Update Manager Hangs Guest OS running VC and its VM

[shurik]

When running Virtual Center in VM, running update reconsiliation against such guest OS causes it to hang. Subsequent attempt to power off its VM gets stuck at 95% (I used vi client power it off)

I had to shutdown all the other guests OSs and restart entire host.

the problem is easily reproduceable. I suspect it happens because update manager attempts rebooting VC's OS

[wharlie]

I get the same results.

My specs are VM running W2K3 VM with VC & license server and update manager server, 1CPU 1GBram.

The database is SQL2K on a differnet VM.

After I initiate a scan it goes OK for a while until it connects to the CDROM and then it hangs.

Can't ping, can't power off (even from console)

I left it overnight and it had returned in the morning but it still hadn't completed the scan (ie patches still unknown).

Currently running as test but have concerns about using in prod.

[MR-T]

I hadn't considered with, but it's a very good point.

I always deploy VirtualCenter as a VM, so how does patch management keep control when the VM is trying to shutdown.

What would happen if you installed patch manager on a seperate VM, do you think it would still hang when the VC went down.

I would suggest the VM is moved to a seperate folder without a baseline attached and patching of this VM is performed manually.

One thing I will say is that I've deployed patch manager into a few companies now and each of them only want to use it for host patching. They already have WSUS in production and are happy to continue patching Windows with this. I know VUM allows the patching of offline vm's, but for now it's not winning much ground.

[wharlie]

In my case I am only doing a scan, not applying any patches (remediate), so it shouldn't be shutting down the VM.

That being said I will follow your advise and patch manually, but it woould be nice to be able to scan the VM to find out what patches it needs.

Either way, VMware needs to fix or disable patching VC so that people don't end up having to reboot hosts that may be running multiple vm's.

[fzimmerb]

Hi Folks,

I do also have the same problem at a customers site when scanning the Win2k3 VM running VC amd VUM. Scanning runs to 65% and the CD is mounted but then the system hangs completely. The only way to "turn it off" is to kill the process...

The database is on a seperate Server and we do use seperate ODBC connections.

Versions are: ESX 3.5 Update 4 (153875) , vCenter 2.5 Update 4 (147633), VI Client 2.5 (147704), VUM 1.0 Update 4 (1.0.0.63995), Win2k3 R2 SP2 - fully manually patched

Does anybody have an idea about that issue?

[Texiwill]

Hello,

Moved to Update Manager forum.

Best regards,
Edward L. Haletky
VMware Communities User Moderator, VMware vExpert 2009
====
Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education.
Blue Gears and SearchVMware Pro Blogs -- Top Virtualization Security Links -- Virtualization Security Round Table Podcast

[Slatfatf]

About (same same), when running remediation or an update

scan against the VM hosting Update Manager it hangs the guest OS, at times to

the point where ESX requires a re-boot to restore. The VC & Oracle database

are homed to physical servers. Do we know anything about this yet? Anyone have

issues when running UM on physical platform? Is anyone running UM & WSUS on

same server, and if so any issues?

UM 1.0.0.63995 (UM Win32 Client 1.0u4)

VIC 2.5.0 b119598

VC 2.5.0 b147633

Oracle 10g v10.2.0.3.0

ESX 3.5.0 b153875

[petkom]

This seems like recursion - you are trying to scan the machine itself.

Why don't you create two mirror VMs with your VC and VUM. From the 1st you can scan the 2nd and vice versa. Just before the scan go to the scanned VM and stop VC and VUM services.