[SOLVED] qmmove not finishing - and way too big size

Dec 8, 2023
3
0
1
Hello!

I have a problem which I'm not able to solve.
I'm trying to moving a lvm based storage (harddrive) from the local Storage to a NFS Storage.
Similar migrations worked before without problems.
The harddrive has a size of 2.2 TB.
The migration never finishes and the Percentage is always dropping down a little bit and going up again (between 99,7X and 99,8X) but never finishes.
But the strange thing is the size. It currently shows a size of 5.5 TB and a Process runtime of 4 days. I interrupted the first try because i thought i had a stale NFS share somewhen in the process but the NFS is reachable without problems regarding to the logs and the statistics in PVE. The usage of the NFS store is not increasing any more since the first "percentage drop" happened.
See some output of the task in the attached files and pve version in the txt file.
Task Status ist running.

Does anyone have an idea?

Thanks!
KR
Alex
 

Attachments

  • pve_version.txt
    1.4 KB · Views: 4
  • Bildschirmfoto vom 2023-12-08 11-27-49.png
    Bildschirmfoto vom 2023-12-08 11-27-49.png
    29.1 KB · Views: 10
  • Bildschirmfoto vom 2023-12-08 11-28-03.png
    Bildschirmfoto vom 2023-12-08 11-28-03.png
    29.6 KB · Views: 10
  • Bildschirmfoto vom 2023-12-08 11-28-12.png
    Bildschirmfoto vom 2023-12-08 11-28-12.png
    29 KB · Views: 9
  • Bildschirmfoto vom 2023-12-08 11-28-23.png
    Bildschirmfoto vom 2023-12-08 11-28-23.png
    29.7 KB · Views: 9
Did nobody ever have this behaviour?
I stopped the migration process after 1 Week and 2 Days at a size of 10TB with a disk size of 2,3 TB.
The space consumed on the target storage was this size (2,3).

Any idea would be welcome!
 
Hi,
how much IO does the VM do and how fast is your network? If the VM is dirtying the disk faster than can be copied to the target, it could be that it never finishes. The reported size is the total to be transferred, so new writes by the VM will count as well.
 
Hi!
Thanks for the help and the hints.
The problem was solved!
It was exactly like you suggested. The disk is holding a mongodb and is dirtying a lot of files on every write.
I didn't imagine that the dirtying was faster than the writing but exactly this was the problem.
The connection is a 10GBit base and it is doing normal IO and no delays.
I stopped the database and the migration finished within expected time frame.

Thanks again for your help, I'm going to mark the thread as solved.
KR
Alex
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!