[PVE-User] Backup to PBS fails

[Konold, Martin]

Hi there,

I have a test setup investigating the migration from oVirt to Proxmox 
which used to work nicely but now only creates single byte length 
entries on PBS datastores.

Firstly backup to a cephfs storage works as expected:
root at hbase01 ~ # vzdump 100 --mode snapshot --node hbase01 --compress 
zstd --all 0 --storage cephfs01 --mailnotification always
INFO: starting new backup job: vzdump 100 --mode snapshot 
--mailnotification always --node hbase01 --storage cephfs01 --all 0 
--compress zstd
INFO: Starting Backup of VM 100 (qemu)
INFO: Backup started at 2021-09-07 13:38:10
INFO: status = stopped
INFO: backup mode: stop
INFO: ionice priority: 7
INFO: VM Name: CentOS-8-100.h.konsec.com
INFO: include disk 'scsi0' 'pool01:base-100-disk-0' 10G
INFO: creating vzdump archive 
'/mnt/pve/cephfs01/dump/vzdump-qemu-100-2021_09_07-13_38_10.vma.zst'
INFO: starting template backup
INFO: /usr/bin/vma create -v -c 
/mnt/pve/cephfs01/dump/vzdump-qemu-100-2021_09_07-13_38_10.tmp/qemu-server.conf 
exec:zstd --rsyncable --threads=1 > 
/mnt/pve/cephfs01/dump/vzdump-qemu-100-2021_09_07-13_38_10.vma.dat 
drive-scsi0=/dev/rbd/pool01/base-100-disk-0
INFO: progress 0% 0/10737418240 0
INFO: progress 1% 107413504/10737418240 64716800
....

Backup to a Proxmox Backup Server PBS fails though
root at hbase01 ~ # vzdump 100 --mode snapshot --node hbase01 --compress 
zstd --all 0 --storage PVE-FE-3 --mailnotification always
INFO: starting new backup job: vzdump 100 --all 0 --node hbase01 
--mailnotification always --storage PVE-FE-3 --mode snapshot --compress 
zstd
INFO: Starting Backup of VM 100 (qemu)
INFO: Backup started at 2021-09-07 13:38:31
INFO: status = stopped
INFO: backup mode: stop
INFO: ionice priority: 7
INFO: VM Name: CentOS-8-100.h.konsec.com
INFO: include disk 'scsi0' 'pool01:base-100-disk-0' 10G
INFO: creating Proxmox Backup Server archive 
'vm/100/2021-09-07T11:38:31Z'
INFO: starting kvm to execute backup task
Use of uninitialized value in split at 
/usr/share/perl5/PVE/QemuServer/Cloudinit.pm line 100.
/dev/rbd45
generating cloud-init ISO
ERROR: VM 100 qmp command 'backup' failed - backup connect failed: 
command error: EACCES: Permission denied
INFO: aborting backup job
INFO: stopping kvm after backup task
trying to acquire lock...
  OK
ERROR: Backup of VM 100 failed - VM 100 qmp command 'backup' failed - 
backup connect failed: command error: EACCES: Permission denied
INFO: Failed at 2021-09-07 13:38:33
INFO: Backup job finished with errors
job errors

The target storage exists and is known to PVE.

root at hbase01 ~ # pvesm status
Name            Type     Status           Total            Used       
Available        %
PVE-FE           pbs     active     59826418816      1498314752     
58328104064    2.50%
PVE-FE-2         pbs     active     58345629568        17525504     
58328104064    0.03%
PVE-FE-3         pbs     active     58345629568        17525504     
58328104064    0.03%
cephfs01      cephfs     active      1128591360        39051264      
1089540096    3.46%
local            dir     active        15718400        10359288         
5359112   65.91%
pool01           rbd     active      1485625252       396081828      
1089543424   26.66%
pool12           rbd     active      4253993533      2619678397      
1634315136   61.58%

Is this a problem introduced with an update of PBS?

I tried adding backup users to PBS and using API-tokens to no avail.

Both PVE and PBS are most uptodate (no-subscription)

-- 
Regards
ppa. Martin Konold

--
Martin Konold - Prokurist, CTO
KONSEC GmbH -⁠ make things real
Amtsgericht Stuttgart, HRB 23690
Geschäftsführer: Andreas Mack
Im Köller 3, 70794 Filderstadt, Germany