install-kvm

Setup a new kvm domain fast

guest setup

or: how to install Debian.

Define a shell function

# set up environment
setup_env() {
 echo -n "New hostname: " &&
 export LC_ALL=C &&
 read guest &&
 target="/mnt/target-$guest" &&
 mirror=`cat /etc/apt/sources.list /etc/apt/sources.list.d/debian.list 2>/dev/null | awk '/^deb.*debian/ {print $2; exit}'` &&
 vgdefault=`vgdisplay -c | awk -F: '{print $1;exit}' | sed 's/ *//g'` &&
 echo -n "Volume group? [$vgdefault]: " &&
 read vg &&
 if [ "$vg" = "" ]; then vg="$vgdefault"; fi &&
 echo -n "Use lvm for non-swap partitions? [Y/n] " &&
 read use_lvm &&
 if [ "$use_lvm" = "n" ]; then
   : SAN, as in the msa2012i at ubcece &&
   : requires that it is already setup &&
   dev_root="/dev/mapper/$guest-root" &&
   dev_boot="/dev/mapper/$guest-boot" &&
   echo "Root device will be $dev_root" &&
   echo "Boot device will be $dev_boot" &&
   echo "Make sure they exist already." &&
   fs=xfs
 else
   use_lvm="y" &&
   dev_root="/dev/mapper/$vg-$guest--root" &&
   dev_boot="/dev/mapper/$vg-$guest--boot" &&
   echo "Root device will be $dev_root" &&
   echo "Boot device will be $dev_boot" &&
   fs=ext4
 fi &&
 if [ "$vg" != "" ]; then
   dev_swap="/dev/mapper/$vg-$guest--swap"
 elif [ -d /SWAPFILES ]; then
   dev_swap=/SWAPFILES/$guest-swap
 else
   echo "No idea how to do swap" && false
 fi &&
 echo -n "Use a /boot filesystem (strongly recommended)? [Y/n] " &&
 read use_boot &&
 echo "Swap device will be $dev_swap" &&
 echo "fs is $fs"
 echo "Chosen mirror is $mirror"
}

And run it

setup_env

Then install base:

The way we lay out the filesystems by default is that we have one 4g / filesystem, a swap, and a tiny boot filesystem. On the host we make a new LVM logical volume for each of the three. Only the LV that will take the guest's boot will actually be partitioned - into a single boot partition. That's so we can install grub into the MBR and have the system start just like a real system. Root and swap are put directly onto the logical volume, without partitioning it at all. This makes getting to the data from the host easier - no need to run kpartx - and it makes growing trivial.

#######
# install base

 apt-get install debootstrap debian-archive-keyring kpartx &&
 if [ "$use_lvm" = "y" ]; then
   if [ "$use_boot" != "n" ] ; then
     lvcreate -L 128m -n "$guest"-boot /dev/"$vg"
   fi &&
   lvcreate -L 4g -n "$guest"-root /dev/"$vg"
 fi &&
 lvcreate -L 4g -n "$guest"-swap /dev/"$vg" &&
 : &&
 if [ "$use_boot" != "n" ] ; then
   ( echo ',,L,*' | sfdisk -D "$dev_boot" ) &&
   kpartx -v -a "$dev_boot" &&
   mkfs."$fs" "$dev_boot"1
 fi &&
 mkfs."$fs" "$dev_root" &&
 mkswap "$dev_swap" &&
 : &&
 mkdir -p "$target" &&
 mount "$dev_root" "$target" &&
 if [ "$use_boot" != "n" ] ; then
   mkdir -p "$target/boot" &&
   mount "$dev_boot"1 "$target/boot"
 fi &&

 cd "$target" &&
 debootstrap --variant=minbase --keyring=/usr/share/keyrings/debian-archive-keyring.gpg stable . "$mirror"

And finalize the setup:

#######
# finish setup
 echo "$guest" > etc/hostname &&
 cat > etc/hosts << EOF &&
127.0.0.1       localhost

# The following lines are desirable for IPv6 capable hosts
::1     localhost ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts
EOF
 rm -fv etc/apt/sources.list &&
 ( ! [ -e /etc/apt/sources.list ] || cp /etc/apt/sources.list etc/apt/sources.list)
 (cp -v /etc/apt/sources.list.d/* etc/apt/sources.list.d/ || true ) &&
 ( ! [ -e /etc/apt/preferences ] || cp -v /etc/apt/preferences etc/apt/ ) &&
 apt-key exportall | chroot . apt-key add - &&
 chroot . apt-get update &&
 echo "Apt::Install-Recommends 0;" > etc/apt/apt.conf.d/local-recommends &&
 chroot . apt-get install -y net-tools iproute ifupdown dialog vim netbase xfsprogs &&
 cp -av `readlink -f $dev_root` dev/`basename $dev_root` &&
 DEBIAN_FRONTEND=noninteractive chroot . apt-get install -y grub2 &&
  cat > etc/kernel-img.conf << EOF &&
do_symlinks = no
do_initrd = yes
EOF
  if [ "$use_boot" != "n" ] ; then
    cp -av `readlink -f $dev_boot` dev/`basename $dev_boot` &&
    cp -av `readlink -f $dev_boot""1` dev/`basename $dev_boot`1 &&
    chroot . grub-install --modules=part_msdos /dev/`basename $dev_boot` &&
    # install a kernel image
    chroot . apt-get install -y linux-image-2.6-amd64 &&
    sed -i -e 's/^#GRUB_TERMINAL=console/GRUB_TERMINAL=console/' etc/default/grub &&
    echo "(hd0) /dev/`basename $dev_boot`" > boot/grub/device.map &&
    chroot . update-grub &&
    sed -i -e "s#dev/`basename $dev_boot`1#dev/hda1#g" boot/grub/grub.cfg &&
    rm -v dev/"`basename $dev_boot`" dev/"`basename $dev_boot`1"
  else
    echo && echo && echo && echo "Hardly tested, expect this to fail." && echo && echo && echo &&
    echo "(hd0) /dev/`basename $dev_root`" > boot/grub/device.map &&
    chroot . grub-install /dev/"`basename $dev_root`" &&
    # install a kernel image
    chroot . apt-get install -y linux-image-2.6-amd64 &&
    sed -i -e 's/^#GRUB_TERMINAL=console/GRUB_TERMINAL=console/' etc/default/grub &&
    chroot . update-grub
  fi &&
  sed -i -e "s#dev/`basename $dev_root`#dev/vda#g" boot/grub/grub.cfg &&
  rm -v boot/grub/device.map &&
  rm -v dev/"`basename $dev_root`"

And a fstab and a boot loader config

 # doesn't work: chroot . update-grub
 rootuuid=`blkid -s UUID -o value "$dev_root"` &&
 swapuuid=`blkid -s UUID -o value "$dev_swap"` &&
 if [ "$fs" = "ext4" ]; then
   rootopts="errors=remount-ro"
 else
   rootopts="defaults"
 fi
 cat > etc/fstab << EOF &&
UUID=$rootuuid    /               $fs    $rootopts 0       1
UUID=$swapuuid    none            swap    sw              0       0
EOF
 if [ "$use_boot" != "n" ] ; then
    bootuuid=`blkid -s UUID -o value "$dev_boot"1` &&
    echo "UUID=$bootuuid    /boot           $fs    defaults        0       2" >> etc/fstab
 fi
 cat > etc/network/interfaces << EOF
auto lo
iface lo inet loopback

auto eth0
iface eth0 inet manual
  pre-up ip link set up dev \$IFACE
  post-down ip link set down dev \$IFACE
EOF

Maybe fix/setup networking properly:

vi etc/network/interfaces

And set a password:

chroot . passwd

Set a nameserver config that works once the VM has booted. Later in the process we will install unbound anyway.

cat > etc/resolv.conf << EOF
nameserver 8.8.8.8
search debian.org
EOF

And unmount:

  cd / &&
 if [ "$use_boot" != "n" ] ; then
   umount "$target"/boot &&
   kpartx -v -d "$dev_boot"
 fi &&
 umount "$target" &&
 rmdir "$target"

virsh setup

Setup a new kvm domain by creating a new file in /etc/dsa-kvm/`hostname`/$guest.xml.

Properly configure hostname
Pick a new uuid (uuidgen)
Setup block devices properly
pick a new and unique mac address (on d.o every kvm host has their own mac address space and the last block is changed for the guests, as in ..:..:..:..:<host byte>:<guest byte>. )
virsh commands:

virsh help
virsh define foo.xml
virsh start foo
virsh destroy foo
virsh vncdisplay foo (and ssh -L 5900:localhost:<5900+x> $host and vnc localhost)

post processing

Do not forget to set a sane root password before installing ssh in the new kvm domain.

when stuff goes wrong

To get to the guest data from the host:

  setup_env
 kpartx -v -a "$dev_boot" &&
 mkdir -p "$target" &&
 mount "$dev_root" "$target" &&
 mkdir -p "$target/boot" &&
 mount "$dev_boot"1 "$target/boot"

and once you're done:

 cd / &&
 umount "$target"/boot &&
 umount "$target" &&
 kpartx -v -d "$dev_boot" &&
 rmdir "$target"

Make sure that the filesystem isn't being mounted twice - i.e. never start the guest while the filesystems are mounted, and never mount them while the guest is running.