How to Create a Virtual Machine on QEMU-KVM

GUI

TUI

You can use virt-install to create a VM via the terminal:

# Install from local ISO
virt-install \
  --virt-type kvm \
  --name vm1 \
  --memory 512 \
  --vcpus 1 \
  --os-type linux \
  --os-variant centos7.0 \
  --network bridge=virbr1 \
  --disk /var/lib/libvirt/images/vm1.qcow2,size=10,bus=virtio,format=qcow2 \
  --location /soft/CentOS-7-x86_64-DVD-2009.iso -x console=ttyS0 \
  --graphics none

# Install from network location
virt-install \
  --virt-type kvm \
  --name vm1 \
  --memory 2048 \
  --vcpus 1 \
  --os-type linux \
  --os-variant centos7.0 \
  --network bridge=virbr1 \
  --disk /var/lib/libvirt/images/vm1.qcow2,size=10,bus=virtio,format=qcow2 \
  --location http://192.168.3.112/assets/tmp/mnt -x console=ttyS0 \
  --graphics none

# To configure the kernel console on ttyS0
grubby --update-kernel=ALL --args="console=ttyS0,115200n8"

# Connect to VM console
virsh console vm1 --force

Press CTRL + ] to escape console.

Command

# Update kernel to enable serial console
grubby --update-kernel=ALL --args="console=ttyS0,115200n8"

Delete VMs

# Kill the VM process
ps aux | grep [v]m1 | awk '{print $2}' | xargs kill -15

# Poweroff VM
virsh destroy vm1

# Remove VM definition and storage
virsh undefine vm1 --remove-all-storage

Troubleshooting

Issue: curl: (7) Failed connect to 192.168.3.112:80; Connection timed out

[    9.842048] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[   12.874243] dracut-initqueue[744]: RTNETLINK answers: File exists
[  140.788907] dracut-initqueue[744]: Warning: can't find installer mainimage path in .treeinfo
...
curl: (7) Failed connect to 192.168.3.112:80; Connection timed out
...
[  268.229474] dracut-initqueue[744]: Warning: Downloading 'http://192.168.3.112/assets/tmp//LiveOS/squashfs.img' failed!
[  268.231758] dracut-initqueue[744]: Warning: anaconda: failed to fetch stage2 from http://192.168.3.112/assets/tmp/
...
[  393.986291] dracut-initqueue[744]: Warning: dracut-initqueue timeout - starting timeout scripts

Cause:

The VM's eth0 cannot connect to the HTTP server.

Workaround:

Check the network settings of the virtual machine.

Issue: curl: (23) Failed writing body (12033 != 16384)

[   13.762605] dracut-initqueue[744]: curl: (23) Failed writing body (12033 != 16384)
...
[   14.040573] dracut-initqueue[744]: /sbin/dmsquash-live-root: line 286: printf: write error: No space left on device
...
[  139.573351] dracut-initqueue[744]: Warning: dracut-initqueue timeout - starting timeout scripts

Cause:

/sbin/dmsquash-live-root: line 286: printf: write error: No space left on device

Workaround:

  • For RHEL 7, increase the VM's memory to 2 GiB.
  • For RHEL 8/9, increase the VM's memory to 3 GiB.

Reference:

・ Kickstart Fails With Curl Error

Issue: qcow2: Image is corrupt; cannot be opened read/write

~]# virsh start vm2
error: Failed to start domain vm2
error: internal error: qemu unexpectedly closed the monitor: 2025-08-27T03:39:34.228948Z qemu-kvm: -drive file=/var/lib/libvirt/images/vm2.qcow2,format=qcow2,if=none,id=drive-virtio-disk0: could not open disk image /var/lib/libvirt/images/vm2.qcow2: qcow2: Image is corrupt; cannot be opened read/write

Workaround:

~]# qemu-img check -r all /var/lib/libvirt/images/vm2.qcow2
Leaked cluster 163869 refcount=1 reference=0
Repairing cluster 163869 refcount=1 reference=0
The following inconsistencies were found and repaired:

    1 leaked clusters
    0 corruptions

Double checking the fixed image now...
No errors were found on the image.
163840/163840 = 100.00% allocated, 0.00% fragmented, 0.00% compressed clusters
Image end offset: 10739318784

Issue: cdrom and floppy device hotplug isn't supported by libvirt

~]# virsh attach-disk vm2 /soft/CentOS-7-x86_64-DVD-2009.iso hdb --driver qemu --type cdrom --mode readonly
error: Failed to attach disk
error: internal error: No device with bus 'ide' and target 'hdb'. cdrom and floppy device hotplug isn't supported by libvirt

Workaround:

~]# virsh shutdown vm2
~]# virsh edit vm2
<disk type='file' device='cdrom'>
  <driver name='qemu' type='raw'/>
  <source file='/soft/CentOS-7-x86_64-DVD-2009.iso'/>
  <target dev='hda' bus='ide'/>
  <readonly/>
</disk>

Issue: XFS (vda3): metadata I/O error in "xfs_buf_iodone_callback_error"

[  334.287041] XFS (vda3): metadata I/O error in "xfs_buf_iodone_callback_error" at daddr 0x13ff6e0 len 32 error 5
[  334.836243] blk_update_request: 993 callbacks suppressed
[  334.839413] blk_update_request: I/O error, dev vda, sector 20971704
[  334.844636] blk_update_request: I/O error, dev vda, sector 20971712
[  334.849579] blk_update_request: I/O error, dev vda, sector 20971744
[  334.855046] blk_update_request: I/O error, dev vda, sector 25165504
[  334.860136] blk_update_request: I/O error, dev vda, sector 25165536
[  334.865183] blk_update_request: I/O error, dev vda, sector 29359296
[  334.870404] blk_update_request: I/O error, dev vda, sector 29359328
[  334.885488] blk_update_request: I/O error, dev vda, sector 20971704
[  334.889038] blk_update_request: I/O error, dev vda, sector 25165504
[  334.894376] blk_update_request: I/O error, dev vda, sector 29359296
[  339.135292] XFS: Failing async write: 4183 callbacks suppressed
[  339.138493] XFS (vda3): Failing async write on buffer block 0xfff8e0. Retrying async write.
[  339.143453] XFS (vda3): Failing async write on buffer block 0xfff8c0. Retrying async write.
[  339.148493] XFS (vda3): Failing async write on buffer block 0x17ff4e0. Retrying async write.
[  339.154292] XFS (vda3): Failing async write on buffer block 0x17ff4c0. Retrying async write.
[  339.159837] XFS (vda3): Failing async write on buffer block 0x13ff6e0. Retrying async write.
[  339.164055] XFS (vda3): Failing async write on buffer block 0x13ff6c0. Retrying async write.
[  339.168882] XFS (vda3): Failing async write on buffer block 0xfff8b8. Retrying async write.
[  339.184288] XFS (vda3): metadata I/O error in "xlog_iodone" at daddr 0x8026a7 len 64 error 5
[  339.189152] XFS (vda3): xfs_do_force_shutdown(0x2) called from line 1238 of file fs/xfs/xfs_log.c.  Return address = ffffffffc0466810
[  339.195251] XFS (vda3): Log I/O Error Detected.  Shutting down filesystem
[  339.199252] XFS (vda3): Please umount the filesystem and rectify the problem(s)

~]# reboot
-bash: reboot: command not found
~]# ls
-bash: /bin/ls: Input/output erro

The root partiton xfs filesystem I/O error and has been force shutdown.

Workaround:

~]# qemu-img check -r all /var/lib/libvirt/images/vm2.qcow2
Repairing refcount block 0 is outside image
ERROR could not resize image: Invalid argument
...
Repairing refcount block 27 is outside image
ERROR could not resize image: Invalid argument
Can't get refcount for cluster 0: Invalid argument
Can't get refcount for cluster 1: Invalid argument
...
Can't get refcount for cluster 163880: Invalid argument
Can't get refcount for cluster 163881: Invalid argument
Rebuilding refcount structure
Repairing cluster 1 refcount=1 reference=0
The following inconsistencies were found and repaired:

    0 leaked clusters
    23 corruptions

Double checking the fixed image now...
No errors were found on the image.
163843/262144 = 62.50% allocated, 0.00% fragmented, 0.00% compressed clusters
Image end offset: 10739777536