GuestOS Discrepancy and the VCenter Database

[evetsreklaw]

I know nothing of VMware and have been tasked with building a web application that retrieves data from several disparate vcenter installations.  I am experiencing a bit of a data quality issue.  Within VSphere I am seeing a discrepancy with regard to which GuestOS is installed.  In one location (all VMs grid 2014-03-25_12-23-02.png)  it shows as Windows 2008 (64 bit) and another location (VM details 2014-03-25_12-22-41.png) Windows 2008 R2 (64 bit).  The latter is correct.  The VPX_VM table show winLonghorn64Guest which translates to Windows 2008 (64 bit) . The VPX_VM_CONFIG_INFO also shows this value.  Where is the correct data stored in the vcenter database?  Any assistance is appreciated.

[evetsreklaw]

I guess I should have added, is this a known bug?  Is there a way to force an update to the VPX_VM and VM_CONFIG_INFO tables so that they show the correct information?

[MKguy]

The value of your first screenshot represents what GuestOS type the VM was configured for from the virtualization-side (you can see that if you edit the VM object settings in the general tab), while the value of the 2nd screenshot is what the VMware Tools service inside the Guest OS detects dynamically at runtime. So it's not really a discrepancy, because the sources of information are different.

Don't mess with the database directly. You can change the GuestOS type of the VM to the correct OS, which involves downtime because the VM must be powered off for that change, but running a VM with Windows Server 2008 R2 x64 installed as "Windows Server 2008 x64"OS type shouldn't cause any issues.

[evetsreklaw]

Thank you for the information.  The odd thing is that these VMs never had anything but R2.  It would be nice if there was a way to automate the update of the database.  Even though there is not a performance problem, there is a bean-counter and security posture issue when reporting the wrong OS.

26 misidentified VMs in one small VCenter and who knows how many in our others.

[evetsreklaw]

If in fact what you said was true, then these numbers would not change unless a manual change was made.  This morning I checked the data and there is a major difference.  Somehow this data is being updated incorrectly.