VMTN

Not to my knowledge.

Does anyone know what the switch is for AMD'S "V" (its Intels VT equivalent)

For Intels Dual Core VT CPU'S Use : monitor_control.vt32 = "TRUE"

For AMD-V CPU'S Use: ?????

I've got ESX3 running as a guest on Workstation 6 - but it runs like a dog...

For AMD-V CPUs, use:

monitor_control.enable_svm = TRUE[/code]

This control flag, like vt32, is entirely unsupported.

Hello Guys,

Been reading this with great interest, this is so great if I would get it to work 

ESX installs like a cham, booting goes ok (no fails) but I cannot cannot with the VI3 client to it.

Pinging (on it's IP Address)  the ESX host is possible for the PC that has the VI3 client.

Created the Guest as following:

\- 2 VCPU's

\- 3 NICS (bridged)

\- 1 SCSI HDD 10G

Everything starts without errors, checked network config with esxcfg util.

All seems ok.

Can anyone help me how to set up networking for this Virtual ESX host.

Thanks.

Message was edited by:

        Marc_Dieters

Hey, already found it.

I added ethernet0.virtualDev = “e1000″ but left ethernet0.virtualDev = “vlance" in the vmx file.

Removing this did it, I can now connect to my V-ESX Server via VI3.

Thanks

Following the information in this thread I have managed to boot ESX without any errors.

My problem right now is that when I put the IP address for ESX in to a browser it doesn't connect.

ESX has a static IP on my network, so is there some further configuration that I need to do?

The option finally leaks!  Yes, folks, AMD's SVM works for virtual machines too in WS6.  The one setting applies to both 32 and 64 bit.  There are slight negative performance implications of SVM on RevF hardware (binary translation on AMD processors is very good); it's only useful for vm-in-a-vm.

What jmattson didn't mention is that the option is not sticking around.  VMware does not "support" first-generation (rev F, "Pacifica") SVM, the option is already removed from the next-generation codebase (vm-in-a-vm has never been a supported mode, it just happens to work) in favor of second-generation SVM hardware (that is, NPT).

Nice doc that was just released on this...

Guide to installing ESX 3 on Workstation 6 - http://www.xtravirt.com/index.php?option=com_remository&Itemid=75&func=fileinfo&id=9

http://www.vmware.com/community/thread.jspa?threadID=86877&tstart=0

Thanks very much! Now I have web access to ESX.

Just FYI tested ESX 3.0.1 running in Fusion Beta 3 on iMac. Runs great!

Now I just need to order my Sager (Clevo) NP9260

Intel Core 2 Extreme X6800 (Quad-Core) 4MB Cache, 2.93Ghz

4GB Memory and  3 SATA 2.5" Disks

http://www.sagernotebook.com/pages/go_np9260.html

Ouch...

Yeah, unfortunately, I've got a NP9860... very sweet machine, but no VT support. 

I've got the NP9750 (Athlon 64 X2 4800) and I don't have the AMD-V (no svm flag in /proc/cpuinfo) ... just hope they get a new AM2 notebook (more like a luggable) out soon...

ksc wrote:

The option finally leaks!  Yes, folks, AMD's SVM works for virtual machines too in WS6. 

The one setting applies to both 32 and 64 bit. There are slight negative performance

implications of SVM on RevF hardware (binary translation on AMD processors is very

good); it's only useful for vm-in-a-vm.

I was curious as to the effect of the monitor_mode.enable_svm in a new

virtual machine. So I installed a new Windows Server 2003 with the flag,

and there is indeed no visible performance.

I wonder if it would speed up MS Virtual Server 2005 VMs just kidding.

What jmattson didn't mention is that the option is not sticking around.  VMware does

not "support" first-generation (rev F, "Pacifica") SVM,

very interesting comment. Is the current first-generation SVM not acceptable

or is the first Pacifica implementation (Opteron Rev F) not a finished product.

I know these are questions you probably won't be able to answer

... the option is already removed from the next-generation codebase (vm-in-a-vm

has never been a supported mode, it just happens to work) in favor of

second-generation SVM hardware (that is, NPT).

Does this mean, that we have to keep presciously our current Workstation 6.0

Build 45731 and not upgrade to another build ?

As I have mentioned in my earlier posts I have been

trying to get ESX to work as Wkst guest on single

core P4 processor PC and despite trying various

recomended vmx file settings but it it fails when

loading the vmkernel during the boot process.

 

I think it can only[/b] work sucessfully if the

Windows PC running the workstation 6 guest is one of

the new dual core processor types. And also the guest

must be configured with 2 CPUs.

 

I get the below error during the startup.

 

VMware VMkernel authorisation daemon, Warning:

date....error:md5 signature mismatch. Running

vmkernel built on -Unknown- (vs.vmkload_app built on

Sept 26 2006 00:00:01)

 

When logon to the console I see an error:

 

Service Console running in fallback mode.

Mayur, I'd be interested in hearing if you got any further with this. I face the same error on startup. Reading through the posts here it's hard to figure whether it's related to the PC hardware (Intel T5200 CPU which doesn't support VT), WS version , vmx contents or something else entirely.