[SOLVED] Filesystem error

[maxprox]

Hello,
alter the Upgrade from Proxmox 4 to Version 5 I get the following filesystem error:

 EXT4-fs error (device dm-1) : ext4_find_entry:1431: inode .....
EXT4-fs (dm-1) previous I/O error to superblock deteced
Buffer I/O error on dev dm-1, logical block 0, lost sync page write
.....
and so on

The Proxmox System freezes after this error and I have to turn off the server by the powerswitch

With parted I get:

root@ouchprox:~# parted -l
Model: ATA HGST HUS726020AL (scsi)
Disk /dev/sda: 2000GB
Sector size (logical/physical): 4096B/4096B
Partition Table: gpt
Disk Flags:

Number  Start   End     Size    File system  Name                  Flags
 1      8389kB  2000GB  2000GB               zfs-e3e40a8b10301513
 9      2000GB  2000GB  67.1MB


.........
!!!### this ERRORs are new after the upgrade ###!!! :

Error: /dev/mapper/pve-data: unrecognised disk label
Model: Linux device-mapper (thin-pool) (dm)                            
Disk /dev/mapper/pve-data: 75.7GB
Sector size (logical/physical): 512B/512B
Partition Table: unknown
Disk Flags:

Error: /dev/mapper/pve-data_tdata: unrecognised disk label
Model: Linux device-mapper (linear) (dm)                               
Disk /dev/mapper/pve-data_tdata: 75.7GB
Sector size (logical/physical): 512B/512B
Partition Table: unknown
Disk Flags:

Error: /dev/mapper/pve-data_tmeta: unrecognised disk label
Model: Linux device-mapper (linear) (dm)                               
Disk /dev/mapper/pve-data_tmeta: 75.5MB
Sector size (logical/physical): 512B/512B
Partition Table: unknown
Disk Flags:
.......
and so on

Herr is the Message befor the Upgrade with version 4:

Model: Linux device-mapper (thin-pool) (dm)
Disk /dev/mapper/pve-data: 75.7GB
Sector size (logical/physical): 512B/512B
...

this is the Message after the Upgrade with version 5:

Error: /dev/mapper/pve-data: unrecognised disk label
Model: Linux device-mapper (thin-pool) (dm)  
...

The Proxmox System is installed on one single NVMe SSD (for example: "
/dev/mapper/pve-data" this is on this SSD)
and four 4kn 4TB SATA disk drives with ZFS RAID 10 for the VMs and data pool.
As I see there is a LVM-thin on the single SSD, in my opinion that is not correct.
Should I make a new installation of Proxmox 5?

regads,
maxprox

 

[maxprox]

After some research, I think that the parted error and file system error are independent.

parted is a partitioning tool - it should be used on the underlying disks NOT on the LVM Volume Group VG or logical volume (LV) mapped devices. That is to say typically you're running parted against things like /dev/sda, /dev/sdb etc... NOT /dev/mappper/... items. From your df it appears that /dev/mapper/pve-data is one of your LVs (presumably the LV is "data" and the VG is "pve".

The actual major error is the I / O errors and therefore I now exchange the system disk after cloning by Clonezilla