Zip file corrupted when I try to download theme from Proxmox Server

I

IWIOS

Member
Dec 3, 2021
28
0
21
34
Hi

I have an issue on my infrastructure

I have a proxmox server cluster (with 3 nodes and CEPH)
I have a physical server with PBS installed

When I try from Proxmox Server UI to download simple file from a backup no problem
But if I try to download a folder with many files (~50Gb), the file downlaod and finishe correctly but zip seems corrupted (and same with .tar.zst extension)


Informations:

All nodes of Proxmox sevrer are in
Virtual Environment 8.4.14

PBS is on Backup Server 3.4.8

Thank all for helping me
 
corrupted how? is it truncated, does it contain bogus data, ..?
 
When I try to open on Arc (Zip fil manager on Debian) I have this error:


"Possibly truncated or corrupted zip archive"

And nothing appear on the archive
 
does the file have a reasonable size or is it actually empty?
 
could you do "stat /path/to/file" and post the output here? thanks
 
Code: 
stat 2023.tar.zst                                                                                                                                                                                                                                                                                              [14:19:35]
  Fichier : 2023.tar.zst
   Taille : 37413602358 Blocs : 73073464   Blocs d'E/S : 4096   regular file
Périphérique : 253/1    Inœud : 9532925     Liens : 1
Accès : (0664/-rw-rw-r--)  UID : ( 1000/   kevin)   GID : ( 1000/   kevin)
 Accès : 2025-12-14 23:15:29.090928861 +0100
Modif. : 2025-12-14 23:15:06.954928468 +0100
Changt : 2025-12-14 23:15:07.006928469 +0100
  Créé : 2025-12-14 22:44:24.706895825 +0100
 
okay, so it's not a 32G file size limit ;) is this a VM or a container backup? are you using the PVE or the PBS UI to download?
 
It's from a VM backup (Nextcloud) and trying to download entire folder from Nextcloud user

I'm using PVE UI to download
 
thanks!

if you retry (please ensure there are no other file restore tasks running in parallel), could you post the /var/log/proxmox-backup/file-restore/qemu.log file corresponding to your file restore?
 
Hum... very strange during backup procedure and ZIP download, no log in the path you have put

Las modification of this file is
Code: 
-rw-r--r-- 1 root root 6.6K Dec 14 21:03 /var/log/proxmox-backup/file-restore/qemu.log
With:

Code: 
watchdog expired, shutting down
 
Last edited:
you need to look on the right node, in case this is a cluster.
 
you need to look on the node that your PVE UI talks to/is connected to
 
fabian said:
you need to look on the node that your PVE UI talks to/is connected to
Click to expand...
Oh, sorry for missunderstood

This is the log:


Code: 
[2025-12-15T14:38:23.677Z INFO  proxmox_restore_daemon::proxmox_restore_daemon::watchdog] Inhibit added: 1
I/O error, dev vda, sector 49346584 op 0x0:(READ) flags 0x80700 phys_seg 2 prio class 0
I/O error, dev vda, sector 49346584 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[2025-12-15T14:38:30.439Z ERROR proxmox_restore_daemon::proxmox_restore_daemon::api] file or dir streaming task failed - Input/output error (os error 5)
[2025-12-15T14:38:30.440Z INFO  proxmox_restore_daemon::proxmox_restore_daemon::watchdog] Inhibit dropped: 0
 
that might be possible, yes. or a connection issue, but if it is reproducibly always the same sector than I'd say it's likely that the corresponding chunk has an issue. did you try verifying the snapshot already?
 
Hi, After many search, and a job of re-verify of backup, Verify Status is failed for all backup for this VM
In my opinion, an hardware issue have broke some chunk (~12000 chunk in .0.bad state)
 
sorry to hear that. if the VM is still around, you could run another backup, that might re-upload some of the missing chunks and thus "heal" some of the backups.
 
You must log in or register to reply here.
Top Bottom