Trying to run a P2V on a Linux (RHEL5) server for first time, can't figure out how to get it *started*.

[MitchK]

Hi - I'm trying to figure out how to do a P2V for a Linux (RHEL5) server, and everything I find for documentation seems to start with, "Now that you already have the Converter Wizard running....."

(Linux Linux Linux, not NOT Windows)

Can I buy a vowel on how/where do I start, from the beginning, please? :smileyhappy:

I'm running VMware VirtualCenter 2.5.0 (build 119598) using VMware Infrastructure Client 2.5.0 (build 119826), and everything is running fine.

I have the VMware Converter Enterprise 4.0.2 plugin installed and enabled on the VC.

I have downloaded the VMware Converter (Standalone) tarball "converter-4-146302.tar.gz", but haven't installed any of it anywhere yet.

I can't seem to find anything in the VC about the converter, beyond the fact that the plugin is installed and enabled, so I'm assuming I don't start the conversion process there.

Here's my question:

The Converter comes with a Server, a Client, and an Agent.

Do I install just the Agent on the Linux Physical box I want to convert, the Agent and the Client, all three pieces, or what?

Once I have whatever piece(s) I need installed on the Linux Physical box, do I initiate the conversion from there, or from the VC, or what?

Is there a page somewhere that describes this in a way that's useful for someone who's never done it before?

Any help, pointers, hints, suggestions, acts of mercy greatly appreciated.

Thanks.

[Troy_Clavell]

In order to convert LInux you need to download VMware vCenter Converter Standalone 4.0

Here is the users guide

http://www.vmware.com/pdf/VMware_Converter_Standalone_guide40.pdf

[MitchK]

Have Standalone Converter already. (Listed above.)

Have PDF file of User's Guide with instructions on my desktop, and printed out. (Listed above.)

Can't find where/how to start. User's Guide seems to think I already know how to do this.

Where do I start Conversion Wizard? Which pieces do I install onto Linux Physical server I'm trying to convert?

The directions in the PDF start in the middle, and switch gears/context without being clear about where they're at.

(Ie, they appear to be written with the assumption that you already know what they're talking about.)

[vmweathers]

The VC-integrated Converter is completely separate from the standalone Converter. You need to install the Converter Standalone 4.0 product in-whole somewhere. For linux p2v the Converter can be installed on either windows or linux. There is no big advantage to installing it on linux directly.

(If your question has been resolved please mark the answers as "Helpful" or "Correct".)

[Troy_Clavell]

...then launch the interface, choose convert machine, linux, then follow the prompts.

[MitchK]

Ok, so I install all three pieces on the target Linux Physical server and run the conversion, right?

Where does the resultant "helper" VM end up? Where does it go, and how does it know how to get there, if all three pieces are installed on the target machine? How do I get it over to the VCenter server I'm already running in-house?

(Thanks for the answers so far.)

Somewhat related question -- If VMware Converter 4 is totally and utterly separate from VMware Converter Enterprise, what good does the Enterprise converter plugin I already have installed do me? Seems kind of wasted/useless to me at the moment.

[vmweathers]

Since you keep saying "target ... server/machine" I'm a little confused. The linux machine you want to convert to a VM is the termed by us as the source not the target. Maybe you're thinking that the VM will be created on the linux machine -- I'm afraid not. It's created on an ESX machine, which is a requirement for linux P2V. That is because of the helper VM you mentioned.

You do not need to do anything for the helper VM, it is deployed automatically. Though if you don't have DHCP service to the target VM/helper VM then you'll need to configure a static IP for the helper VM. See the user guide for details on this. Also you'll need to ensure there is connectivity between the Converter machine, source machine, and helper VM; i.e., open holes in any intervening firewall. See the user guide for the list of ports you need to open.

Also, as I noted before, you do not need to install Converter on the same linux machine you are converting. All Linux P2Vs are done "remotely" over ssh, so the machine that Converter runs on doesn't matter.

As for the VC plugin -- yeah, I don't think it will prove very useful if you are ok with having a separate Converter product. The most useful feature you lose is probably having it directly integrated with the VI client so that you can right click on VMs and launch Converter for them.

(If your question has been resolved please mark the answers as "Helpful" or "Correct".)

[MitchK]

Ok, regrouping for a better explanation here.

I am running Virtual Center on physical box A. I am using A to control two physical boxes running ESX servers (3.5.0, build 123630) on them, boxes B and C.

I have a physical box D, running RHEL5 on it. All of these boxes are on the same subnet.

I need to turn physical box D (the "source") into a VM (the "target"?) running on ESX server B or C, and controlled by A. (Let's call the VM 'E'.)

To virtualize physical machine D into VM E, running on ESX server B or C, controlled by VC A, do I install a "Local Install" or a "Client-Server Install"? The documentation assumes I already know which one I want, and how to use it.

If I run a "Local Install", obviously, all three components (Server, Client, and Agent) go on the physical box D.

If I run a "Client-Server Install", which components go on D, and where do the others then go??

What I can't figure out from the "User's Guide, vCenter Converter Standalone 4.0" (the PDF file referred to earlier), Chapter 4, page 35, "Converting Machines", is how I end up with the VM (E) back running on ESX servers B or C.

I don't see where the pieces go, and the documentation isn't clear about it at all.

• Where/on which piece of physical hardware do I install the Standalone Converter Server to accomplish my goal?

• Where/on which piece of physical hardware do I install the Standalone Converter Client to accomplish my goal?

• Where/on which piece of physical hardware do I install the Standalone Converter Agent to accomplish my goal?

• Which component(s) of Standalone Converter 4 do I have to run to create VM 'E' on ESX boxes B or C so the new VM is managed by VC "A"?

• How do I point <whatever process it is I have to run> so that it creates VM 'E' on ESX boxes B or C so the new VM is managed by VC "A"?

[vmweathers]

Ok, it's really not as complicated as you are making it out to be:

1. You run the installer on some machine that can reach the source and the ESX on the network. You don't try to install "only this" or "only that". You install everything. The only difference between a "local install" and a "client server install" is that it allows you to install the client on another machine and control the Converter Server from that other machine. This may not be necessary for you, but it doesn't hurt so I always do this.

2. You start the Converter Client (vmware-converter-client) on a machine where you have installed everything.

3. You select the remotely reachable IP (from ESX) of the source Linux machine for the powered-on source.

4. You select the target ESX (or the VC, but there is really no point in that you can just point directly to the target ESX).

5. You enter a helper VM static IP if necessary.

6. You watch the progress counter and see the conversion complete. You do not need to worry about anything else unless you have firewalls preventing traffic.

P.S. Another user has published this little guide to doing a LInux P2V with converter, you might find it useful:

http://www.vladan.fr/how-to-p2v-linux-into-vmware-esx-server/

(If your question has been resolved please mark the answers as "Helpful" or "Correct".)

[MitchK]

Ok, got it figured out, have been successfully P2V-ing for a couple of days now.

(Also been using it to do V2Vs, moving VMs that aren't on the current VCenter onto the current VCenter )

The part I had been missing was to install the Converter to a third, uninvolved box and run it from there.

All the boxes I'm working with are Linux boxes -- No Windows boxes in the current environment.

What I did was:

• Copy the Converter tarball to an uninvolved box

• Create /home/vmconverter to work in

• Untar the converter there via "tar xvf VMware-converter-4.0.0-146302.tar.gz"

• Install the Converter via "./vmware-install.pl"

• Start the Converter via "vmware-converter-client &"

When the Converter opened its Xwindows session, I selected the "Localhost" option.

On the next page, I followed the prompts and gave the hostname, username, and password of the source box (the box to convert)

On the next page, I followed the prompts and gave the hostname, username, and password of the ESX server to place the newly converted VM onto.

From there I chose the name for the newly converted VM, and after that it was just a question of making sure the options were correct. (# CPUs, memory, disk space, etc.)

Hit "Finish", and waited for it to do its magic. Was quick, clean, and simple.

Just need to know whether Converter went on the box to be converted or not, and which pieces to install. (The documentation makes it sound like you have the option of only installing certain pieces of the overall Converter -- You just want to install the whole thing in one go. (In fact, you aren't even given a choice: You either install the Converter or you don't. No choices about which parts to install.))

All in all, you were right in that it's way simpler than I originally thought -- My confusion was mostly caused by reading (and believing) the documentation, which is "Microsoft-helpful: If you understand it, you didn't need it in the first place. If you don't understand it, it won't be of any use."

Thanks for the pointers!

:smileyhappy:

[vmweathers]

Glad to hear you got it working.

The reason for the confusion in the documentation is probably that the product has historically been a Windows product, so the documentation is Windows-centric. In the Windows case, you can indeed install 3 separate components (client, server, or just agent). Linux does not require a "source agent" for performing Linux P2V. But you can choose install either the client or the server, so you can install certain pieces if you wish to.

Also it is notable that you can install the Converter on the source Linux machine that you are converting, you do not need to install it on a separate machine.

(If your question has been resolved please mark the answers as "Helpful" or "Correct".)

[Troy_Clavell]

don't forget to award points by marking answers as "correct" or "helpful"

there was some valuable time spent by Erik to help in troubleshooting your issue.