Backup failed to restore

[yakka]

hello i have problem now for one moth all my VM are down and i have i9 one VM 101 very important data need t be restored ..the web interface all time failed to restore gives error i guess proxmox team work again to solve make it possible

in my case i had backup dump vzdump-qemu-101-2021_03_11-01_56_38 when i try restore gives error so i extract this and made QM import to get new hdd to new vm with same number as it was i original machine and i also configure the dummy vm t be same as old one then i add HDD when i start vm all sound good but never boot it stop in this dracut: i read more than 100 pages on forum and everywhere no any solution can help if i exit dracut it spend all day scanning .....no reult at end and i try wrote regenerate it say command not found anybody can hep .....i can hire if any body can help solve it

 

[Ivan Dimitrov]

First it will easier if you post the exact error message when you try to restore the VM.
Also in PVE interface you can go to Storage --> "PBS volume" --> Backups. There you will see all the existing VM backups. You can select the one you want and restore it is as different VM. Basically create a clone from backup.
This can be helpful if you have messed the original VM.

 

[yakka]

Thank you for the reply my case was i had to move from server to other so when i made backup i move the dump by scp t new server there we could not do the restore
when i do restore from web interface here error log

*****************
restore vma archive: vma extract -v -r /var/tmp/vzdumptmp8435.fifo /var/lib/vz/dump/vzdump-qemu-101-2021_03_11-01_56_38.vma /var/tmp/vzdumptmp8435
CFG: size: 437 name: qemu-server.conf
DEV: dev_id=1 size: 1086626725888 devname: drive-scsi0
CTIME: Thu Mar 11 02:56:38 2021
Formatting '/var/lib/vz/images/106/vm-106-disk-0.raw', fmt=raw size=1086626725888
new volume ID is 'local:106/vm-106-disk-0.raw'
map 'drive-scsi0' to '/var/lib/vz/images/106/vm-106-disk-0.raw' (write zeros = 0)
progress 1% (read 10866327552 bytes, duration 40 sec)
progress 2% (read 21732589568 bytes, duration 41 sec)
progress 3% (read 32598851584 bytes, duration 159 sec)
vma: restore failed - short vma extent (1148416 < 3801600)
temporary volume 'local:106/vm-106-disk-0.raw' sucessfuly removed
no lock found trying to remove 'create' lock
TASK ERROR: command 'set -o pipefail && vma extract -v -r /var/tmp/vzdumptmp8435.fifo /var/lib/vz/dump/vzdump-qemu-101-2021_03_11-01_56_38.vma /var/tmp/vzdumptmp8435' failed: got signal 5
**************************************

 

[Ivan Dimitrov]

Do you use PBS at all? I can see option to manually import a VM backup.

Alternative solution - install PBS on SiteA (original) and siteB (target). Connect the two instance. Make backup on SiteA and sync the pbs on SiteB. Restore from SiteB.

 

[fabian]

vma: restore failed - short vma extent (1148416 < 3801600)

indicates that your backup archive is corrupt..

 

[yakka]

i do not have anymore access to the original server so what i have now are 2 version of backup

vzdump-qemu-101-2021_03_10-18_11_42.vma.zst 28gb size and one

vzdump-qemu-101-2021_03_11-01_56_38.vma 17 gb size booth dump was successful when i made them and size was this one so how i can be corrupted ..fabian

Do you use PBS at all? I can see option to manually import a VM backup.

Alternative solution - install PBS on SiteA (original) and siteB (target). Connect the two instance. Make backup on SiteA and sync the pbs on SiteB. Restore from SiteB.

 

[yakka]

when i try extract the second backup also got this error


zstd -d vzdump-qemu-101-2021_03_10-18_11_42.vma.zst /root/23/


_10-18_11_42.vma.zst : Read error (39) : premature end


zstd: /root/23/: unknown suffix -- ignored


root@prox:/var/lib/vz/dump# zstd -d vzdump-qemu-101-2021_03_10-18_11_42.vma.zst /root/23/


_10-18_11_42.vma.zst : Read error (39) : premature end


zstd: /root/23/: unknown suffix -- ignored

 

[yakka]

my problem now is how to get back some php file and msql for some web project was hosted in this VM 101 anybody can help i can pay for the job please

 

[fabian]

that sounds like there was some issue with your backup target storage and you never tested your backups?

the 'vma' binary is compiled with our fork of Qemu, so you can build a copy that has those checks disabled to attempt some form of recovery.

 

[yakka]

Thank you fabian but seams the git has problem


root@prox:~# git://git.proxmox.com/git/pve-qemu.git


-bash: git://git.proxmox.com/git/pve-qemu.git: No such file or directory


root@prox:~# wget https://git.proxmox.com/git/pve-qemu.git


--2021-04-27 12:48:37-- https://git.proxmox.com/git/pve-qemu.git


Resolving git.proxmox.com (git.proxmox.com)... 79.133.36.253, 2a01:7e0:0:424::5


Connecting to git.proxmox.com (git.proxmox.com)|79.133.36.253|:443... connected.


HTTP request sent, awaiting response... 404 Not Found


2021-04-27 12:48:38 ERROR 404: Not Found.

 

[fabian]

git works fine.. but if you are not able to clone the repo you probably won't be able to patch qemu for a recovery build..