Are you new to the company, datacenter or a third party who is responsible for deploying a fleet of servers from scratch.
The first step is to normally login to the KVM so you can perhaps manually reinstall, PXE boot the Cloud Image or reimage/reinstall an OS but you need access to the KVM/IP or what Dell calls iDRAC.
It's common that you may have forgotten this information or that another employee or colleague has changed the info and did not tell you, that they have left the........
It's best not to mix the two technologies. Here is how to fix things but break Docker.
If you do an iptables -L you will notice even if you deleted all the Docker chains that the iptables FORWARD policy is enabled and is set to drop, this causes your VMs to not have networking, at least not outside the host machine.
Chain FORWARD (policy DROP)
target prot opt source&nb........
This can be used on almost anything, since Gluster is a userspace tool, based on FUSE. This means that all Gluster appears as to any application is just a directory.
Applications don't need specific support for Gluster, so long as you can tell the application to use a certain directory for storage.
One application can be for redundant and scaled storage, including for within Docker and Kubernetes, LXC, Proxmox, OpenStack, etc or just your image/web/video files or even da........
Let's say you have a VM file that uses 200G of dynamic space, but really only has 40G in usage. If you add fles and delete, at some point the file will be larger than the current space you are using.
Take this image which shows is using 71G of space on the host:
The actual space being used inside the image is about 43G as we can see:........
Datacenters, Cloud, VPS and Dedicated Servers in Hong Kong are ping neutral to the world and rest of Asia
This article is the best guide on the internet for all things Hong Kong internet and how it applies to your business, based on decades of experience and research, to help you make........
For a lot of reasons, it may be convenient to detach or attach live disks to a running VM without having to reboot it. Sure, you can use some network based storage, but when performance counts, attaching a new virtual disk will usually give you better throughput and lower latency in a quick testing situation.
This doesn't work, why not?
drive_add 0 if=virtio,file=/tmp/vm.qcow2,if=virtio,format=qcow2,id=rtt
Here is the scenario, you are using QEMU/KVM and are using something like the AC97 sound driver to pass the host audio to the guest via pulseaudio. This is useful because you can transparently pass your mic input from the host which means you can mute your microphone from the host, which prevents the guest from receiving any mic input even if unmuted.
Mute / Unmute Fix
This issue also seems to happen even if you press the mute button on the microphone and then unmute,........
It seems that QEMU/KVM's default PS2 mouse and keyboard doesn't work right in most cases. I have especially observed issues using Ctrl+C and Ctrl+V and in Linux you may see repeated key presses in the terminal and you will wonder why you copied something and it's not in the clipboard when you try to paste. The way to temporarily fix it is to press the key that is repeating once(works in Linux but not really in Windows).
Sometimes when moving your mouse it will also s........
How To Compile QEMU Manually (using sensible options)
1.) Download the QEMU source file you want.
2.) Extract The Source File
tar -Jxvf qemu-4.2.0.tar.xz
3.) Switch to the extracted source
4.) Make sure we have the right libraries and tools to compile QEMU manually
sudo apt install build-e........
Interestingly enough Windows 2000 works fine on QEMU 64-bit but you have to specify Pentium as your CPU otherwise it doesn't complete the install (it will not pass the detecting/setting up devices phase).
-vga cirrus is wise because it is supported by Windows 2000 and allows higher resolutions and 24-bit color.
-cpu Pentium emulates an old computer and is necessary for install to complete
-device rtl8139 is important as this oldschool Realtek 8139 NIC is supported by W........
In QEMU 4 or higher you can no longer use the normal "-soundhw ac97" flag and it is much more complicated but here is a simple copy and paste on Linux that will just work:
-audiodev you have to use -audiodev to specify the driver and id
-device you have to specify the same audiodev id you used in -audiodev and driver
-audiodev driver=pa,id=pa1 -devic........
If you've come here, don't be embarraassed, working in IT, this is the MOST common computer problem that almost everyone will encounter. The reason why I'm doing this post is because I've seen an increase from colleagues and admins having this problem and many times it's not even your fault. A common scenario is that someone acquires a new or used computer which they weren't given the password for. Fortunately Ihave a detailed list of all the options whether free or pa........
/usr/libexec/qemu-kvm -enable-kvm -boot order=cd,once=dc -vga cirrus -m 4096 -drive file=~/23815135.img,if=virtio -usbdevice tablet -net nic,macaddr=DE:AD:BE:EF:D4:AB -netdev bridge,br=br0,id=net0
qemu-kvm: -usbdevice tablet: '-usbdevice' is deprecated, please use '-device usb-...' instead
access denied by acl file
qemu-kvm: bridge helper failed
[root@CentOS-82-64-minimal 23815135]# /usr/libexec/qemu-kvm -enable-kvm -boot order=cd,once=dc -vga cirrus -........
FATAL: kernel too old
This happens for example if you are in Centos 6 and trying to chroot into a system based on a newer kernel like 4.x+
You'll have to use a newer OS/kernel system to chroot into the environment or a VM running a newer kernel.........
Iam going to build this based on a series of small howto QEMU / KVMposts I've made as I feel much of the information is actually hard to find and piece together from the rest of the web.
What I'm going to focus on is how to use virtio as the NIC because if you don't you get very slow NIC speeds but with the virtio NIC model you basically get host speeds.
/usr/libexec/qemu-kvm -enable-kvm -smp 8 -m 16000 -net user -net nic,model=virtio -drive file=ubuntu-gpt2l........
This is caused because the user is running as qemu for virt-resize and if qemu does not have privileges to read from the source and write to the destination, it will fail with the below. So either change the uid of qemu or change the ownership of the source and target.
virt-resize --expand /dev/sda2 /root/kvmtemplates/windows2019-eval-template.img /root/kvmguests/kvmkvmuser4515........
It's as simple as below where you just specify the dev device of the CDROM which is usually /dev/sr0. You can boot actual bootable discs like Windows, Linux, etc straight from a physical drive this way.
sudo qemu-system-x86_64 -cdrom /dev/sr0 -m 4096
sudo usermod -a -G groupname username
It's really simple like above, the -a is for append so that you are not changing their main group, but adding them to another additonal group. Just change "groupname" to your group and "username" to the user you want to be added to "groupname".
A common task these days is getting your user access to kvm for virtualization so the KVM/QEMUprocess........
This should work but the key thing is having the "-cpu host" flag.
Once you add the correct -cpu host flag then it should boot just fine on KVM.
qemu-system-x86_64 --enable-kvm -cpu host -smp 8 -m 8192 -drive format=raw,file=the-file.img
Examples can be found here on how to boot Windows properly with KVM.........
Now older versions of qemu-kvm didn't throw this error say if you just had "-video cirrus" when starting qemu-kvm. But newer versions do care.
And this probably only applies to you if you are running from bash/terminal with remote kvm images.
What you need to do is remove the "-video" part and just add -vnc :5
eg. this would fix the error:
qemu-system-x86_64 -enable-kvm -boot order=cd,once=dc -m 1024 -drive........
CPU:Intel(R) Core(TM) i3-2120 CPU @ 3.30GHz
MOBO: Manufacturer: ASUSTeK COMPUTER INC.
Product Name: P8H61-M LX3 PLUS R2.0
This is weird but the only OS I've found this machine doesn't work with is Windows 2019 Server. Ihave no idea, when 2008, 2012 work f........
Install Errors on Version 12:
This error happened on QEMU emulator version 2.11.1 pve-qemu-kvm_2.11.1-5
on Proxmox/Debian but installing on QEMU.12 on Centos 6 did not produce the error.
*Update it is not related to the OS or QEMU version. This happened in Centos 6 too after a second install.
What really causes this even though you successfully install........
The strange thing is that usually the first install or two will work on any new machine but then it suddenly won't. I had this experience on QEMU 2.13 on a different machine. There is something finicky or buggy about the CUCM installer even when choosing the same virtual hardware specs.
QEMU PC emulator version 0.12.1 (qemu-kvm-0.12.1.2-2.506.el6_10.1), Copyright (c) 2003-2008 Fabrice Bellard
Pass QEMU this flag:
Or you could use (for a Ensoniq soundcard):
However on some machines I get this error:
pulseaudio: pa_simple_new for capture failed
pulseaudio: Reason: Connection terminated
pulseaudio: pa_simple_new for capture failed
pulseaudio: Reason: Connection terminated
audio: Failed to create voice `ac9........
You may have noticed if you are running QEMU/KVMmanually that in Windows the the position of the physical mouse does not match where the mouse is positioned within Windows.
There is an easy command to pass to qemu-kvm or qemu-system (whatever you call your binary):
The above flag will fix your mouse pointing problems whether you are running Windows 95, 98, NT, XP 2000, 2003, Vista, 7, 8 10 or Server 2000, 2003, 2008, 2012, 2016 or 2........
WARNING: Image format was not specified for '/mnt/space/cucm12.img' and probing guessed raw.
Automatically detecting the format is dangerous for raw images, write operations on block 0 will be restricted.
Specify the 'raw' format explicitly to remove the restrictions.
#you should manually specify the format
The key thing is that you must use a "machine"id of "pc-1.3" or it will say your hardware is not supported.
Additionally you MUST use a virtio disk or you will get a ks_pre.sh error as soon as the install starts (a look at logs will show it can't find a disk). This is funny because even though the OS finds the disk and an fdisk -l shows it, it looks like the script looks for a /dev/vda device (virtio) and nothing else, so if you didn't use Virtio as you........
What you need to do if you have taken a dd or real raw image dump of a hard disk:
VBoxManage convertdd windows2019-eval-template.img windows2019.vdi --format VDI
The .img is the raw dd dump and the .vdi is the output file.
--format VDIspecifies to output to .vdi format
If you are in a pinch you can always use qemu-kvm binary and manually specify the .img as your disk and i........
1.) Create Image using qemu-img
qemu-img create -f qcow2 skype.img 40G
2.) Start VM using flags
-m = memory in MB
qemu-system-x86_64 -m 4096 -drive file=skype.img -cdrom ~/Downloads/SfB-E-9319.0-enUS.ISO
Enable Bridged Networking........
sudo apt-get install hwloc-nox
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following NEW packages will be installed:
0 upgraded, 1 newly installed, 0 to remove and 530 not upgraded.
Need to get 151 kB of archives.
After this operation, 453 kB of additional disk space will be used.
How to check if nested KVM is enabled by doing a cat on nested inside sys
Nested KVM is mainly important for testing for example if you wanted to install VMWare or Proxmox Nodes in a virtual environment for testing. Without nesting, the performance will be extremely slow, since the VMs within the nodes will not be using Virtualization extensions.
I've used wildcard on kvm_ because it could be kvm_intel or kvm_amd depending on whether y........
It could just be my specific Java but other KVM/IP works ok on my machine but with Supermicro's IPMI for some reason the console window doesn't resize and even putting it in full screen leaves the window cut off as shown below:
As you can see above the screen is cut off you can't see the "Iagree" on the right side.
The window resi........
The Correct Way To Resize In Place
qemu-img resize kvmuserwindows2008dcetest.img +1G
Below is a common mistake that some users make they are trying to specify a new image name but it can be resized in place (just make sure the VMis NOT running and you've backed up the data in case something goes wrong).
qemu-img resize kvmuser453111.img kvmuser453111-larger.img +5G
I've read a few guides about this but they didn't work for me.
sudo apt-get install bridge-utils
#don't think the above is enough it won't work still even though you have by default an /etc/qemu-ifup that handles it if you have the right tools and setup
sudo qemu-system-x86_64 -net tap -net nic -enable-kvm -cpu host,vmx=on ~/VirtualBox VMs/vsphere-vcenter/vsphere-vcenter.vdi
W: /etc/qemu-ifup: no bridge for guest interface foun........
I can't get vmx cpu extensions to show up in Virtualbox guests despite enabling nested paging and
enable vmx in virtualbox guest but this doesn't help that you check VT-X or the AMD Virtualization SVM it enables it for the guest to use BUT does not pass it through. This means if you check cat /proc/cpuinfo in the guest you will see the CPUdoesn't support virtualization. It looks like VirtualBox still hasn't implemented this!
But there is good news I&n........
Intel VT-X is enabled in Virtualbox but it doesn't seem to pass through the needed vmx extension despite the following variables on the host confirming it is enabled:
OVF Tool: Disk progress: 99%
OVF Tool: Transfer Completed
OVF Tool: Powering on VM: Embedded-vCenter-Server-Appliance-
OVF Tool: Task p........
guestmount -a kvmuserscra.img -m /dev/sda1 mount
libguestfs: error: mount_options: /dev/sda1 on / (options: ''): mount: unknown filesystem type 'ntfs'
guestmount: '/dev/sda1' could not be mounted.
guestmount: Did you mean to mount one of these filesystems?
guestmount: /dev/sda1 (ntfs)
guestmount: /dev/sda2 (ntfs)
yum -y install ntfs-3g
/usr/bin/supermin-helper exited with error status 1.
To see full error messages you may need to enable debugging.
See http://libguestfs.org/guestfs-faq.1.html#debugging-libguestfs at /usr/bin/virt-list-partitions line 177.
Just a note before you do this you should have a sure, guaranteed way into the system such as local, KVMor preferably publickey making bruteforce SSH absolutely impossible since there is no password to bruteforce and even if someone knew the password they wouldn't be able to login except from the local console (presumably you should make sure no one unauthorized has physical access).
1. Edit /etc/ssh/sshd_config
Find the section like this:........
Iam not sure why this is happening neither the hostnode or VM changed. All I did was reboot the hostnode and startup the Centos VM again, also note it happened with the original kernel on the VM and also the latest 6.9 kernel as of this writing as shown below.
Host Node: Centos 6.9
Same result in any kernel above........
When running cudaminer once it tries to initialize the card the entire screen freezes. The computer itself is still running but the Xorg is done for, you cannot even switch to another console window and must reboot (even an mdm or Xorg restart does not help).
At first cudaminer will give you these errors:
...retry after 15 seconds
GPU #0: Geforce 210 with compute ca........
This error is commonly due to Java security or TLS settings but there is a second issue with forwarded ports that also causes it.
1. Java Security/TLS Settings issue:
This article has the solution to change them all in Linux automatically
2. Port Forwarding Issue if your Avocent DSR is behind NAT/private IP........
# yum -y install qemu-kvm
Loaded plugins: fastestmirror
Setting up Install Process
Loading mirror speeds from cached hostfile
* base: mirror.lzu.edu.cn
* elrepo: ftp.utexas.edu
* epel: ftp.jaist.ac.jp
* extras: mirrors.aliyun.com
* openvz-kernel-rhel6: mirror.fdcservers.net
* openvz-utils: mirror.fdcservers.net
* updates: mirrors.nwsuaf.edu.cn
This command in Debian/Ubuntu/Mint will get everything need installed for most Java based KVM viewers:
sudo apt install icedtea-netx
The following additional packages will be installed:
ca-certificates-java icedtea-netx-common openjdk-8-jre openjdk-8-jre-headless
It seems every other updated version of Java or Icedtea breaks things and I will save the frustration of Java for another post.........
Here is the scenario you or a client have a remote machine that was installed as a standard/default minimal Centos 6.x machine on a single disk with LVM for whatever reason. Often many people do not know how to install it to a RAID array so it is common to have this problem and why reinstall if you don't need to? In some cases on a remote system you can't easily reinstall without physical or KVM access.
So in this case you add a second physical or disk or already ha........
windows cannot find the microsoft license terms windows 2012 server install
This is actually because you are using less than 576MB of RAM. In my case I was installing on a KVM VPS with 512MB of RAM. This issue would apply to any physical or virtual server with less than 576MB of RAM. This includes Virtualized VPS Servers with XENHVM, KVM, VBOX, VMWare etc.. or Dedicated Servers with such little RAM.
It's very misleading of course since it has not........
This is useful if you are installing Windows and need virtio but of course Windows needs the virtio driver which is on a second iso.
This is the line of code you would add to your kvm startup script and then you get this iso as second cdrom you can browse to for your Windows install of virtio driver's or whatever other use you need it for.
Thsi is very handy when doing your own kernel development.
-m specifies how much ram (in the example it is 768MB)
-kernel specifies the path to the kernel file
-net tap,ifname=tap1,script=no (the ifname=tap1 is what you need to change and setup manually).
*Run "tunctl -b" to create a tap device and use the one it gives you for ifname=
Enable networking to the outside like this:
*Note we assume that your bridge is br0 i........
I messed up the bootloader by accident on a standard Centos 6.3 install because I turned the /dev/vda1 boot partition into an mdadm raid 1. This was all done correctly aside from one point Ididn't realize was an issue metadata=00.90 is the only thing that will allow you to boot (otherwise grub won't work and you won't boot).
So the next step is rescue mode from a CD right? The problem you will find is that grub does not detect your hard drives, this is Ibelieve is be........
If the below is happening on KVM (a very weird and scary looking error) it's probably because of Windows. This has happened countless times to me where the bootsector on Windows 7/2008 becomes corrupted easily (even by a crash or shutdown).
KVM: unknown exit, hardware reason 0x80000021
kvm_run returned -22
rax 0000000000000010 rbx 0000000000000080 rcx 0000000000000000 rdx 0000000000000080
rsi 000000000025db2a rdi 000000000007db2a rsp 0000000000000200 rbp........
libguestfs tools howto guide for managing virtual machine images.
libguestfs-tools aka guestfs tools has a lot of tools that make this very easy for you. You can easily mount partitons from an image with some of the commands below.
To mount a partition
#mount the kvmuser102821.img image and the /dev/sda1 partition from it to the local directory "mount"
guestmount -a kvmuser102821.img -m /dev/sda1 mount
error: Failed to create domain from /home/kvm/kvm101/kvm101.xml
error: cannot open file '/dev//dev/kvmcontainer/kvm101_img': No such file or directory
This is caused by what we consider a quark in SolusVMthat Ihelped a client with.
SolusVMhas as config for the "LVMvolume name" and does not enforce any convention.
Naturally most technical people would use the actual path eg "/dev/kvmcontainer".
OpenVZ has made vzctl version 4.7 default to using ploop which is a big annoyance. No one wants it otherwise we'd use Xen or KVM.
Make sure to manually specify vzctl 4.6.1 or you will have issues with old scripts breaking since it defaults to using ploop (a single image like Xen/KVM).
Here's a list to old versions of vzctl.........
# first we need a physical volume which we use the pvcreate tool to create
# I create mine on /dev/sdb3
dev_is_mpath: failed to get device for 8:19
Physical volume "/dev/sdb3" successfully created
# pvdisplay shows the newly created volume
"/dev/sdb3" is a new physical volume of "1.35 TiB"
--- NEW Physical volume ---
#count=10000 makes an image of 10000MB make sure your image is at least the same as your existing
dd if=/dev/zero of=yourimage.img bs=1M count=10000
# losetup -fv newimage.raw
# fdisk -cu /dev/loop0
# kpartx -a /dev/loop0
# dd if= of=/dev/mapper/loop0p1
# e2fsck -f /dev/mapper/loop0p1
# resize2fs /dev/mapper/loop0p1
# a lot of guides tell you to edit /etc/fst........
Current Direct stable URL from Fedora: https://fedoraproject.org/wiki/Windows_Virtio_Drivers
Direct Downoad of Stable Virtio: https://fedorap........
lvextend -L +10G /dev/kvmvm/w2k8r2evalstandard
Extending logical volume w2k8r2evalstandard to 20.00 GiB
Logical volume w2k8r2evalstandard successfully resized
The above adds 10GB to the logical volume. Of course you must resize the filesystem using other tools to take advantage of the space.........
./configure: cannot locate gcc 3.x. please install it or specify with --qemu-cc
yum -y install gcc make
./configure: cannot locate gcc 3.x. please install it or specify with --qemu-cc
yum -y install compat-gcc-*
Error: Could not find alsa
Make sure to have the alsa libs and headers installed.
yum -y install alsa-lib-devel
On the same OS with the same kernels basically I cannot boot FreeBSD installer, I believe the issue is the host CPU.
One host runs an Intel and works, while an Opteron 2373 does not work despite suggestions from this thread: http://188.8.131.52/showthread.php?t=36761
I've tried cpu options like "kvm64" "Opteron_G3" and none........
open /dev/kvm: No such file or directory
failed to initialize KVM: Operation not permitted
[ 96.084502] kvm: disabled by bios
Any of the above means that "Virtualization" is not enabled in your BIOS. Most servers and Desktops (non-mainstream) can do this no problem but leave it off by default. Note that some systems even though the CPU supports it, don't allow enabling of Virtualization for some reason (especially some lap........
qemu-img create -f qcow2 isotest.qcow2 10G
The above creates a "qcow2" format image called "isotest.qcow2" that is 10GB in size.
If you want to preallocate all of the space for increased performance do this:
qemu-img create -f qcow2 isotest.qcow2 -o preallocation=full 10G
It is controlled with the preallocation=full option, of course by default this is set to off........
Once I set a new MAC address everything was fine and finally anything can be pinged etc.., I've had this happen with many clients in different datacenters on different hardware.........
This seems to be a verified bug on KVM with no solution. I wasn't able to get through the installer with Ubuntu 12.04, it would crash while "configuring apt". I was able to install it with Ubuntu 12.10 but it crashes randomly at times.
This happens with the latest Centos 6.3 kernel and KVM package, I even compiled the latest version from the KVM website and used that, but the results are still the same.
Here's some links to discussion about the bugs:........
This booting error is because the Xen PV guest image uses the Xen kernel, this is not compatible with anything but a host running a Xen kernel.
I did a kpartx -av virtual.img and then it created some partitions that showed up in fdisk.
I mounted it and did a chroot into it and removed the xen kernel and installed a normal kernel but Xen still shows the same kernel in Grub (only the Xen one).
This is strange but it seems like this Xen PV guest has some sort of hidden or........
The example below converts youvm.img into a vdi.
The -O vdi specifies the new format will be vdi but you can use any other valid formats/vice versa.
qemu-img convert yourvm.img -O vdi yourvm.vdi
This is very handy for those who need to switch between virtualization technologies.........
Let's keep this simple, it's horrible, not only is the picture quality horrible (it's not a big deal) but the keyboard input is an issue. You have to type slowly, lightly and carefully otherwise 1-keystroke could end up being dozens of keystrokes.
It's also a confusing unit for many datacenter techs, I've seen these units hooked up wrong and the instructions themselves aren't clear. I had to resort to shipping one to my office and take my phone pictures of the setup for th........
I've run into two issues with Lantronix based KVMs on various servers and here's how I solved them (with that said I like these units as they are Java based and OS independent and work very well, unike some other models like Startech).
1.) If you are connected by USB only and you're sure USB support from the BIOS is enabled, you just need to click the following in Lantronix
Interfaces -> Keyboard/Mouse
Check "Force USB Full Speed Mode", this fixed the issu........
Disabling PIE due to missing toolchain support
glib-2.12 required to compile QEMU
Solution install glib2
*Don't confuse glib2 with glibc, they are different and it may catch some off guard.
yum -y install glib2*
After that KVM should compile and install just fine.........
This seems to be common for some reason when trying to run Windows 7 under KVM and 7 has always had strange issues with KVM. What happens for me is after the install it simply won't boot, it stays on the POST screen and does nothing.
For some reason it appears the boot sector either doesn't get installed or gets corrupted with KVM.
The solution is to boot the install disc for Windows 7 and reinstall the windows MBR into the boot sector:
1. Boot t........
qemu-img create -b centos.5-8.x86.20120308.qcow2 -f qcow2 ../kvmguests/25000-centos5.8x86.qcow2
Formatting '../kvmguests/25000-centos5.8x86.qcow2', fmt=qcow2, backing_file=centos.5-8.x86.20120308.qcow2, size=10485760 kB
-b the source/base image
-f format is qcow2 and the location of the destination image
What is so special about this? It's even quicker than creating a template with OpenVZ but this is an actual OS.
It saves time a........
I am running a 64-bit host and was running a 32-bit XPguest which was crawling no matter what I was doing with it,in fact it frequently used 100% CPU power while having no real work load. I experienced with this other guests and wondered why KVM was slow compared to say Xen or Virtualbox.
This is because you need to use the "-cpu" flag.
For a full list of options run:
/usr/libexec/qemu-kvm -cpu ?
Failed to open a session for the virtual machine XP.
AMD-V is being used by another hypervisor. (VERR_SVM_IN_USE).
VirtualBox can't enable the AMD-V extension. Please disable the KVM kernel extension, recompile your kernel and reboot (VERR_SVM_IN_USE)
Solution - Unload KVM module (you can't use two hardware virtualization suites at once)
sudo rmmod kvm_amd kvm........