Search results

Yeah that is what Thomas discussed in this, one approach would be trying to do this with the qemu guest agent. This is installed on basically all VMs anyway and doesn't require any more configuration.

Hi, check this: https://bugzilla.proxmox.com/show_bug.cgi?id=3501 That might answer your question :)

HAProxy does also support HTTP/2 and might be a bit more performant than apache2.

Hi, no there is currently no built-in option for this. You could code a script for this yourself using the API, but I would like to see this as a feature implemented as well :)

So you mean use an iSCSI LUN as the storage for PBS? Just mount the lun on your PBS server and point a datastore at it, done.

Yeah, the proxmox-backup-client map doesn't work that well with lvm because you have two LVMs with the same ID and name on your VM. Maybe a future improvement for the proxmox-file-restore client? :)

Is there a "CLI Browser" for the file systems? So you can navigate in the folders and select the files you want to extract?

You don't need PBS 2.x to use the file restore feature, only an up-to-date PVE 7.x installation

Hello, take a look at this solution I posted: https://forum.proxmox.com/threads/copy-backups-to-external-drive.88791/#post-389048 That is probably the way you want to go about this.

You could also use the API, that way you don't need to execute a cli command on the pbs host

Hi, also wanted to chime in here, I ran into the same issue. Have you decided on which option will be implemented yet? Or is the 2nd option already implemented?

Both the GUI and Remote Sync/Backup Jobs use the API. The API is avaible on Port 8007. It would probably be best for you to implement a firewall so only authorized IPs can access Port 8007 on your Server.

You can do a file-level restore in a bit more manual way by using the proxmox-backup-client map command. This maps the disk to a read-only loopback device. You should then be able to mount refs by hand. This requires a refs driver... There is a company well known for their filesystem ports that...

That is expected. When VMs are shut down the qemu process is closed. The qemu process is the one tracking your changes in the blocks of the disk. It will need to read the entire disk again, but it will only transfer missing chunks (as you already said). The Proxmox Team said they want to think...

thats true :) You don't want to have too much io wait. Just wanted to point that part with network traffic out as a lot of people think that if the dirty bitmap is destroyed it will also upload all the chunks to the PBS server again. That is not what is happening.

Hi, the install from ISO is also a relativly vanilla debian install. Is PBS reachable with its IPv4 adress? You only show the netstat output, read here why netstat doesn't always list services bound to both IPv4 and v6...

This would only stress the disk the VM is on. There is no additional traffic for VMs that don't have dirty bitmaps, it just takes longer and needs to read the entire disk.

This is what a command could look like: proxmox-backup-client map vm/117/2021-01-04T03:00:55Z drive-scsi0.img --repository root@pam@localhost:8007:store02 The <snapshot> is vm/117/2021-01-04T03:00:55Z, the <archive-name> is drive-scsi0.img which is how your disk in the VM is called. You can...

Yes you can do that. The datastore just points to a directory. I don't know how exactly you set up your first datastore. Is it zfs? if yes, you can just create another dataset on that zpool and point a datastore to it. If it is something else, you will have to see if you can create another...

Both PVE Storages are pointed at the same PBS Datastore. When the Backup was first created, it was created with the user root@pam (because you did the backup to the PBS storage). When you want to create another backup to the UserPBS storage, it detects there already is a backup on that datastore...