I am currently in the process of creating a home server, made up of several virtual machines, each running a different type of server (i.e. a file server, a multimedia server, a firewall and a print server as of now).

Today I discovered this new software, Zentyal, which looks interesting for system and network administration. However, I haven’t understood one thing: can it be used only with its preconfigured packages or can it also be used to control other servers running other programs (in my case, the four virtual machines)?.

Also, would you suggest to use such a program or do you think it’s better to control each server on its own?

I’m new to sysadmin and this site, but I’ll try my best to answer this based on my experience with that distro.

You don’t specify what you’re using it for, and how many boxes you have to work with (physical or virtual) so we’ll say 2 possibilities. First, you just want to run some servers at home and this isn’t live and essential to some venture. Then just use one box, with Zentyal, and redundancy and survival is not an issue. Second, this is for some business or operational function and therefore necessary. Then do not host all servers on one box for standard practice reasons.

As to whether to use Zentyal or not overall, it’s a decent distro but I wouldn’t use the free version for a massive network. The commercial version is much more robust with better support. Having all services on one box is not typically allowed in production networks but from home, it’s not as big of a deal and you make the call.

It sounds like you’re asking whether other servers can be used on a Zentyal installation. It’s still just Ubuntu Linux so you can put anything else on there. But if you want access by way of the web interface, then I believe you’ll have to code that into the files (it’s written in PHP I think) to allow those additional servers to be shown. Otherwise just access them by CLI and save time.

Check more discussion of this question.

I am trying to virtualize an Ubuntu machine using VMWare vCenter Converter, but ran into permissions issues. I followed the instructions of part 1 and 2 on this page but when I got to “For Ubuntu operating systems further configuration is needed” I started running into trouble. I’m decent at Linux, but I’m not an experienced sysadmin.

• How do I Target the specific user you will be using and assign it
user id 0/group 0?
• How do I Ensure that you also still enable Allow
root to ssh even though you are not using the root account?

Thanks for your help.

How do I Target the specific user you will be using and assign it user id 0/group 0?

Don’t assign the user id or group id 0 to the user you are targeting. Add the user to /etc/sudoers as specified. Then flag that sudo will be uses also as documented. sudo will be used to run command as root (user id 0, group 0) when needed.

How do I Ensure that you also still enable Allow root to ssh even though you are not using the root account?

This is controlled in the file /etc/ssh/sshd_config. If keys are used for logins, then you need a line like:

PermitRootLogin without-password

Check more discussion of this question.

How can I install Centos 6 on SCSI hard drive under Hyper-V? I added a IDE drive to boot from but CentOS doesn’t recognize SCSI drive during install. How can I fix this please?

Using Linux Integration Services for Hyper-V is the solution. Last version is really easy to install and there is only one problem, After upgrading Linux Kernel OS will crash while booting. to fix this you can switch back to older kernel in grub listings.

Check more discussion of this question.

I added a Volume group in Proxmox VE as Storage for KVM VMs. A new Logical Volume will be created for each VM, but in Proxmox VE under The Defined Storage which shows list of created Logical Volumes I see my Logical Volume marked as RAW format. Is this a RAW image or a Logical Volume (LVM)?

But my main question, Does this really avoid the overhead of going through the filesystem layer on the host system ?

In this case, “RAW format” means that it’s a bit-by-bit image of what your VM will see as a disk, without any extra information, headers, metadata, or anything else. A RAW format disk may be stored in a file, but it can also be directly in a block device. You might call it an “image” in both cases, although it’s more common to call it an “image” when it’s in a file.

In your case, the answer is yes: the disk is backed by a LVM volume, which qemu sees as a block device, and the overhead is certainly much lower than what you would have by going through the filesystem layer.

Check more discussion of this question.

I’ve installed WindowsServer 2008 R2 on the computer with Xeon3430 2.40GHz processor. VT was activated in BIOS. Hyper-Role was installed and a VM could be created and started. Running pidenu32.msi-tool results, that this processor type doesn’t support VT. What is wrong?

That’s correct. The virtual processor that ran the program does not support VT. Intel x86 CPUs are not fully virtualizable, and once virtualized no longer support virtualization. You can’t virtualize again inside a virtual environment. So from the point of view of the program, VT is not supported because it’s already been “used up” by the environment running the OS that ran the program.

Strictly speaking, it is possible to do this — ESX(i) 5.0 can do it.

(I believe there’s a note to that effect somewhere in the tab.)

Check more discussion of this question.

I’m following this construction to install (version 4.1.2) on a CentOS 6.2 (x86_64) system.
When I use this command

yum install http://au1.mirror.crc.id.au/repo/kernel-xen-release-6-3.noarch.rpm

I face this error:

Loaded plugins: fastestmirror, security
Loading mirror speeds from cached hostfile
 * base: ftp.sh.cvut.cz
 * extras: ftp.sh.cvut.cz
 * updates: ftp.sh.cvut.cz
Setting up Install Process
Cannot open: http://au1.mirror.crc.id.au/repo/kernel-xen-release-6-3.noarch.rpm. Skipping.
Error: Nothing to do

there is a third-party repository with some mirror list, but I don’t know if I could use them? and how would that possible?
I would appreciate if any one could lead me to a available link to download Xen Repository.

You received the error because the file does not exist.

The reason the file does not exist is that the file in the repo is named kernel-xen-release-6-4.noarch.rpm, while your howto has a 3 instead of the 4.

Check more discussion of this question.

I recently setup a virtualization server for the small company I’m running.
This server runs few virtual machines that are used for development, testing, etc…

My business partner works from a remote location, thus I also installed a vpn server on the virtualization host to make it possible for him to safely reach the company services. Moreover, again on the virtualization host, I installed bacula to perform the backup of the data.

Is it advisable/good practice to do so or should I create one more virtual machine to do backups and VPN? Is it a bad idea to run these services on the host itself? If yes, why?

The virtualizaton host should be the most secure machine you have. The most secure machine is one that is not connected to a network at all

Having that in mind it is best not to offer any services on your public interfaces.
You should not even have an IP there (a bridge for VMs is layer2).

Think of the VM-host as DMZ: traffic into it is forbidden, originating no problem.

So in your example:

• VNC: Bad – this is an incoming service
• Backup: No problem – sessions are initiated from here to the outside

But even then – you should only run services that will not eat up RAM/CPU/IO on your VM-host – else your VMs will suffer from lack of resources.

Check more discussion of this question.

I’m hoping for a high level discussion of this problem I know is quickly approaching my application.

I have a server that binds on a specific port and manages TCP Sockets from my clients. I know that it is common practice to use VMWare to virtualize servers and run multiple servers at once.

How can a single server design be changed to support multiple servers?

Multiple servers can’t bind to the same port. If I had to guess I would say a proxy server binds to the port and then sends connections off to the other servers to be handled as if it was still a single server application.

I’m wondering what options there are and what are the common practices for solving this problem?

Thanks in advance!

Each VMWare machine is essentially a standalone PC, as far as the OS running inside a VM is concerned. Like on a real PC, a server can bind to the same port on different IPs, and each running VM can have its own IP on the virtualized network. VMWare has a built-in DHCP router that manages the IPs and network traffic of its running VMs. Each VM can have its own real independant IP on the Host machine’s actual network, or they can share the Host’s IP via VMWare’s built-in router. The OS in each VM doesn’t know the difference. Go ahead and bind your server to the same port on each running VM, it will work fine. VMWare will handle the packet routing as needed.

Check more discussion of this question.

Our company gives a VM for each developer which is exclusive for them but doesnt give root access.

I am planning to ask for root password so I can install different softwares on the VM.

Just want to make sure, having a root password on VM doesnt give me any control on the host system.

Nope. It shouldn’t give you any host access, assuming you’re on one of the major virtualization suites (VMWare, KVM, Xen, etc.).

Check more discussion of this question.

Is there a way to know if the machine I’m working on is virtual or physical?
(I’m connecting with RDP to the machine. If it’s a virtual machine it is working and handled by VMWare).

Thank you in advance.

If it’s Windows, just have a look at the hardware screens. It’ll have a billion and five VMWare-branded virtual devices.

Check more discussion of this question.