New Import Wizard Available for Migrating VMware ESXi Based Virtual Machines

[suckmyhardware]

This comes a little too late as I've already imported (except for 3 vms) with the usual method. Nonetheless, this is super nice! Glad I've made the switch from vmWare to Proxmox!

 

[Darkk]

That has been my approach in recent years

Then tell me what trick you used to teach Windows to use the driver for booting. I have already set the bootable flag in the registry, but it is gone again after a reboot.

You have to create a small disk and then use that to load those drivers. It's all explained in

https://support.tuxis.nl/en/knowledgebase/article/convert-windows-boot-disk-from-virtio-to-scsi

Once Windows loads the virtio-scsi drivers it will get loaded whenever it encounters those disks.

 

[Darkk]

This comes a little too late as I've already imported (except for 3 vms) with the usual method. Nonetheless, this is super nice! Glad I've made the switch from vmWare to Proxmox!

Ya, I've used the CLI method a year ago to move several VMs over from vmware to ProxMox. All went through without issues. I was using vSAN at the time so CLI made it easy to export the vms out.

 

[Falk R.]

You have to create a small disk and then use that to load those drivers. It's all explained in

https://support.tuxis.nl/en/knowledgebase/article/convert-windows-boot-disk-from-virtio-to-scsi

Once Windows loads the virtio-scsi drivers it will get loaded whenever it encounters those disks.

This is the way I have done it hundreds of times. But I would like to get by without the extra reboot and controller replacement.

Nutanix has a way to do this, they inject the driver so that you can boot directly from it. Unfortunately they don't give any information on how they do this.

 

[Darkk]

This is the way I have done it hundreds of times. But I would like to get by without the extra reboot and controller replacement.

Nutanix has a way to do this, they inject the driver so that you can boot directly from it. Unfortunately they don't give any information on how they do this.

Sounds like they use a PowerShell script to make the necessary changes before the VM gets imported over. There might be open source scripts out there that does the same thing.

 

[Falk R.]

Sounds like they use a PowerShell script to make the necessary changes before the VM gets imported over. There might be open source scripts out there that does the same thing.

have not yet found a script that really works. Do you have a link?

 

[Gilberto Ferreira]

Finally it's works fine with live-import.
In my case, it took 22 minutes to import a 10G Debian 12, with nothing but the OS!
Awesome.

 

[Darkk]

have not yet found a script that really works. Do you have a link?

No but if I run across one I will let you know.

 

[lee.fishlock]

Also, once I get the error it never seems to come back. Clicking on the storage gets me (vim.fault.HostConnectFault) { (500)

I'm experiencing the same error during live migration and at least 50% of the time I lose the ability to connect to ESXi storage from PVE until I reboot the ESX host. It also seemed to brick my mgmt access to vsphere for the host on one attempt, but not sure if that was directly related.

 

[ofelbar]

Tried again today.
Only one proxmox pve connected to ESXi
Waited during the night, so no traffic towards ESXi.

I get 24% in on a 64 GB disk.

Mar 28 09:22:29 pve03 pvedaemon[283672]: <root@pam> starting task UPIDve03:0008B13A:00486EC0:660528C5:qmcreate:103:root@pam:
Mar 28 09:29:08 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: rate limited, retrying (1 of 5)...
Mar 28 09:29:18 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: rate limited, retrying (2 of 5)...
Mar 28 09:29:28 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: rate limited, retrying (3 of 5)...
Mar 28 09:29:38 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: rate limited, retrying (4 of 5)...
Mar 28 09:29:48 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: rate limited => Response { status: 503, version: HTTP/1.1, headers: {"date": "Thu, 28 Mar 2024 08:29:48 GMT", "connection": "close", "content-security-policy": "block-all-mixed-content", "content-type": "text/plain; charset=utf-8", "strict-transport-security": "max-age=31536000", "x-content-type-options": "nosniff", "x-frame-options": "DENY", "x-xss-protection": "1", "content-length": "0"}, body: Body(Empty) }
Mar 28 09:29:48 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: error handling request: cached read failed: rate limited

After that I get ratelimited out of the local mgmt also, I cant login to ESXi
Login screen gets:
HTTP/1.1 503 Service Unavailable
Date: Thu, 28 Mar 2024 08:33:22 GMT
Cache-Control: no-cache
Connection: close
Content-Type: text/plain; charset=utf-8
X-Frame-Options: DENY
Content-Length: 0


Dell R620 / VMWare ESXi 6.7.0 U2, Enterprise plus license, No vCenter.

Anything I can try? I really want this to work since I have 20 VMs left to convert.

 

[d4gli]

Hm, running on PVE 8.1.4 and installed the pve-esxi-import-tools package - restarted the interface (and later even the test node), I do not see any 'esxi' import option under storage?! what might be missing?

 

[ofelbar]

Hm, running on PVE 8.1.4 and installed the pve-esxi-import-tools package - restarted the interface (and later even the test node), I do not see any 'esxi' import option under storage?! what might be missing?

Q: How can I get the new import wizard?
A: Update your Proxmox VE to pve-manager version 8.1-8, libpve-storage-perl version 8.1.3 or newer, which should pull in the new pve-esxi-import-tools package.

 

[buco]

I'm experiencing the same error during live migration and at least 50% of the time I lose the ability to connect to ESXi storage from PVE until I reboot the ESX host. It also seemed to brick my mgmt access to vsphere for the host on one attempt, but not sure if that was directly related.

Also "time" seems to fix this. I waited for a couple of hours and I no longer get the "error 500". I can get to the migration wizard already but migration still fails.

 

[Stoiko Ivanov]

Hm, running on PVE 8.1.4 and installed the pve-esxi-import-tools package - restarted the interface (and later even the test node), I do not see any 'esxi' import option under storage?! what might be missing?

You add a new storage of type 'ESXi' under Datacenter->Storage.

I hope this helps!

 

[d4gli]

Q: How can I get the new import wizard?
A: Update your Proxmox VE to pve-manager version 8.1-8, libpve-storage-perl version 8.1.3 or newer, which should pull in the new pve-esxi-import-tools package.

I did... hm, nvm maybe caching. Now, I see it

 

[Falk R.]

Tried again today.
Only one proxmox pve connected to ESXi
Waited during the night, so no traffic towards ESXi.

I get 24% in on a 64 GB disk.

Mar 28 09:22:29 pve03 pvedaemon[283672]: <root@pam> starting task UPIDve03:0008B13A:00486EC0:660528C5:qmcreate:103:root@pam:
Mar 28 09:29:08 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: rate limited, retrying (1 of 5)...
Mar 28 09:29:18 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: rate limited, retrying (2 of 5)...
Mar 28 09:29:28 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: rate limited, retrying (3 of 5)...
Mar 28 09:29:38 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: rate limited, retrying (4 of 5)...
Mar 28 09:29:48 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: rate limited => Response { status: 503, version: HTTP/1.1, headers: {"date": "Thu, 28 Mar 2024 08:29:48 GMT", "connection": "close", "content-security-policy": "block-all-mixed-content", "content-type": "text/plain; charset=utf-8", "strict-transport-security": "max-age=31536000", "x-content-type-options": "nosniff", "x-frame-options": "DENY", "x-xss-protection": "1", "content-length": "0"}, body: Body(Empty) }
Mar 28 09:29:48 pve03 esxi-folder-fus[63523]: pve03 esxi-folder-fuse[63523]: error handling request: cached read failed: rate limited

After that I get ratelimited out of the local mgmt also, I cant login to ESXi
Login screen gets:
HTTP/1.1 503 Service Unavailable
Date: Thu, 28 Mar 2024 08:33:22 GMT
Cache-Control: no-cache
Connection: close
Content-Type: text/plain; charset=utf-8
X-Frame-Options: DENY
Content-Length: 0


Dell R620 / VMWare ESXi 6.7.0 U2, Enterprise plus license, No vCenter.

Anything I can try? I really want this to work since I have 20 VMs left to convert.

This is not a Licensing issue. I think this is a Issue from your very old ESXi version.
I Know, many years before, ESXi 6.7 had any Performance issues on the Management Interface (very slow Backups).
You can try, add a new vmkernel Interface and let PVE connect to this different Interface.

 

[ofelbar]

This is not a Licensing issue. I think this is a Issue from your very old ESXi version.
I Know, many years before, ESXi 6.7 had any Performance issues on the Management Interface (very slow Backups).
You can try, add a new vmkernel Interface and let PVE connect to this different Interface.

Tried on another vmk (vmk2 in another vSwitch and physical NIC, instead of vmk1). Got to 16% before rate-limited this time.

I can try to spin up a vCenter and see if that helps.

 

[stv.Bundesadmin]

Tried a few very small Test-Imports and a live import.

I'm impressed, very nice work!
We'll wait a litte more until some of the "bugs" are worked out, e.g. a proper way to handle the rate limiting on Proxmox side?


One Problem we encountered:
Live-Importing a VM that was still running on the VMware side led to an error (expectedly).
I think such cases could/should be handled by the Proxmox workflow - check for it and display a "can't continue until you do $x"
Because it doesn't check that, the import will fail immediately but it leaves a "locked" VM on the Proxmox side behind.
-> while Import works in the GUI, you now have to switch to the cli to "qm unlock xxx" in order to be able to delete this failed VM.
Maybe clean up failed imports automatically or present a user choice on "clean up or leave behind"?

And please please please provide more GUI-actions for "VM unlock" or "HA Error Reset" etc.

 

[juliano.morona]

Note that you can live-import VMs. This means that you can stop the VM on the ESXi source and then immediately start it on the Proxmox VE target with the disk data required for booting then being fetched on demand. The feature works similarly to live-restore that we provide for PBS backups since a while.

Live-import naturally requires a very stable and fast network between the PVE and ESXi host and taking some care with device driver/model selection, but it can reduce the downtime on import significantly.

This is fantastic, it will boost the adoption of Proxmox to replace VMWare.

 

[t.lamprecht]

I think such cases could/should be handled by the Proxmox workflow - check for it and display a "can't continue until you do $x"

FWIW, we show a warning in the UI if the source VM is still running, but yeah re-checking on the API site if it's still running might be dooable and provide slightly better UX.