Annotation of wikisrc/ports/xen/howto.mdwn, revision 1.22

1.5       mspo        1: Introduction
1.13      gdt         2: ============
1.1       mspo        3: 
                      4: [![[Xen
1.7       mspo        5: screenshot]](http://www.netbsd.org/gallery/in-Action/hubertf-xens.png)](../../gallery/in-Action/hubertf-xen.png)
1.1       mspo        6: 
1.12      gdt         7: Xen is a virtual machine monitor or hypervisor for x86 hardware
                      8: (i686-class or higher), which supports running multiple guest
                      9: operating systems on a single physical machine.  With Xen, one uses
                     10: the Xen kernel to control the CPU, memory and console, a dom0
                     11: operating system which mediates access to other hardware (e.g., disks,
                     12: network, USB), and one or more domU operating systems which operate in
                     13: an unprivileged virtualized environment.  IO requests from the domU
                     14: systems are forwarded by the hypervisor (Xen) to the dom0 to be
                     15: fulfilled.
                     16: 
                     17: Xen supports two styles of guests.  The original is Para-Virtualized
                     18: (PV) which means that the guest OS does not attempt to access hardware
                     19: directly, but instead makes hypercalls to the hypervisor.  This is
                     20: analogous to a user-space program making system calls.  (The dom0
                     21: operating system uses PV calls for some functions, such as updating
                     22: memory mapping page tables, but has direct hardware access for disk
                     23: and network.)   PV guests must be specifically coded for Xen.
                     24: 
                     25: The more recent style is HVM, which means that the guest does not have
                     26: code for Xen and need not be aware that it is running under Xen.
                     27: Attempts to access hardware registers are trapped and emulated.  This
                     28: style is less efficient but can run unmodified guests.
                     29: 
1.19      gdt        30: Generally any amd64 machine will work with Xen and PV guests.  For HVM
1.20      gdt        31: guests, the VT or VMX cpu feature (Intel) or SVM/HVM/VT (amd64) is
                     32: needed; "cpuctl identify 0" will show this.  TODO: Clean up and check
                     33: the above features.  TODO: Explain if i386 (non-amd64) machines can
                     34: still be used - I think that witthe requirement to use PAE kernels is
                     35: about the hypervisor being amd64 only.
1.19      gdt        36: 
1.12      gdt        37: At boot, the dom0 kernel is loaded as module with Xen as the kernel.
                     38: The dom0 can start one or more domUs.  (Booting is explained in detail
                     39: in the dom0 section.)
                     40: 
                     41: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
                     42: and that Xen kernels and tools are available in pkgsrc.  This HOWTO
                     43: attempts to address both the case of running a NetBSD dom0 on hardware
                     44: and running NetBSD as a domU in a VPS.
                     45: 
1.20      gdt        46: Some versions of Xen support "PCI passthrough", which means that
                     47: specific PCI devices can be made available to a specific domU instead
                     48: of the dom0.  This can be useful to let a domU run X11, or access some
                     49: network interface or other peripheral.
                     50: 
1.12      gdt        51: Prerequisites
1.13      gdt        52: -------------
1.12      gdt        53: 
                     54: Installing NetBSD/Xen is not extremely difficult, but it is more
                     55: complex than a normal installation of NetBSD.
1.15      gdt        56: In general, this HOWTO is occasionally overly restrictive about how
                     57: things must be done, guiding the reader to stay on the established
                     58: path when there are no known good reasons to stray.
1.12      gdt        59: 
                     60: This HOWTO presumes a basic familiarity with the Xen system
1.16      gdt        61: architecture.  This HOWTO presumes familiarity with installing NetBSD
                     62: on i386/amd64 hardware and installing software from pkgsrc.
                     63: See also the [Xen website](http://www.xen.org/).
1.1       mspo       64: 
1.19      gdt        65: History
                     66: -------
                     67: 
                     68: NetBSD used to support Xen2; this has been removed.
                     69: 
                     70: Before NetBSD's native bootloader could support Xen, the use of
                     71: grub was recommended.  If necessary, see the
                     72: [old grub information](/xen/howto-grub/).
                     73: 
1.15      gdt        74: Versions of Xen and NetBSD
                     75: ==========================
                     76: 
                     77: Most of the installation concepts and instructions are independent of
                     78: Xen version.  This section gives advice on which version to choose.
                     79: Versions not in pkgsrc and older unsupported versions of NetBSD are
                     80: inentionally ignored.
                     81: 
                     82: Xen
                     83: ---
                     84: 
                     85: In NetBSD, xen is provided in pkgsrc, via matching pairs of packages
                     86: xenkernel and xentools.  We will refer only to the kernel versions,
                     87: but note that both packages must be installed together and must have
                     88: matching versions.
                     89: 
                     90: xenkernel3 and xenkernel33 provide Xen 3.1 and 3.3.  These no longer
1.20      gdt        91: receive security patches and should not be used.  Xen 3.1 supports PCI
                     92: passthrough.
1.15      gdt        93: 
                     94: xenkernel41 provides Xen 4.1.  This is no longer maintained by Xen,
                     95: but as of 2014-12 receives backported security patches.  It is a
                     96: reasonable although trailing-edge choice.
                     97: 
                     98: xenkernel42 provides Xen 4.2.  This is maintained by Xen, but old as
                     99: of 2014-12.
                    100: 
                    101: Ideally newer versions of Xen will be added to pkgsrc.
                    102: 
1.20      gdt       103: Note that NetBSD support is called XEN3; it works with 3.1 through
                    104: 4.2, because the hypercall interface has been stable.
                    105: 
1.19      gdt       106: Xen command program
                    107: -------------------
                    108: 
                    109: Early Xen used a program called "xm" to manipulate the system from the
                    110: dom0.  Starting in 4.1, a replacement program with similar behavior
                    111: called "xl" is provided.  In 4.2, "xm" is no longer available.
                    112: 
1.15      gdt       113: NetBSD
                    114: ------
                    115: 
                    116: The netbsd-5, netbsd-6, netbsd-7, and -current branches are all
                    117: reasonable choices, with more or less the same considerations for
                    118: non-Xen use.  Therefore, netbsd-6 is recommended as the stable version
                    119: of the most recent release.
                    120: 
                    121: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
                    122: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
                    123: need SMP; the lack of support is really a problem when using a dom0 as
                    124: a normal computer.)
                    125: 
1.18      gdt       126: Architecture
                    127: ------------
                    128: 
                    129: Xen is basically amd64 only at this point.  One can either run i386
                    130: domains or amd64 domains.  If running i386, PAE versions are required,
                    131: for both dom0 and domU.  These versions are built by default in NetBSD
                    132: releases.  While i386 dom0 works fine, amd64 is recommended as more
                    133: normal.  (Note that emacs (at least) fails if run on i386 with PAE when
                    134: built without, and vice versa, presumably due to bugs in the undump
                    135: code.)
                    136: 
1.15      gdt       137: Recommendation
                    138: --------------
                    139: 
1.18      gdt       140: Therefore, this HOWTO recommends running xenkernel42 (and xentools42),
1.19      gdt       141: xl, the NetBSD 6 stable branch, and to use amd64 as the dom0.  Either
                    142: the i386 or amd64 of NetBSD may be used as domUs.
1.15      gdt       143: 
                    144: NetBSD as a dom0
                    145: ================
                    146: 
                    147: NetBSD can be used as a dom0 and works very well.  The following
                    148: sections address installation, updating NetBSD, and updating Xen.
1.19      gdt       149: Note that it doesn't make sense to talk about installing a dom0 OS
                    150: without also installing Xen itself.  We first address installing
                    151: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
                    152: NetBSD install to a dom0 install by just changing the kernel and boot
                    153: configuration.
1.15      gdt       154: 
                    155: Styles of dom0 operation
                    156: ------------------------
                    157: 
                    158: There are two basic ways to use Xen.  The traditional method is for
                    159: the dom0 to do absolutely nothing other than providing support to some
                    160: number of domUs.  Such a system was probably installed for the sole
                    161: purpose of hosting domUs, and sits in a server room on a UPS.
                    162: 
                    163: The other way is to put Xen under a normal-usage computer, so that the
                    164: dom0 is what the computer would have been without Xen, perhaps a
                    165: desktop or laptop.  Then, one can run domUs at will.  Purists will
                    166: deride this as less secure than the previous approach, and for a
                    167: computer whose purpose is to run domUs, they are right.  But Xen and a
                    168: dom0 (without domUs) is not meaingfully less secure than the same
                    169: things running without Xen.  One can boot Xen or boot regular NetBSD
                    170: alternately with little problems, simply refraining from starting the
                    171: Xen daemons when not running Xen.
                    172: 
                    173: Note that NetBSD as dom0 does not support multiple CPUs.  This will
                    174: limit the performance of the Xen/dom0 workstation approach.
                    175: 
1.19      gdt       176: Installation of NetBSD
                    177: ----------------------
1.13      gdt       178: 
1.19      gdt       179: First,
                    180: [install NetBSD/amd64](../../docs/guide/en/chap-inst.html)
                    181: just as you would if you were not using Xen.
                    182: However, the partitioning approach is very important.
                    183: 
                    184: If you want to use RAIDframe for the dom0, there are no special issues
                    185: for Xen.  Typically one provides RAID storage for the dom0, and the
1.22    ! gdt       186: domU systems are unaware of RAID.  The 2nd-stage loader bootxx_* skips
        !           187: over a RAID1 header to find /boot from a filesystem within a RAID
        !           188: partition; this is no different when booting Xen.
1.19      gdt       189: 
                    190: There are 4 styles of providing backing storage for the virtual disks
                    191: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN,
                    192: 
                    193: With raw partitions, one has a disklabel (or gpt) partition sized for
                    194: each virtual disk to be used by the domU.  (If you are able to predict
                    195: how domU usage will evolve, please add an explanation to the HOWTO.
                    196: Seriously, needs tend to change over time.)
                    197: 
                    198: One can use lvm(8) to create logical devices to use for domU disks.
                    199: This is almost as efficient sa raw disk partitions and more flexible.
                    200: Hence raw disk partitions should typically not be used.
                    201: 
                    202: One can use files in the dom0 filesystem, typically created by dd'ing
                    203: /dev/zero to create a specific size.  This is somewhat less efficient,
                    204: but very convenient, as one can cp the files for backup, or move them
                    205: between dom0 hosts.
                    206: 
                    207: Finally, in theory one can place the files backing the domU disks in a
                    208: SAN.  (This is an invitation for someone who has done this to add a
                    209: HOWTO page.)
1.1       mspo      210: 
1.19      gdt       211: Installation of Xen
                    212: -------------------
1.1       mspo      213: 
1.20      gdt       214: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
                    215: pkgsrc (or another matching pair).
                    216: See [the pkgsrc
                    217: documentation](http://www.NetBSD.org/docs/pkgsrc/) for help with pkgsrc.
                    218: 
                    219: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm.  More
                    220: recent versions have HVM support integrated in the main xentools
                    221: package.  It is entirely reasonable to run only PV guests.
                    222: 
                    223: Next you need to install the selected Xen kernel itself, which is
                    224: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz".  Copy it to /.
                    225: For debugging, one may copy xen-debug.gz; this is conceptually similar
                    226: to DIAGNOSTIC and DEBUG in NetBSD.  xen-debug.gz is basically only
                    227: useful with a serial console.  Then, place a NetBSD XEN3_DOM0 kernel
                    228: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
                    229: of a NetBSD build.  Both xen and NetBSD may be left compressed.  (If
                    230: using i386, use releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz.)
                    231: 
                    232: In a dom0 kernel, kernfs is mandatory for xend to comunicate with the
                    233: kernel, so ensure that /kern is in fstab.
                    234: 
                    235: Because you already installed NetBSD, you have a working boot setup
                    236: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
                    237: beginning of your root filesystem, /boot present, and likely
                    238: /boot.cfg.  (If not, fix before continuing!)
                    239: 
                    240: See boot.cfg(5) for an example.  The basic line is
                    241: 
                    242: "menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M"
                    243: 
                    244: which specifies that the dom0 should have 256M, leaving the rest to be
                    245: allocated for domUs.
                    246: 
                    247: As with non-Xen systems, you should have a line to boot /netbsd (a
                    248: kernel that works without Xen) and fallback versions of the non-Xen
                    249: kernel, Xen, and the dom0 kernel.
1.1       mspo      250: 
1.21      gdt       251: Configuring Xen
                    252: ---------------
                    253: 
                    254: Now, you have a system that will boot Xen and the dom0 kernel, and
                    255: just run the dom0 kernel.  There will be no domUs, and none can be
                    256: started because you still have to configure the dom0 tools.
                    257: 
                    258: For 3.3 (and probably 3.1), add to rc.conf (but note that you should
                    259: have installed 4.2):
                    260:   xend=YES
                    261:   xenbackendd=YES
                    262: 
                    263: For 4.1 and 4.2, add to rc.conf:
                    264:   xend=YES
                    265:   xencommons=YES
                    266: 
1.15      gdt       267: Updating NetBSD in a dom0
                    268: -------------------------
                    269: 
                    270: This is just like updating NetBSD on bare hardware, assuming the new
                    271: version supports the version of Xen you are running.  Generally, one
                    272: replaces the kernel and reboots, and then overlays userland binaries
                    273: and adjusts /etc.
                    274: 
                    275: Note that one must update both the non-Xen kernel typically used for
                    276: rescue purposes and the DOM0 kernel used with Xen.
                    277: 
1.22    ! gdt       278: To convert from grub to /boot, install an mbr bootblock with fdisk,
        !           279: bootxx_ with installboot, /boot and /boot.cfg.  This really should be
        !           280: no different than completely reinstalling boot blocks on a non-Xen
        !           281: system.
        !           282: 
1.15      gdt       283: Updating Xen versions
                    284: ---------------------
                    285: 
1.21      gdt       286: Updating Xen is conceptually not difficult, but can run into all the
                    287: issues found when installing Xen.  Assuming migration from 4.1 to 4.2,
                    288: remove the xenkernel41 and xentools41 packages and install the
                    289: xenkernel42 and xentools42 packages.  Copy the 4.2 xen.gz to /.
                    290: 
                    291: Ensure that the contents of /etc/rc.d/xen* are correct.  Enable the
                    292: correct set of daemons.  Ensure that the domU config files are valid
                    293: for the new version.
1.15      gdt       294: 
1.14      gdt       295: Creating unprivileged domains (domU)
                    296: ====================================
                    297: 
                    298: Creating domUs is almost entirely independent of operating system.  We
                    299: first explain NetBSD, and then differences for Linux and Solaris.
                    300: 
                    301: Creating an unprivileged NetBSD domain (domU)
                    302: ---------------------------------------------
1.1       mspo      303: 
                    304: Once you have *domain0* running, you need to start the xen tool daemon
1.5       mspo      305: (`/usr/pkg/share/examples/rc.d/xend start`) and the xen backend daemon
                    306: (`/usr/pkg/share/examples/rc.d/xenbackendd start` for Xen3\*,
                    307: `/usr/pkg/share/examples/rc.d/xencommons start` for Xen4.\*). Make sure
                    308: that `/dev/xencons` and `/dev/xenevt` exist before starting `xend`. You
                    309: can create them with this command:
1.1       mspo      310: 
1.3       mspo      311:     # cd /dev && sh MAKEDEV xen
1.1       mspo      312: 
1.5       mspo      313: xend will write logs to `/var/log/xend.log` and
                    314: `/var/log/xend-debug.log`. You can then control xen with the xm tool.
                    315: 'xm list' will show something like:
1.1       mspo      316: 
1.3       mspo      317:     # xm list
                    318:     Name              Id  Mem(MB)  CPU  State  Time(s)  Console
                    319:     Domain-0           0       64    0  r----     58.1
1.1       mspo      320: 
                    321: 'xm create' allows you to create a new domain. It uses a config file in
                    322: PKG\_SYSCONFDIR for its parameters. By default, this file will be in
1.5       mspo      323: `/usr/pkg/etc/xen/`. On creation, a kernel has to be specified, which
                    324: will be executed in the new domain (this kernel is in the *domain0* file
                    325: system, not on the new domain virtual disk; but please note, you should
                    326: install the same kernel into *domainU* as `/netbsd` in order to make
                    327: your system tools, like MAN.SAVECORE.8, work). A suitable kernel is
                    328: provided as part of the i386 and amd64 binary sets: XEN3\_DOMU.
1.1       mspo      329: 
                    330: Here is an /usr/pkg/etc/xen/nbsd example config file:
                    331: 
1.3       mspo      332:     #  -*- mode: python; -*-
                    333:     #============================================================================
                    334:     # Python defaults setup for 'xm create'.
                    335:     # Edit this file to reflect the configuration of your system.
                    336:     #============================================================================
1.5       mspo      337: 
1.3       mspo      338:     #----------------------------------------------------------------------------
                    339:     # Kernel image file. This kernel will be loaded in the new domain.
                    340:     kernel = "/home/bouyer/netbsd-XEN3_DOMU"
                    341:     #kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
1.5       mspo      342: 
1.3       mspo      343:     # Memory allocation (in megabytes) for the new domain.
                    344:     memory = 128
1.5       mspo      345: 
1.3       mspo      346:     # A handy name for your new domain. This will appear in 'xm list',
                    347:     # and you can use this as parameters for xm in place of the domain
                    348:     # number. All domains must have different names.
                    349:     #
                    350:     name = "nbsd"
1.5       mspo      351: 
1.3       mspo      352:     # The number of virtual CPUs this domain has.
                    353:     #
                    354:     vcpus = 1
1.5       mspo      355: 
1.3       mspo      356:     #----------------------------------------------------------------------------
                    357:     # Define network interfaces for the new domain.
1.5       mspo      358: 
1.3       mspo      359:     # Number of network interfaces (must be at least 1). Default is 1.
                    360:     nics = 1
1.5       mspo      361: 
1.3       mspo      362:     # Define MAC and/or bridge for the network interfaces.
                    363:     #
                    364:     # The MAC address specified in ``mac'' is the one used for the interface
                    365:     # in the new domain. The interface in domain0 will use this address XOR'd
                    366:     # with 00:00:00:01:00:00 (i.e. aa:00:00:51:02:f0 in our example). Random
                    367:     # MACs are assigned if not given.
                    368:     #
                    369:     # ``bridge'' is a required parameter, which will be passed to the
                    370:     # vif-script called by xend(8) when a new domain is created to configure
                    371:     # the new xvif interface in domain0.
                    372:     #
                    373:     # In this example, the xvif is added to bridge0, which should have been
                    374:     # set up prior to the new domain being created -- either in the
                    375:     # ``network'' script or using a /etc/ifconfig.bridge0 file.
                    376:     #
                    377:     vif = [ 'mac=aa:00:00:50:02:f0, bridge=bridge0' ]
1.5       mspo      378: 
1.3       mspo      379:     #----------------------------------------------------------------------------
                    380:     # Define the disk devices you want the domain to have access to, and
                    381:     # what you want them accessible as.
                    382:     #
                    383:     # Each disk entry is of the form:
                    384:     #
1.5       mspo      385:     #   phy:DEV,VDEV,MODE
1.3       mspo      386:     #
                    387:     # where DEV is the device, VDEV is the device name the domain will see,
                    388:     # and MODE is r for read-only, w for read-write.  You can also create
                    389:     # file-backed domains using disk entries of the form:
                    390:     #
1.5       mspo      391:     #   file:PATH,VDEV,MODE
1.3       mspo      392:     #
                    393:     # where PATH is the path to the file used as the virtual disk, and VDEV
                    394:     # and MODE have the same meaning as for ``phy'' devices.
                    395:     #
                    396:     # VDEV doesn't really matter for a NetBSD guest OS (it's just used as an index),
                    397:     # but it does for Linux.
                    398:     # Worse, the device has to exist in /dev/ of domain0, because xm will
                    399:     # try to stat() it. This means that in order to load a Linux guest OS
                    400:     # from a NetBSD domain0, you'll have to create /dev/hda1, /dev/hda2, ...
                    401:     # on domain0, with the major/minor from Linux :(
                    402:     # Alternatively it's possible to specify the device number in hex,
                    403:     # e.g. 0x301 for /dev/hda1, 0x302 for /dev/hda2, etc ...
1.5       mspo      404: 
1.3       mspo      405:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
                    406:     #disk = [ 'file:/var/xen/nbsd-disk,0x01,w' ]
                    407:     #disk = [ 'file:/var/xen/nbsd-disk,0x301,w' ]
1.5       mspo      408: 
1.3       mspo      409:     #----------------------------------------------------------------------------
                    410:     # Set the kernel command line for the new domain.
1.5       mspo      411: 
1.3       mspo      412:     # Set root device. This one does matter for NetBSD
                    413:     root = "xbd0"
                    414:     # extra parameters passed to the kernel
                    415:     # this is where you can set boot flags like -s, -a, etc ...
                    416:     #extra = ""
1.5       mspo      417: 
1.3       mspo      418:     #----------------------------------------------------------------------------
                    419:     # Set according to whether you want the domain restarted when it exits.
                    420:     # The default is False.
                    421:     #autorestart = True
1.5       mspo      422: 
1.3       mspo      423:     # end of nbsd config file ====================================================
1.1       mspo      424: 
                    425: When a new domain is created, xen calls the
1.5       mspo      426: `/usr/pkg/etc/xen/vif-bridge` script for each virtual network interface
                    427: created in *domain0*. This can be used to automatically configure the
                    428: xvif?.? interfaces in *domain0*. In our example, these will be bridged
                    429: with the bridge0 device in *domain0*, but the bridge has to exist first.
                    430: To do this, create the file `/etc/ifconfig.bridge0` and make it look
                    431: like this:
1.1       mspo      432: 
1.3       mspo      433:     create
                    434:     !brconfig $int add ex0 up
1.1       mspo      435: 
1.5       mspo      436: (replace `ex0` with the name of your physical interface). Then bridge0
                    437: will be created on boot. See the MAN.BRIDGE.4 man page for details.
1.1       mspo      438: 
1.5       mspo      439: So, here is a suitable `/usr/pkg/etc/xen/vif-bridge` for xvif?.? (a
                    440: working vif-bridge is also provided with xentools20) configuring:
1.1       mspo      441: 
1.5       mspo      442:     #!/bin/sh
1.3       mspo      443:     #============================================================================
1.22    ! gdt       444:     # $NetBSD: howto.mdwn,v 1.21 2014/12/24 01:12:49 gdt Exp $
1.3       mspo      445:     #
                    446:     # /usr/pkg/etc/xen/vif-bridge
                    447:     #
                    448:     # Script for configuring a vif in bridged mode with a dom0 interface.
                    449:     # The xend(8) daemon calls a vif script when bringing a vif up or down.
                    450:     # The script name to use is defined in /usr/pkg/etc/xen/xend-config.sxp
                    451:     # in the ``vif-script'' field.
                    452:     #
                    453:     # Usage: vif-bridge up|down [var=value ...]
                    454:     #
                    455:     # Actions:
1.5       mspo      456:     #    up     Adds the vif interface to the bridge.
                    457:     #    down   Removes the vif interface from the bridge.
1.3       mspo      458:     #
                    459:     # Variables:
1.5       mspo      460:     #    domain name of the domain the interface is on (required).
                    461:     #    vifq   vif interface name (required).
                    462:     #    mac    vif MAC address (required).
                    463:     #    bridge bridge to add the vif to (required).
1.3       mspo      464:     #
                    465:     # Example invocation:
                    466:     #
                    467:     # vif-bridge up domain=VM1 vif=xvif1.0 mac="ee:14:01:d0:ec:af" bridge=bridge0
                    468:     #
                    469:     #============================================================================
1.5       mspo      470: 
1.3       mspo      471:     # Exit if anything goes wrong
                    472:     set -e
1.5       mspo      473: 
1.3       mspo      474:     echo "vif-bridge $*"
1.5       mspo      475: 
1.3       mspo      476:     # Operation name.
                    477:     OP=$1; shift
1.5       mspo      478: 
1.3       mspo      479:     # Pull variables in args into environment
                    480:     for arg ; do export "${arg}" ; done
1.5       mspo      481: 
1.3       mspo      482:     # Required parameters. Fail if not set.
                    483:     domain=${domain:?}
                    484:     vif=${vif:?}
                    485:     mac=${mac:?}
                    486:     bridge=${bridge:?}
1.5       mspo      487: 
1.3       mspo      488:     # Optional parameters. Set defaults.
                    489:     ip=${ip:-''}   # default to null (do nothing)
1.5       mspo      490: 
1.3       mspo      491:     # Are we going up or down?
                    492:     case $OP in
1.5       mspo      493:     up) brcmd='add' ;;
1.3       mspo      494:     down)   brcmd='delete' ;;
                    495:     *)
1.5       mspo      496:         echo 'Invalid command: ' $OP
                    497:         echo 'Valid commands are: up, down'
                    498:         exit 1
                    499:         ;;
1.3       mspo      500:     esac
1.5       mspo      501: 
1.3       mspo      502:     # Don't do anything if the bridge is "null".
                    503:     if [ "${bridge}" = "null" ] ; then
1.5       mspo      504:         exit
1.3       mspo      505:     fi
1.5       mspo      506: 
1.3       mspo      507:     # Don't do anything if the bridge doesn't exist.
                    508:     if ! ifconfig -l | grep "${bridge}" >/dev/null; then
1.5       mspo      509:         exit
1.3       mspo      510:     fi
1.5       mspo      511: 
1.3       mspo      512:     # Add/remove vif to/from bridge.
                    513:     ifconfig x${vif} $OP
                    514:     brconfig ${bridge} ${brcmd} x${vif}
1.1       mspo      515: 
                    516: Now, running
                    517: 
1.3       mspo      518:     xm create -c /usr/pkg/etc/xen/nbsd
1.1       mspo      519: 
1.5       mspo      520: should create a domain and load a NetBSD kernel in it. (Note: `-c`
                    521: causes xm to connect to the domain's console once created.) The kernel
                    522: will try to find its root file system on xbd0 (i.e., wd0e) which hasn't
                    523: been created yet. wd0e will be seen as a disk device in the new domain,
                    524: so it will be 'sub-partitioned'. We could attach a ccd to wd0e in
                    525: *domain0* and partition it, newfs and extract the NetBSD/i386 or amd64
                    526: tarballs there, but there's an easier way: load the
                    527: `netbsd-INSTALL_XEN3_DOMU` kernel provided in the NetBSD binary sets.
                    528: Like other install kernels, it contains a ramdisk with sysinst, so you
                    529: can install NetBSD using sysinst on your new domain.
1.1       mspo      530: 
                    531: If you want to install NetBSD/Xen with a CDROM image, the following line
1.5       mspo      532: should be used in the `/usr/pkg/etc/xen/nbsd` file:
1.1       mspo      533: 
1.3       mspo      534:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
1.1       mspo      535: 
                    536: After booting the domain, the option to install via CDROM may be
1.5       mspo      537: selected. The CDROM device should be changed to `xbd1d`.
1.1       mspo      538: 
1.5       mspo      539: Once done installing, `halt -p` the new domain (don't reboot or halt, it
                    540: would reload the INSTALL\_XEN3\_DOMU kernel even if you changed the
1.1       mspo      541: config file), switch the config file back to the XEN3\_DOMU kernel, and
1.5       mspo      542: start the new domain again. Now it should be able to use `root on xbd0a`
                    543: and you should have a second, functional NetBSD system on your xen
                    544: installation.
1.1       mspo      545: 
                    546: When the new domain is booting you'll see some warnings about *wscons*
                    547: and the pseudo-terminals. These can be fixed by editing the files
1.5       mspo      548: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
                    549: `/etc/ttys`, except *console*, like this:
1.1       mspo      550: 
1.3       mspo      551:     console "/usr/libexec/getty Pc"         vt100   on secure
                    552:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
                    553:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
                    554:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
                    555:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
1.1       mspo      556: 
1.5       mspo      557: Finally, all screens must be commented out from `/etc/wscons.conf`.
1.1       mspo      558: 
                    559: It is also desirable to add
                    560: 
1.3       mspo      561:     powerd=YES
1.1       mspo      562: 
1.5       mspo      563: in rc.conf. This way, the domain will be properly shut down if
                    564: `xm shutdown -R` or `xm shutdown -H` is used on the domain0.
1.1       mspo      565: 
                    566: Your domain should be now ready to work, enjoy.
                    567: 
1.14      gdt       568: Creating an unprivileged Linux domain (domU)
1.5       mspo      569: --------------------------------------------
1.1       mspo      570: 
                    571: Creating unprivileged Linux domains isn't much different from
                    572: unprivileged NetBSD domains, but there are some details to know.
                    573: 
                    574: First, the second parameter passed to the disk declaration (the '0x1' in
                    575: the example below)
                    576: 
1.3       mspo      577:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
1.1       mspo      578: 
                    579: does matter to Linux. It wants a Linux device number here (e.g. 0x300
                    580: for hda). Linux builds device numbers as: (major \<\< 8 + minor). So,
                    581: hda1 which has major 3 and minor 1 on a Linux system will have device
                    582: number 0x301. Alternatively, devices names can be used (hda, hdb, ...)
                    583: as xentools has a table to map these names to devices numbers. To export
                    584: a partition to a Linux guest we can use:
                    585: 
1.3       mspo      586:     disk = [ 'phy:/dev/wd0e,0x300,w' ]
                    587:     root = "/dev/hda1 ro"
1.1       mspo      588: 
                    589: and it will appear as /dev/hda on the Linux system, and be used as root
                    590: partition.
                    591: 
                    592: To install the Linux system on the partition to be exported to the guest
                    593: domain, the following method can be used: install sysutils/e2fsprogs
                    594: from pkgsrc. Use mke2fs to format the partition that will be the root
                    595: partition of your Linux domain, and mount it. Then copy the files from a
1.5       mspo      596: working Linux system, make adjustments in `/etc` (fstab, network
                    597: config). It should also be possible to extract binary packages such as
                    598: .rpm or .deb directly to the mounted partition using the appropriate
                    599: tool, possibly running under NetBSD's Linux emulation. Once the
                    600: filesystem has been populated, umount it. If desirable, the filesystem
                    601: can be converted to ext3 using tune2fs -j. It should now be possible to
                    602: boot the Linux guest domain, using one of the vmlinuz-\*-xenU kernels
                    603: available in the Xen binary distribution.
1.1       mspo      604: 
                    605: To get the linux console right, you need to add:
                    606: 
1.3       mspo      607:     extra = "xencons=tty1"
1.1       mspo      608: 
                    609: to your configuration since not all linux distributions auto-attach a
                    610: tty to the xen console.
                    611: 
1.14      gdt       612: Creating an unprivileged Solaris domain (domU)
1.5       mspo      613: ----------------------------------------------
1.1       mspo      614: 
                    615: Download an Opensolaris [release](http://opensolaris.org/os/downloads/)
                    616: or [development snapshot](http://genunix.org/) DVD image. Attach the DVD
1.5       mspo      617: image to a MAN.VND.4 device. Copy the kernel and ramdisk filesystem
                    618: image to your dom0 filesystem.
1.1       mspo      619: 
1.3       mspo      620:     dom0# mkdir /root/solaris
                    621:     dom0# vnconfig vnd0 osol-1002-124-x86.iso
                    622:     dom0# mount /dev/vnd0a /mnt
1.5       mspo      623: 
1.3       mspo      624:     ## for a 64-bit guest
                    625:     dom0# cp /mnt/boot/amd64/x86.microroot /root/solaris
                    626:     dom0# cp /mnt/platform/i86xpv/kernel/amd64/unix /root/solaris
1.5       mspo      627: 
1.3       mspo      628:     ## for a 32-bit guest
                    629:     dom0# cp /mnt/boot/x86.microroot /root/solaris
                    630:     dom0# cp /mnt/platform/i86xpv/kernel/unix /root/solaris
1.5       mspo      631: 
1.3       mspo      632:     dom0# umount /mnt
1.5       mspo      633:           
                    634: 
                    635: Keep the MAN.VND.4 configured. For some reason the boot process stalls
                    636: unless the DVD image is attached to the guest as a "phy" device. Create
                    637: an initial configuration file with the following contents. Substitute
                    638: */dev/wd0k* with an empty partition at least 8 GB large.
1.1       mspo      639: 
1.4       mspo      640:     memory = 640
                    641:     name = 'solaris'
                    642:     disk = [ 'phy:/dev/wd0k,0,w' ]
                    643:     disk += [ 'phy:/dev/vnd0d,6:cdrom,r' ]
                    644:     vif = [ 'bridge=bridge0' ]
                    645:     kernel = '/root/solaris/unix'
                    646:     ramdisk = '/root/solaris/x86.microroot'
                    647:     # for a 64-bit guest
                    648:     extra = '/platform/i86xpv/kernel/amd64/unix - nowin -B install_media=cdrom'
                    649:     # for a 32-bit guest
                    650:     #extra = '/platform/i86xpv/kernel/unix - nowin -B install_media=cdrom'
1.5       mspo      651:           
                    652: 
1.1       mspo      653: Start the guest.
                    654: 
1.4       mspo      655:     dom0# xm create -c solaris.cfg
                    656:     Started domain solaris
                    657:                           v3.3.2 chgset 'unavailable'
                    658:     SunOS Release 5.11 Version snv_124 64-bit
                    659:     Copyright 1983-2009 Sun Microsystems, Inc.  All rights reserved.
                    660:     Use is subject to license terms.
                    661:     Hostname: opensolaris
                    662:     Remounting root read/write
                    663:     Probing for device nodes ...
                    664:     WARNING: emlxs: ddi_modopen drv/fct failed: err 2
                    665:     Preparing live image for use
                    666:     Done mounting Live image
1.5       mspo      667:           
1.1       mspo      668: 
                    669: Make sure the network is configured. Note that it can take a minute for
                    670: the xnf0 interface to appear.
                    671: 
1.4       mspo      672:     opensolaris console login: jack
                    673:     Password: jack
                    674:     Sun Microsystems Inc.   SunOS 5.11      snv_124 November 2008
                    675:     jack@opensolaris:~$ pfexec sh
                    676:     sh-3.2# ifconfig -a
                    677:     sh-3.2# exit
1.5       mspo      678:           
1.1       mspo      679: 
                    680: Set a password for VNC and start the VNC server which provides the X11
                    681: display where the installation program runs.
                    682: 
1.4       mspo      683:     jack@opensolaris:~$ vncpasswd
                    684:     Password: solaris
                    685:     Verify: solaris
                    686:     jack@opensolaris:~$ cp .Xclients .vnc/xstartup
                    687:     jack@opensolaris:~$ vncserver :1
1.5       mspo      688:           
1.1       mspo      689: 
1.5       mspo      690: From a remote machine connect to the VNC server. Use `ifconfig xnf0` on
                    691: the guest to find the correct IP address to use.
1.1       mspo      692: 
1.4       mspo      693:     remote$ vncviewer 172.18.2.99:1
1.5       mspo      694:           
1.1       mspo      695: 
                    696: It is also possible to launch the installation on a remote X11 display.
                    697: 
1.4       mspo      698:     jack@opensolaris:~$ export DISPLAY=172.18.1.1:0
                    699:     jack@opensolaris:~$ pfexec gui-install
1.5       mspo      700:            
1.1       mspo      701: 
                    702: After the GUI installation is complete you will be asked to reboot.
                    703: Before that you need to determine the ZFS ID for the new boot filesystem
                    704: and update the configuration file accordingly. Return to the guest
                    705: console.
                    706: 
1.4       mspo      707:     jack@opensolaris:~$ pfexec zdb -vvv rpool | grep bootfs
                    708:                     bootfs = 43
                    709:     ^C
                    710:     jack@opensolaris:~$
1.5       mspo      711:            
1.1       mspo      712: 
                    713: The final configuration file should look like this. Note in particular
                    714: the last line.
                    715: 
1.4       mspo      716:     memory = 640
                    717:     name = 'solaris'
                    718:     disk = [ 'phy:/dev/wd0k,0,w' ]
                    719:     vif = [ 'bridge=bridge0' ]
                    720:     kernel = '/root/solaris/unix'
                    721:     ramdisk = '/root/solaris/x86.microroot'
                    722:     extra = '/platform/i86xpv/kernel/amd64/unix -B zfs-bootfs=rpool/43,bootpath="/xpvd/xdf@0:a"'
1.5       mspo      723:            
1.1       mspo      724: 
                    725: Restart the guest to verify it works correctly.
                    726: 
1.4       mspo      727:     dom0# xm destroy solaris
                    728:     dom0# xm create -c solaris.cfg
                    729:     Using config file "./solaris.cfg".
                    730:     v3.3.2 chgset 'unavailable'
                    731:     Started domain solaris
                    732:     SunOS Release 5.11 Version snv_124 64-bit
                    733:     Copyright 1983-2009 Sun Microsystems, Inc.  All rights reserved.
                    734:     Use is subject to license terms.
                    735:     WARNING: emlxs: ddi_modopen drv/fct failed: err 2
                    736:     Hostname: osol
                    737:     Configuring devices.
                    738:     Loading smf(5) service descriptions: 160/160
                    739:     svccfg import warnings. See /var/svc/log/system-manifest-import:default.log .
                    740:     Reading ZFS config: done.
                    741:     Mounting ZFS filesystems: (6/6)
                    742:     Creating new rsa public/private host key pair
                    743:     Creating new dsa public/private host key pair
1.5       mspo      744: 
1.4       mspo      745:     osol console login:
1.5       mspo      746:            
1.1       mspo      747: 
                    748: Using PCI devices in guest domains
1.14      gdt       749: ----------------------------------
1.1       mspo      750: 
                    751: The domain0 can give other domains access to selected PCI devices. This
                    752: can allow, for example, a non-privileged domain to have access to a
                    753: physical network interface or disk controller. However, keep in mind
                    754: that giving a domain access to a PCI device most likely will give the
                    755: domain read/write access to the whole physical memory, as PCs don't have
                    756: an IOMMU to restrict memory access to DMA-capable device. Also, it's not
                    757: possible to export ISA devices to non-domain0 domains (which means that
                    758: the primary VGA adapter can't be exported. A guest domain trying to
                    759: access the VGA registers will panic).
                    760: 
                    761: This functionality is only available in NetBSD-5.1 (and later) domain0
                    762: and domU. If the domain0 is NetBSD, it has to be running Xen 3.1, as
                    763: support has not been ported to later versions at this time.
                    764: 
                    765: For a PCI device to be exported to a domU, is has to be attached to the
1.5       mspo      766: `pciback` driver in domain0. Devices passed to the domain0 via the
                    767: pciback.hide boot parameter will attach to `pciback` instead of the
                    768: usual driver. The list of devices is specified as `(bus:dev.func)`,
                    769: where bus and dev are 2-digit hexadecimal numbers, and func a
                    770: single-digit number:
1.1       mspo      771: 
1.4       mspo      772:     pciback.hide=(00:0a.0)(00:06.0)
1.1       mspo      773: 
                    774: pciback devices should show up in the domain0's boot messages, and the
1.5       mspo      775: devices should be listed in the `/kern/xen/pci` directory.
1.1       mspo      776: 
1.5       mspo      777: PCI devices to be exported to a domU are listed in the `pci` array of
                    778: the domU's config file, with the format `'0000:bus:dev.func'`
1.1       mspo      779: 
1.4       mspo      780:     pci = [ '0000:00:06.0', '0000:00:0a.0' ]
1.1       mspo      781: 
1.5       mspo      782: In the domU an `xpci` device will show up, to which one or more pci
                    783: busses will attach. Then the PCI drivers will attach to PCI busses as
                    784: usual. Note that the default NetBSD DOMU kernels do not have `xpci` or
                    785: any PCI drivers built in by default; you have to build your own kernel
                    786: to use PCI devices in a domU. Here's a kernel config example:
1.1       mspo      787: 
1.4       mspo      788:     include         "arch/i386/conf/XEN3_DOMU"
                    789:     #include         "arch/i386/conf/XENU"           # in NetBSD 3.0
1.5       mspo      790: 
1.4       mspo      791:     # Add support for PCI busses to the XEN3_DOMU kernel
                    792:     xpci* at xenbus ?
                    793:     pci* at xpci ?
1.5       mspo      794: 
1.4       mspo      795:     # Now add PCI and related devices to be used by this domain
                    796:     # USB Controller and Devices
1.5       mspo      797: 
1.4       mspo      798:     # PCI USB controllers
                    799:     uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
1.5       mspo      800: 
1.4       mspo      801:     # USB bus support
                    802:     usb*    at uhci?
1.5       mspo      803: 
1.4       mspo      804:     # USB Hubs
                    805:     uhub*   at usb?
                    806:     uhub*   at uhub? port ? configuration ? interface ?
1.5       mspo      807: 
1.4       mspo      808:     # USB Mass Storage
                    809:     umass*  at uhub? port ? configuration ? interface ?
                    810:     wd*     at umass?
                    811:     # SCSI controllers
                    812:     ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
1.5       mspo      813: 
1.4       mspo      814:     # SCSI bus support (for both ahc and umass)
                    815:     scsibus* at scsi?
1.5       mspo      816: 
1.4       mspo      817:     # SCSI devices
                    818:     sd*     at scsibus? target ? lun ?      # SCSI disk drives
                    819:     cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
1.1       mspo      820: 
                    821: Links and further information
1.5       mspo      822: =============================
1.1       mspo      823: 
1.9       gdt       824: -   The [HowTo on Installing into RAID-1](http://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
1.8       gdt       825:     explains how to set up booting a dom0 with Xen using grub 
                    826:     with NetBSD's RAIDframe.  (This is obsolete with the use of
                    827:     NetBSD's native boot.)
1.1       mspo      828: -   An example of how to use NetBSD's native bootloader to load
1.9       gdt       829:     NetBSD/Xen instead of Grub can be found in the i386/amd64 boot(8)
                    830:     and boot.cfg(5) manpages.

CVSweb for NetBSD wikisrc <wikimaster@NetBSD.org> software: FreeBSD-CVSweb