Backup error Input/output error

t89s

New Member
Jan 11, 2021
4
0
1
35
Hi everyone, Sorry for my English, I am French speaking.

I recently installed a proxmox backup server to add a backup solution for my three proxmox ve. I encountered a problem with a physical host backup.

Version proxmox-backup-client Installed: 1.0.6-1

On pve1 I get the error below, while on pve2 the backup is done correctly


Code:
proxmox-backup-client backup root.pxar:/ --repository 'Backup_user@pbs!Backup_Agent@172.20.5.253:Backup'
Starting backup: host/pve1/2021-01-11T21:09:05Z
Client name: pve1
Starting backup protocol: Mon Jan 11 22:09:05 2021
No previous manifest available.
Upload directory '/' to 'Backup_user@pbs!Backup_Agent@172.20.5.253:8007:Backup' as root.pxar.didx
skipping mount point: "dev"
skipping mount point: "etc/pve"
skipping mount point: "opt/VM"
skipping mount point: "opt/VM_SSD"
skipping mount point: "proc"
skipping mount point: "rpool"
skipping mount point: "run"
skipping mount point: "sys"
Error: error at "usr/lib/x86_64-linux-gnu/libdcerpc-binding.so.0.0.1": Input/output error (os error 5)

GUI

Code:
2021-01-11T22:09:05+01:00: starting new backup on datastore 'Backup': "host/pve1/2021-01-11T21:09:05Z"
2021-01-11T22:09:05+01:00: created new dynamic index 1 ("host/pve1/2021-01-11T21:09:05Z/root.pxar.didx")
2021-01-11T22:09:05+01:00: created new dynamic index 2 ("host/pve1/2021-01-11T21:09:05Z/catalog.pcat1.didx")
2021-01-11T22:09:18+01:00: backup ended and finish failed: backup ended but finished flag is not set.
2021-01-11T22:09:18+01:00: removing unfinished backup
2021-01-11T22:09:18+01:00: TASK ERROR: backup ended but finished flag is not set.

PVE2

Code:
proxmox-backup-client backup root.pxar:/ --repository 'Backup_user@pbs!Backup_Agent@172.20.5.253:Backup'
Starting backup: host/pve2/2021-01-11T21:09:18Z
Client name: pve2
Starting backup protocol: Mon Jan 11 22:09:18 2021
Downloading previous manifest (Mon Jan 11 00:32:02 2021)
Upload directory '/' to 'Backup_user@pbs!Backup_Agent@172.20.5.253:8007:Backup' as root.pxar.didx
skipping mount point: "dev"
skipping mount point: "etc/pve"
skipping mount point: "proc"
skipping mount point: "rpool"
skipping mount point: "run"
skipping mount point: "sys"
skipping mount point: "var/lib/lxcfs"
skipping mount point: "var/log"
skipping mount point: "zfs_data"
root.pxar: had to upload 78.48 MiB of 3.11 GiB in 92.14s, average speed 872.17 KiB/s).
root.pxar: backup was done incrementally, reused 3.04 GiB (97.5%)
Uploaded backup catalog (1.49 MiB)
Duration: 92.27s
End Time: Mon Jan 11 22:10:50 2021

Does anyone have any idea to solve on this problem?
 
Error: error at "usr/lib/x86_64-linux-gnu/libdcerpc-binding.so.0.0.1": Input/output error (os error 5)
there is a problem accessing that file, os error 5 means input/output error which most often indicates a broken disk/filesystem
is there anything in the syslog/dmesg when you try to backup?
 
I hadn't seen anything in dmesg or syslog, that's why I redirected to the community.

I hadn't thought of a simple test

Code:
cp /usr/lib/x86_64-linux-gnu/libdcerpc-binding.so.0.0.1 / root / 

cp: error reading '/usr/lib/x86_64-linux-gnu/libdcerpc-binding.so.0.0.1': Input / output error

I just have a zfs raid which is in error and waiting for a disk replacement.

I hope that it is not too badly in point and that it will be necessary to redo a pve installation

I will try to see to replace the disk with a temporary one and see if it works fine