vm migration with big hdd

[informant]

hi all, i have a question. i do migrate a vm with 1TB hdd qcow2 to a other host over eth, at moment the new qcow2 file is over 1.7tb and was bigger every minute, but original file is only 1 Tb and in configuration of vm hdd size is 1024 gb. why hdd is on new host very big in migrating time? do this shrink automatical after migration? regards

 

[mira]

You have to run a `TRIM` inside the guest after the migration. Also your disks need to have `Discard` enabled.

 

[informant]

hi ok it is, a 2nd question pls, why vmdk i cant migrate to other hot?

 

[mira]

Please provide the VM config qm config <VMID> and the storage config cat /etc/pve/storage.cfg.

 

[informant]

here the confis:

cat /etc/pve/storage.cfg
dir: local
        path /var/lib/vz
        maxfiles 0
        content rootdir,iso,images,vztmpl

nfs: SLS-001
        path /mnt/pve/SLS-001
        server 10.11.12.50
        export /volume1/storage
        options vers=3
        maxfiles 3
        content iso,backup,vztmpl

qm config 5532
bootdisk: virtio0
cores: 4
ide2: none,media=cdrom
memory: 10240
name: mail.myhostname.com
net0: e1000=C2:D1:EE:23:87:C2,bridge=vmbr0
onboot: 1
ostype: other
scsihw: virtio-scsi-pci
smbios1: uuid=2fcc5ad8-d70f-43f3-91bc-a7c91b969a1c
sockets: 1
virtio0: local:5532/vm-5532-disk-3.qcow2,format=qcow2,size=150G
virtio1: local:5532/vm-5532-disk-4.vmdk,format=vmdk,backup=no,size=1024G

 

[informant]

@mira or others, any ideas?
regards

 

[mira]

Ah, sorry for the late response.

Could you provide the complete migration task log?

 

[informant]

hi here is the log...
https://www.file-upload.net/download-14686496/log-failed.txt.html
online well to big for attache...

 

[informant]

@mira any ideas? regards

 

[oguz]

2021-08-31 11:56:14 starting migration of VM 5532 to node 'prometheus' (10.11.12.3)
2021-08-31 11:56:15 found local disk 'local:5532/vm-5532-disk-1.raw' (via storage)
2021-08-31 11:56:15 found local disk 'local:5532/vm-5532-disk-2.qcow2' (via storage)
2021-08-31 11:56:15 found local disk 'local:5532/vm-5532-disk-3.qcow2' (in current VM config)
2021-08-31 11:56:15 found local disk 'local:5532/vm-5532-disk-4.vmdk' (in current VM config)
2021-08-31 11:56:15 copying local disk images
2021-08-31 11:56:17 0+0 records in
...
drive-virtio1: transferred 1024.0 GiB of 1.0 TiB (100.00%) in 7h 42m 15s
drive-virtio1: transferred 1.0 TiB of 1.0 TiB (100.00%) in 7h 42m 16s, ready
all 'mirror' jobs are ready
2021-09-01 02:06:27 starting online/live migration on unix:/run/qemu-server/5532.migrate
2021-09-01 02:06:27 set migration capabilities
2021-09-01 02:06:27 migration downtime limit: 100 ms
2021-09-01 02:06:27 migration cachesize: 1.0 GiB
2021-09-01 02:06:27 set migration parameters
2021-09-01 02:06:27 start migrate command to unix:/run/qemu-server/5532.migrate
2021-09-01 02:06:27 migrate uri => unix:/run/qemu-server/5532.migrate failed: VM 5532 qmp command 'migrate' failed - The vmdk format used by node '#block353' does not support live migration
2021-09-01 02:06:28 ERROR: online migrate failure - VM 5532 qmp command 'migrate' failed - The vmdk format used by node '#block353' does not support live migration
2021-09-01 02:06:28 aborting phase 2 - cleanup resources
2021-09-01 02:06:28 migrate_cancel
drive-virtio1: Cancelling block job

vmdk files aren't supported.
remove the vmdk file from config, or convert it to qcow and edit the configuration file afterwards.

you can convert with something like this:

qemu-img convert -f vmdk /path/to/your/disk.vmdk -O qcow2 file.qcow2

 

[informant]

ok vmdk cant also migrate. ok but all qcow files with big size i have problems on backup. if hdd in qcow2 is only 30 gb and qcow2 hdd is 1tb i have on backup over 600 gb backup file.

 

[informant]

@oguz i have convert on node and change in conf of vm. but on vm-start i get error:

kvm: -drive file=/var/lib/vz/images/5532/vm-5532-disk-4.qcow2,if=none,id=drive-virtio1,format=qcow2,cache=none,aio=native,detect-zeroes=on: Unsupported qcow2 version 1 TASK ERROR: start failed: QEMU exited with code 1

how can i solve it? regards

 

[oguz]

Unsupported qcow2 version 1

ah, sorry about that! the option after the -O should have been qcow2, i've edited my post now

 

[informant]

hi i have show it and have make re convert with
qemu-img convert -f qcow /var/lib/vz/images/5532/vm-5532-disk-4.qcow2 -O qcow2 vm-5532-disk-4-n.qcow2and move to org filename. than its works.
regards^^

 

[oguz]

great!

does it migrate now?

if the issue is solved please mark the thread [SOLVED]

 

[informant]

hi migrate is running at moment, can you tell me a solution for this in this threade her pls.
https://forum.proxmox.com/threads/vm-migration-with-big-hdd.95421/post-414126

 

[oguz]

hi migrate is running at moment, can you tell me a solution for this in this threade her pls.
https://forum.proxmox.com/threads/vm-migration-with-big-hdd.95421/post-414126

what do you mean? that's the same thread where we are

 

[informant]

hi @oguz, i mean qcow2 trim dont work on backup. hdd is 13% full but backup is more as 60% of hdd...

 

[informant]

@oguz, or other users any ideas?
regards

 

[informant]

hi all, a question to this thread, if i make a migration of hdd, they is very big, but if i make a complete backup and retore it, hdd was smaller as before and have only size of my data.
how can i make this size/cleaning with a command before migrating?
regards