclaim?
You need to put the domain to maintenance, uncheck the option and activate the domain. If you have vms using this domain, you need to shut them down, or migrate the disk to another storage, and then migrate the vm back to the storage. Since migrating the disk requires coping the disks twice, maybe it is better to create a new LUN(s) on the server side, create a new storage domain (without wipe-after-delete), and migrate the disks to the new storage domain (no downtime required). Finally when old storage domain is unused, remove it. With this flow you need to copy the data only once. If you can have short downtime, maybe during a maintenance window shutting down the vms and modifying the domain may be better. I'm not sure why we require moving a storage domain to maintenance for changing such configuration, this sounds completely unnessercery, so I think RFE for allowing online configuration changes is a good idea. Nir
Thanks, Ranesh..
On Mon, Jul 12, 2021 at 3:46 AM Nir Soffer <nsoffer@redhat.com> wrote:
On Mon, Jul 12, 2021 at 12:27 AM Ranesh Tharanga Perera <raneshtharanga@gmail.com> wrote:
We have enabled both options discard-after delete and wipe after delete but nothing is working , I have tried to creating small disk ( like 1 GB ). after deleting that disk space has reclaimed succsuuufully. Seems this happens for larger size disk
If it worked for 1g disk, it should work for a bigger disk.
If you can reproduce this again, please file a ovirt/vdsm bug and attach vdsm logs and /var/log/messages (or output of journalcltl).
BTW, do you have a real need for wipe-after-delete? This can make deletion much slower, since it may need to allocate the entire disk on the server side for writing zeroes just to deallocate all space right after that when we discard the disk.
This option is needed only if you want to ensure that sensitive data cannot leak to another disk on the same storage, maybe used by another tenant.
On Mon, Jul 12, 2021 at 1:10 AM Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Jul 11, 2021 at 5:03 PM Ranesh Tharanga Perera <raneshtharanga@gmail.com> wrote:
I have attached 7TB disk ( pre allocated) for VM ( Redhat 8 ).. Due to some space issue we have unattached disk from VM and deleted it.
Did you enable the "discard after delete" option in the storage domain advanced options?
If you did not, the logical volume is deleted, but the blocks allocated on the server side from this logical volume are not discarded.
Since you deleted the logical volume, it is not possible to discard it now, but you do this:
Try this: - Put the storage domain to maintenance - Enable "discard-after delete" - Active the storage domain - Create new raw preallocated disk filling the entire domain - Delete the disk
Nir
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/privacy-policy.html oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/CQKAZBOHAPR4IB...
participants (3)
-
Nir Soffer -
Ranesh Tharanga Perera -
Strahil Nikolov