ESXi 4.1 VM guests not starting after QNAP update

[nasman77]

Hi,

after I updated our TS-259Pro+ to 3.4.4 (from 3.3.4), the Windows guests are not starting anymore.
After Windows boot splash boot the screen gets black again. Some guests can be pinged, but they're not accessible via RDP for example.
Already downgraded back to 3.3.4 and did several reboots of ESXi as well as of the NAS without any success.

Logfile shows for guests with SCSI drives:

Code: Select all

Aug 04 15:29:56.518: vcpu-0| DISKLIB-DDB   : "longContentID" = "022d9af3aecf55cb26ffdbcb0d68d81e" (was "7dcb83966229138e18702393eb90888a")
Aug 04 15:29:56.519: vcpu-0| DISKLIB-CHAIN : DiskChainUpdateContentID: old=0xeb90888a, new=0xd68d81e (022d9af3aecf55cb26ffdbcb0d68d81e)

Logfile shows for guests with IDE drives:

Code: Select all

Aug 04 14:47:49.488: vcpu-0| VIDE: (0x1f0) OUTB Cmd 0xf5, Unknown ATA Command on drive 0
Aug 04 14:47:49.494: vcpu-0| VIDE: (0x1f0) OUTB Cmd 0xf5, Unknown ATA Command on drive 1
Aug 04 14:47:49.499: vcpu-0| VIDE: (0x170) OUTB Cmd 0xf5, Unknown ATA Command on drive 0

I found a similar problem here:
http://communities.vmware.com/message/1724836

Using a BartPE or linux Boot-CD, I can access the data on the virtual disks, so there must be any problem with change disk UID or something like that...

I'm planning to setup two 659Pro+ for another location, but first, I need to get that problem fixed...

[banders5144]

I want to say that I just recently had this problem as well. I upgraded to the latest version of the QNAP firmware and it caused some of our VMWare Machines that were hosted on the QNAP to fail. After i reverted back everything started working again.

[nasman77]

Hi,

I'm not alone!
Only that bad thing is, that the downgrade didn't change anything.
I tried several things yesterday to get the guests run again, all of them failed:

- Enabled windows bootlogging on the guests - got some logs, but they don't show anything critical
- enabled and disabled drivers via repair console - no success
- did fixboot and fixmbr - no success
- created new disks (same and different in size, IDE and SCSI) and transferred the original system drive to them (Acronis, DriveSnapshot) - no success

Nevertheless installing a new guest is no problem at all, and a guest which is running on local storage on the ESXi host shows no problems too, so there must be a QNAP related problem...

[nasman77]

Has anyone @QNAP seen this problem before? There's no chance to get these VMs back running again, so I had to rebuild them from backups.

[banders5144]

nasman77

After downgrading, I was able to fix mine by having a backup copy of my VMDK on another share on the QNAP. I just overwrote it to the share where i actually run the VM's

[nasman77]

The VMs were not so important, that I needed a backup - but is there a general problem to upgrade to 3.4? Or did you manage to do the upgrade lateron?

[banders5144]

I did not actually have a problem upgrading to it. It didn't hang like other people have described. The only thing that stopped working was the NFS part of it for the VM's we have running from it.

[nasman77]

Ok, all other services except NFS were running after my upgrade to 3.4.4 too. But what's your current configuration now? Did you repeat the upgrade and copied your VMs back to the NFS share or did you stay at your old version?