Aborted qvm-backup does not reclaim disk space

User Support
1

I need a little help concerning qvm-backup…

A full backup has been taken to my NAS storage and been verified with qvm-backup-restore --verify-only.

But this backup process has stored over 15GB on my local SSD as well, which lead to nearly full disk space. I had to abort backup operation to avoid more bad…

  1. Is it normal that backup verification takes so much local disk space?
  2. How can I free and clean up dom0 from the aborted backup data?

A reboot did not hide or clean up disk, warning still shows up and shows 98% usage.

[user@dom0 ~]$ sudo lvs
  LV                                                  VG         Pool   Origin                                              Data%  Meta%  Move Log Cpy%Sync Convert
  pool00                                              qubes_dom0                                                            97.42  79.57

EDIT:
Qubes OS 4.0

Thanks!

2

More infos, suggested here:

[user@dom0 ~]$ sudo pvs
  PV                                                    VG         Fmt  Attr PSize   PFree
  /dev/mapper/luks-xxx qubes_dom0 lvm2 a--  117.11g 23.18g

[user@dom0 ~]$ sudo vgs
  VG         #PV #LV #SN Attr   VSize   VFree
  qubes_dom0   1 172   0 wz--n- 117.11g 23.18g

[user@dom0 ~]$ sudo lvs
  LV                                                  VG         Attr       LSize  Pool   Origin                                              Data%  Meta%  Move Log Cpy%Sync Convert
  pool00                                              qubes_dom0 twi-aotz-- 85.13g                                                            97.31  74.44
  root                                                qubes_dom0 Vwi-aotz-- 85.13g pool00                                                     25.98
  swap                                                qubes_dom0 -wi-ao----  7.41g    
# other qubes omitted

[user@dom0 ~]$ df -h
Filesystem                   Size  Used Avail Use% Mounted on
/dev/mapper/qubes_dom0-root   84G   21G   59G  26% /

There is a similar issue, which suggests

sudo lvextend -l +100%FREE /dev/mapper/qubes_dom0-pool00

I invoked qvm-backup + qvm-backup-restore --verify-only and did not alter logical volumes, so not sure about this.

Does Qubes by default create swap logical volume as seen in the third terminal output or is this a consequence of not enough RAM during backup run?

How might I reclaim disk space, that was available before backup started?

3

Update 2:

There has been free space left on the disk, which I could assign to logical volume pool00:

sudo lvextend -l +100%FREE /dev/mapper/qubes_dom0-pool00

The acute full disk warning now is solved.

Though I still have “lost” up to ~ 10GB of disk space due to aborted backup.

Is there any location for existent temporary files of qvm-backup and qvm-backup-restore --verify-only ?

Or do you have any resources, which help me research further on this issue?