nixpkgs-suyu/modules/virtualisation/qemu-vm.nix
Eelco Dolstra ecaf1d9f08 * Using hpet no longer seems necessary. Maybe upstream fixed it.
svn path=/nixos/trunk/; revision=25489
2011-01-10 13:32:09 +00:00

401 lines
13 KiB
Nix

# This module creates a virtual machine from the NixOS configuration.
# Building the `config.system.build.vm' attribute gives you a command
# that starts a KVM/QEMU VM running the NixOS configuration defined in
# `config'. The Nix store is shared read-only with the host, which
# makes (re)building VMs very efficient. However, it also means you
# can't reconfigure the guest inside the guest - you need to rebuild
# the VM in the host. On the other hand, the root filesystem is a
# read/writable disk image persistent across VM reboots.
{ config, pkgs, ... }:
with pkgs.lib;
let
vmName = config.networking.hostName;
options = {
virtualisation.memorySize =
mkOption {
default = 384;
description =
''
Memory size (M) of virtual machine.
'';
};
virtualisation.diskSize =
mkOption {
default = 512;
description =
''
Disk size (M) of virtual machine.
'';
};
virtualisation.diskImage =
mkOption {
default = "./${vmName}.qcow2";
description =
''
Path to the disk image containing the root filesystem.
The image will be created on startup if it does not
exist.
'';
};
virtualisation.graphics =
mkOption {
default = true;
description =
''
Whether to run QEMU with a graphics window, or access
the guest computer serial port through the host tty.
'';
};
virtualisation.pathsInNixDB =
mkOption {
default = [];
description =
''
The list of paths whose closure is registered in the Nix
database in the VM. All other paths in the host Nix store
appear in the guest Nix store as well, but are considered
garbage (because they are not registered in the Nix
database in the guest).
'';
};
virtualisation.vlans =
mkOption {
default = [ 1 ];
example = [ 1 2 ];
description =
''
Virtual networks to which the VM is connected. Each
number <replaceable>N</replaceable> in this list causes
the VM to have a virtual Ethernet interface attached to a
separate virtual network on which it will be assigned IP
address
<literal>192.168.<replaceable>N</replaceable>.<replaceable>M</replaceable></literal>,
where <replaceable>M</replaceable> is the index of this VM
in the list of VMs.
'';
};
virtualisation.writableStore =
mkOption {
default = false;
description =
''
If enabled, the Nix store in the VM is made writable by
layering an AUFS/tmpfs filesystem on top of the host's Nix
store.
'';
};
networking.primaryIPAddress =
mkOption {
default = "";
internal = true;
description = "Primary IP address used in /etc/hosts.";
};
virtualisation.qemu.options =
mkOption {
default = "";
example = "-vga std";
description = "Options passed to QEMU.";
};
virtualisation.useBackdoor =
mkOption {
default = false;
description =
''
If enabled, the virtual machine makes a connection through TCP port 23
to a daemon running on the host system acting as a proxy.
This option makes it possible to connect to a VM through a socket file.
'';
};
virtualisation.useBootLoader =
mkOption {
default = false;
description =
''
If enabled, the virtual machine will be booted using the
regular boot loader (i.e., GRUB 1 or 2). This allows
testing of the boot loader. If
disabled (the default), the VM directly boots the NixOS
kernel and initial ramdisk, bypassing the boot loader
altogether.
'';
};
};
cfg = config.virtualisation;
qemuGraphics = if cfg.graphics then "" else "-nographic";
kernelConsole = if cfg.graphics then "" else "console=ttyS0";
ttys = [ "tty1" "tty2" "tty3" "tty4" "tty5" "tty6" ];
# Shell script to start the VM.
startVM =
''
#! ${pkgs.stdenv.shell}
NIX_DISK_IMAGE=$(readlink -f ''${NIX_DISK_IMAGE:-${config.virtualisation.diskImage}})
if ! test -e "$NIX_DISK_IMAGE"; then
${pkgs.qemu_kvm}/bin/qemu-img create -f qcow2 "$NIX_DISK_IMAGE" \
${toString config.virtualisation.diskSize}M || exit 1
fi
${pkgs.lib.optionalString cfg.useBackdoor ''
# Remember the current working directory
WORKDIR=$(pwd)
''}
# Start Samba (which wants to put its socket and config files in TMPDIR).
if [ -z "$TMPDIR" -o -z "$USE_TMPDIR" ]; then
TMPDIR=$(mktemp -d nix-vm-smbd.XXXXXXXXXX --tmpdir)
fi
cd $TMPDIR
${pkgs.vmTools.startSamba}
${pkgs.lib.optionalString cfg.useBackdoor ''
# Create a shell socket file to which the VM can connect and create in the
# current working directory a socket file which can be used to remotely access
# the VM through the shell interface
${pkgs.socat}/bin/socat UNIX-LISTEN:./shell UNIX-LISTEN:$WORKDIR/${vmName}.socket,fork &
while [ ! -e ./shell ]; do sleep 0.1; done # Wait until the socket file is there
''}
# Start QEMU.
exec ${pkgs.qemu_kvm}/bin/qemu-system-x86_64 \
-name ${vmName} \
-m ${toString config.virtualisation.memorySize} \
-net nic,vlan=0,model=virtio \
-chardev socket,id=samba,path=./samba \
-net user,vlan=0,guestfwd=tcp:10.0.2.4:445-chardev:samba${if cfg.useBackdoor then ",guestfwd=tcp:10.0.2.6:23-chardev:shell" else ""}''${QEMU_NET_OPTS:+,$QEMU_NET_OPTS} \
${if cfg.useBackdoor then "-chardev socket,id=shell,path=./shell" else ""} \
${if cfg.useBootLoader then ''
-drive index=0,file=$NIX_DISK_IMAGE,if=virtio,cache=writeback,werror=report \
-drive index=1,file=${bootDisk}/disk.img,if=virtio,boot=on \
'' else ''
-drive file=$NIX_DISK_IMAGE,if=virtio,boot=on,cache=writeback,werror=report \
-kernel ${config.system.build.toplevel}/kernel \
-initrd ${config.system.build.toplevel}/initrd \
-append "$(cat ${config.system.build.toplevel}/kernel-params) init=${config.system.build.toplevel}/init regInfo=${regInfo} ${kernelConsole} $QEMU_KERNEL_PARAMS" \
''} \
${qemuGraphics} \
$QEMU_OPTS \
${config.virtualisation.qemu.options}
'';
regInfo = pkgs.runCommand "reginfo"
{ exportReferencesGraph =
map (x: [("closure-" + baseNameOf x) x]) config.virtualisation.pathsInNixDB;
buildInputs = [ pkgs.perl ];
}
''
printRegistration=1 perl ${pkgs.pathsFromGraph} closure-* > $out
'';
# Generate a hard disk image containing a /boot partition and GRUB
# in the MBR. Used when the `useBootLoader' option is set.
bootDisk =
pkgs.vmTools.runInLinuxVM (
pkgs.runCommand "nixos-boot-disk"
{ preVM =
''
mkdir $out
diskImage=$out/disk.img
${pkgs.vmTools.kvm}/bin/qemu-img create -f qcow2 $diskImage "32M"
'';
buildInputs = [ pkgs.utillinux ];
}
''
# Create a single /boot partition.
${pkgs.parted}/sbin/parted /dev/vda mklabel msdos
${pkgs.parted}/sbin/parted /dev/vda -- mkpart primary ext2 1M -1s
. /sys/class/block/vda1/uevent
mknod /dev/vda1 b $MAJOR $MINOR
. /sys/class/block/vda/uevent
${pkgs.e2fsprogs}/sbin/mkfs.ext3 -L boot /dev/vda1
${pkgs.e2fsprogs}/sbin/tune2fs -c 0 -i 0 /dev/vda1
# Mount /boot.
mkdir /boot
mount /dev/vda1 /boot
# This is needed for GRUB 0.97, which doesn't know about virtio devices.
mkdir /boot/grub
echo '(hd0) /dev/vda' > /boot/grub/device.map
# Install GRUB and generate the GRUB boot menu.
touch /etc/NIXOS
mkdir -p /nix/var/nix/profiles
${config.system.build.toplevel}/bin/switch-to-configuration boot
umount /boot
''
);
in
{
require = options;
boot.loader.grub.device = mkOverride 50 "/dev/vda";
# All the modules the initrd needs to mount the host filesystem via
# CIFS. Also use paravirtualised network and block devices for
# performance.
boot.initrd.availableKernelModules =
[ "cifs" "virtio_net" "virtio_pci" "virtio_blk" "virtio_balloon" "nls_utf8" ]
++ optional cfg.writableStore [ "aufs" ];
boot.extraModulePackages =
optional cfg.writableStore config.boot.kernelPackages.aufs2;
boot.initrd.extraUtilsCommands =
''
# We need mke2fs in the initrd.
cp ${pkgs.e2fsprogs}/sbin/mke2fs $out/bin
# And `ifconfig'.
cp ${pkgs.nettools}/sbin/ifconfig $out/bin
'';
boot.initrd.postDeviceCommands =
''
# Set up networking. Needed for CIFS mounting.
ifconfig eth0 up 10.0.2.15
# If the disk image appears to be empty, run mke2fs to
# initialise.
FSTYPE=$(blkid -o value -s TYPE /dev/vda || true)
if test -z "$FSTYPE"; then
mke2fs -t ext3 /dev/vda
fi
'';
boot.initrd.postMountCommands =
''
mkdir -p $targetRoot/boot
mount -o remount,ro $targetRoot/nix/store
${optionalString cfg.writableStore ''
mkdir /mnt-store-tmpfs
mount -t tmpfs -o "mode=755" none /mnt-store-tmpfs
mount -t aufs -o dirs=/mnt-store-tmpfs=rw:$targetRoot/nix/store=rr none $targetRoot/nix/store
''}
'';
# After booting, register the closure of the paths in
# `virtualisation.pathsInNixDB' in the Nix database in the VM. This
# allows Nix operations to work in the VM. The path to the
# registration file is passed through the kernel command line to
# allow `system.build.toplevel' to be included. (If we had a direct
# reference to ${regInfo} here, then we would get a cyclic
# dependency.)
boot.postBootCommands =
''
( source /proc/cmdline
if [ -n "$regInfo" ]; then
${config.environment.nix}/bin/nix-store --load-db < $regInfo
fi
)
'';
virtualisation.pathsInNixDB = [ config.system.build.toplevel ];
# Mount the host filesystem via CIFS, and bind-mount the Nix store
# of the host into our own filesystem. We use mkOverride to allow
# this module to be applied to "normal" NixOS system configuration,
# where the regular value for the `fileSystems' attribute should be
# disregarded for the purpose of building a VM test image (since
# those filesystems don't exist in the VM).
fileSystems = mkOverride 50 (
[ { mountPoint = "/";
device = "/dev/vda";
}
{ mountPoint = "/hostfs";
device = "//10.0.2.4/qemu";
fsType = "cifs";
options = "guest,sec=none,noperm,noacl";
neededForBoot = true;
}
{ mountPoint = "/nix/store";
device = "/hostfs/nix/store";
fsType = "none";
options = "bind";
neededForBoot = true;
}
] ++ optional cfg.useBootLoader
{ mountPoint = "/boot";
device = "/dev/disk/by-label/boot";
fsType = "ext3";
options = "ro";
noCheck = true; # fsck fails on a r/o filesystem
});
swapDevices = mkOverride 50 [ ];
# Starting DHCP brings down eth0, which kills the connection to the
# host filesystem and thus deadlocks the system.
networking.useDHCP = false;
networking.defaultGateway = mkOverride 200 "10.0.2.2";
networking.nameservers = [ "10.0.2.3" ];
networking.interfaces = singleton
{ name = "eth0";
ipAddress = "10.0.2.15";
};
# Don't run ntpd in the guest. It should get the correct time from KVM.
services.ntp.enable = false;
system.build.vm = pkgs.runCommand "nixos-vm" {}
''
ensureDir $out/bin
ln -s ${config.system.build.toplevel} $out/system
ln -s ${pkgs.writeScript "run-nixos-vm" startVM} $out/bin/run-${vmName}-vm
'';
# sendfile() is currently broken over CIFS, so fix it here for all
# configurations that use Apache.
services.httpd.extraConfig =
''
EnableSendFile Off
'';
# When building a regular system configuration, override whatever
# video driver the host uses.
services.xserver.videoDriver = mkOverride 50 null;
services.xserver.videoDrivers = mkOverride 50 [ "cirrus" "vesa" ];
services.xserver.defaultDepth = mkOverride 50 0;
services.xserver.monitorSection =
''
# Set a higher refresh rate so that resolutions > 800x600 work.
HorizSync 30-140
VertRefresh 50-160
'';
services.mingetty.ttys = ttys ++ optional (!cfg.graphics) "ttyS0";
# Wireless won't work in the VM.
networking.enableWLAN = mkOverride 50 false;
}