Hi,
I hope someone can help as I have been struggling with the following issue regard backups of LXC instances from PVE 7 to a Proxmox Backup Server.
Up to now, we have been doing our backups to a NAS via a NFS share and this is working well.
We decided to try out Proxmox Backup Server as an alternative, but Ihave come across a strange issue with regards to backups of LXC instances.
We have both Ubuntu 20.04 and 22.04 LXC instances running on the PVE host. All instances use a 4GB root volume running off 'local-lvm' on the PVE host.
Doing initial backups of these instances I found that some took much longer than others to backup.
On further investigation I saw that this relates to the 'size' of the backup as reported in PBS. Some instances are reporting their 'size' in PBS in the range 200MB - 400MB which is 'correct', but others report their 'size' as being more than 200GB which is completely incorrect (root volume is 4GB in all cases)
I attempted to restore one of these bloated backups and it restores correctly (except for the time it takes to do the restore). The restored root volume reports the correct size of 4GB.
It makes no difference if I backup using 'snapshot' or 'shut down' mode the results are the same.
Does anyone have any idea what is going on?
Backing instances up to NFS or even 'local' does them quickly and with an appropriate size.
Both PVE7 and PBS are fully updated. They are different physical hosts.
Thanks
Shaun
I hope someone can help as I have been struggling with the following issue regard backups of LXC instances from PVE 7 to a Proxmox Backup Server.
Up to now, we have been doing our backups to a NAS via a NFS share and this is working well.
We decided to try out Proxmox Backup Server as an alternative, but Ihave come across a strange issue with regards to backups of LXC instances.
We have both Ubuntu 20.04 and 22.04 LXC instances running on the PVE host. All instances use a 4GB root volume running off 'local-lvm' on the PVE host.
Doing initial backups of these instances I found that some took much longer than others to backup.
On further investigation I saw that this relates to the 'size' of the backup as reported in PBS. Some instances are reporting their 'size' in PBS in the range 200MB - 400MB which is 'correct', but others report their 'size' as being more than 200GB which is completely incorrect (root volume is 4GB in all cases)
I attempted to restore one of these bloated backups and it restores correctly (except for the time it takes to do the restore). The restored root volume reports the correct size of 4GB.
It makes no difference if I backup using 'snapshot' or 'shut down' mode the results are the same.
Does anyone have any idea what is going on?
Backing instances up to NFS or even 'local' does them quickly and with an appropriate size.
Both PVE7 and PBS are fully updated. They are different physical hosts.
Thanks
Shaun
