Offline migrated VM doesn't boot completely

[J-D]

Hi,

We have an ESX server which is not connected to a SAN and 3 other ESX'es which are and are member of a HA-DRS cluster. I wanted to move a VM from that separate ESX to the ESX cluster using Virtual Center's migrate.

VMotion of course didn't work because the VM is not on the SAN (it's 3.0.X). I shutted the VM, used migrate and let it move the files to the SAN. There I try to boot up the VM but after I see the Windows splash screen (the Windows logo on the black background) the VM freezes. Choosing power off just hangs the command at 95%. I have to kill the vmkload_app process to have it shut.

The seperate ESX runs 3.0.1. and the others in the cluster either 3.0.1 or 3.0.2u1. None of the ESX's in the cluster can boot up the VM completely. When I migrate the VM back to my original VM, the VM boots up nicely!

I now cloned the VM to the SAN and I've put the VM on some internal-only vswitch. Every time I try to boot this VM this one just freezes and in the console all I see is a black screen. I am also pretty sure this VM original got migrated from the SAN to this ESX...

Why oh why is this VM not booting? Is it CPU hardware difference? In the vmware.log I see:

00) and 0xf4000000(0xf4000000)

Jan 22 16:29:49.172: vcpu-0| SVGA: Registering IOSpace at 0x1060 (0x0)

Jan 22 16:29:49.172: vcpu-0| SVGA: Registering MemSpace at 0xf8000000(0xf8000000) and 0xf4000000(0xf4000000)

Jan 22 16:29:49.685: vcpu-0| CDROM: Mode Sense for Unsupported Page 0x1B

Jan 22 16:29:51.767: mks| Ignoring update request in VGA_Expose (mode change pending).

Jan 22 16:29:58.907: vcpu-0| Guest: VMXNET: Initialization completed successfully. Version 2.0

Jan 22 16:29:58.908: vcpu-0| Guest: VMXNET: Features: zerocopy csum

Jan 22 16:29:59.015: vcpu-0| SVGA: Unregistering IOSpace at 0x1060 (0x1060)

Jan 22 16:29:59.015: vcpu-0| SVGA: Unregistering MemSpace at 0xf8000000(0xf8000000) and 0xf4000000(0xf4000000)

Jan 22 16:29:59.016: vcpu-0| SVGA: Registering IOSpace at 0x1060 (0x0)

Jan 22 16:29:59.016: vcpu-0| SVGA: Registering MemSpace at 0xf8000000(0xf8000000) and 0xf4000000(0xf4000000)

Jan 22 16:29:59.034: vcpu-0| Guest OS = 0x5016

Jan 22 16:33:23.455: mks| Ignoring update request in VGA_Expose (mode change pending).

Jan 22 16:33:48.368: mks| SOCKET 2 send error 32: Broken pipe

Jan 22 16:33:55.306: mks| SOCKET 2 destroying VNC backend on socket error: 32

Jan 22 16:33:55.306: mks| POLL device deleted

Jan 22 16:34:05.928: vmx| Caught signal 15 -- tid 101229 (eip 0x080f36b6)

What's this signal 15?

Also for some reason, one of the times the VM reached a grey window...just before the Windows 2003 login windows but it hanged there.

Anyone any thoughts?

I am just afraid that this VM won't boot if I upgrade this separate ESX to 3.5

TIA

[Texiwill]

Hello,

Did you update the virtual hardware within the VM after bringing it to VI3?

Best regards,

Edward L. Haletky

VMware Communities User Moderator

====

Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education. As well as the Virtualization Wiki at http://www.astroarch.com/wiki/index.php/Virtualization

[J-D]

Everything is already on VI3. All ESX Servers are 3.0.X with a VC running 2.0.2 Update2.

I can't update any virtual hardware in a VM that doesn't boot ... or not? How should I update that? Using VMware tools installer?

[Texiwill]

Hello,

If the VM was built for v3 then the hardware is already updated. You update the hardware before boot, right click on the VM in the VIC and there should be an update hardware option. Or open the console and go to the VM menu. Same thing. If that does not work, I think you have a different issue going on.

Best regards,

Edward L. Haletky

VMware Communities User Moderator

====

Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education. As well as the Virtualization Wiki at http://www.astroarch.com/wiki/index.php/Virtualization

[J-D]

the hardware is probably updated....I don't see the option anywhere...not even grayed out.

I wish this was the solution, it would have been easy to solve...

[J-D]

I tried to boot in safe mode and saw the boot procedure with vmscsi.sys, SCSIPORT.SYS, disk.sys, ntfs.sys,..,drvmain.sdb and then as last:

....\Windows\system32\acpitabl.dat

I assume ACPI is to blame here...

[J-D]

for anyone who cares: same issue when doing an offline V2V using the VMware converter.

[ericdaly]

Had same issue but could boot ok into safemode. After hours of tryning various troubleshooting techniques I changed the HAL to "Standard PC" in device manager. It was set at ACPI Uniprocessor. Once I changed to Standard PC in safemode and reboot all was ok.