[SOLVED] ProLiant MicroServer Gen10 High CPU IO Delay Version 7.4.3 restore takes verry long.

[Badkey]

HPE ProLiant MicroServer Gen10 Plus (P16005-421) 32Gb RAM and 3TB Drive.

Start testing before in production. No ZFS. Just installed Proxmox 7.4 default and fresh. Created a VM did a backup and deleted the VM.
No compression. Then when I do a restore there is a more then 50% IO CPU delay.

Then after restore the VM (32Gb) took 30 Minutes !
Loginfo:
restore vma archive: vma extract -v -r /var/tmp/vzdumptmp37012.fifo /var/lib/vz/dump/vzdump-qemu-100-2023_05_17-21_34_22.vma /var/tmp/vzdumptmp37012
CFG: size: 491 name: qemu-server.conf
DEV: dev_id=1 size: 34359738368 devname: drive-scsi0
CTIME: Wed May 17 21:34:23 2023
Logical volume "vm-100-disk-0" created.
new volume ID is 'local-lvm:vm-100-disk-0'
map 'drive-scsi0' to '/dev/pve/vm-100-disk-0' (write zeros = 0)
progress 1% (read 343605248 bytes, duration 1 sec) etc .........
progress 100% (read 34359738368 bytes, duration 641 sec)
total bytes read 34359738368, sparse bytes 26144788480 (76.1%)
space reduction due to 4K zero blocks 0.376%
rescan volumes...
TASK OK

What can be wrong ? Any BIOS Setting of the Proliant itself ?

Regards John

 

[onenerdyguy]

do you have Virt extensions turned on?

 

[Badkey]

Yes all is in Vrt modes.

 

[ITT]

High I/O = slow Disks!

 

[_gabriel]

1) if backup is restored on same disk, i/o delay is expected.
2) iirc, hp smart array s100i disable disk writecache in raid mode. even in non raid disk writecache can be disabled by default, Enable it to get regular performance.

 

[Badkey]

I used PBS on an other server all working ok now.