Repair VM: systemId does not match

[Joshua Kaplan_1]

I have an HP DL380 running ESXi 4.1, with iSCSI storage provided from an HP MSA 2312i. I recently needed to reload ESXi, and then reattach the iSCSI storage by configuring both the iSCSI adapter in ESXi and the MSA. The storage returned intact, with the folders and files of two VMs I had on them. What I didn't realize at the time was that the VMs actually are split between two folders. One folder lives on local SAS storage, and contains the VMX, VMDK, NVRAM, VMXF and log files. The other folder, which lives on the iSCSI storage, contains *_1.vmdk, *_2.vmdk, etc. What makes that significant is that each time the MSA iSCSI detects a host, it assigns a new ID string to it, such as iqn.1998-01.com.vmware:-4355c0ba.


After configuring both the MSA and the host, and adding the VMs back into the inventory by right clicking the vmx file and clicking 'add to inventory', I attempted to start the VM. I received the following error message:


The systemId does not match the current system or the deviceId, and the vendorId does not match the device currently at 000:29.7


It seems to me that there is a reference (or possibly references) to the previous string by which the iSCSI storage was known, and that is making ESXi unhappy. I could be completely off about that, but it's my best guess.


The question is how to resolve this error, so that I can restart the VM.


Any help is greatly appreciated, but effective help is particularly appreciated.

[Joshua Kaplan_1]

Solved my own problem, which is sometimes the best way to do it. It has been a long road, but I learned a lot along the way.



The reference to device 000:29.7 turns out to be a PCI device. In ESX, there is a list in /proc/vmware/pci of discovered devices, but that turns out not to be true in ESXi 4.1



There is, however, a command lspci, which provides exactly that information. Unfortunately, it goes by much too fast, and the screen capture facility in HP iLo 2 doesn't work very well. Fortunately, although it seems that in previous versions of ESXi the more command was not available, it has been restored in 4.1



The device in question turns out to be the host server's USB controller. I had stumbled across a writeup somewhere in all of this about boot issues related to not being able to remove pass-thru USB devices, and then it all clicked.



I went back to the failing VM, and checked the config. Sure enough, the USB port was included as a PCI device, but the host was not configured for USB pass-thru. I enabled that and rebooted, but the VM still failed to boot. I then removed the PCI device from the VM config, and up it came.