File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.17: download - view: text, annotated - select for diffs
Wed Dec 24 00:06:31 2014 UTC (4 years, 11 months ago) by gdt
Branches: MAIN
CVS tags: HEAD
tighten up spacing

    1: Introduction
    2: ============
    3: 
    4: [![[Xen
    5: screenshot]](http://www.netbsd.org/gallery/in-Action/hubertf-xens.png)](../../gallery/in-Action/hubertf-xen.png)
    6: 
    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: 
   30: At boot, the dom0 kernel is loaded as module with Xen as the kernel.
   31: The dom0 can start one or more domUs.  (Booting is explained in detail
   32: in the dom0 section.)
   33: 
   34: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
   35: and that Xen kernels and tools are available in pkgsrc.  This HOWTO
   36: attempts to address both the case of running a NetBSD dom0 on hardware
   37: and running NetBSD as a domU in a VPS.
   38: 
   39: Prerequisites
   40: -------------
   41: 
   42: Installing NetBSD/Xen is not extremely difficult, but it is more
   43: complex than a normal installation of NetBSD.
   44: In general, this HOWTO is occasionally overly restrictive about how
   45: things must be done, guiding the reader to stay on the established
   46: path when there are no known good reasons to stray.
   47: 
   48: This HOWTO presumes a basic familiarity with the Xen system
   49: architecture.  This HOWTO presumes familiarity with installing NetBSD
   50: on i386/amd64 hardware and installing software from pkgsrc.
   51: See also the [Xen website](http://www.xen.org/).
   52: 
   53: Versions of Xen and NetBSD
   54: ==========================
   55: 
   56: Most of the installation concepts and instructions are independent of
   57: Xen version.  This section gives advice on which version to choose.
   58: Versions not in pkgsrc and older unsupported versions of NetBSD are
   59: inentionally ignored.
   60: 
   61: Xen
   62: ---
   63: 
   64: In NetBSD, xen is provided in pkgsrc, via matching pairs of packages
   65: xenkernel and xentools.  We will refer only to the kernel versions,
   66: but note that both packages must be installed together and must have
   67: matching versions.
   68: 
   69: xenkernel3 and xenkernel33 provide Xen 3.1 and 3.3.  These no longer
   70: receive security patches and should not be used.
   71: 
   72: xenkernel41 provides Xen 4.1.  This is no longer maintained by Xen,
   73: but as of 2014-12 receives backported security patches.  It is a
   74: reasonable although trailing-edge choice.
   75: 
   76: xenkernel42 provides Xen 4.2.  This is maintained by Xen, but old as
   77: of 2014-12.
   78: 
   79: Ideally newer versions of Xen will be added to pkgsrc.
   80: 
   81: NetBSD
   82: ------
   83: 
   84: The netbsd-5, netbsd-6, netbsd-7, and -current branches are all
   85: reasonable choices, with more or less the same considerations for
   86: non-Xen use.  Therefore, netbsd-6 is recommended as the stable version
   87: of the most recent release.
   88: 
   89: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
   90: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
   91: need SMP; the lack of support is really a problem when using a dom0 as
   92: a normal computer.)
   93: 
   94: Recommendation
   95: --------------
   96: 
   97: Therefore, this HOWTO recommends running xenkernel42 (and xentools42)
   98: and NetBSD 6 stable branch.
   99: 
  100: NetBSD as a dom0
  101: ================
  102: 
  103: NetBSD can be used as a dom0 and works very well.  The following
  104: sections address installation, updating NetBSD, and updating Xen.
  105: 
  106: Styles of dom0 operation
  107: ------------------------
  108: 
  109: There are two basic ways to use Xen.  The traditional method is for
  110: the dom0 to do absolutely nothing other than providing support to some
  111: number of domUs.  Such a system was probably installed for the sole
  112: purpose of hosting domUs, and sits in a server room on a UPS.
  113: 
  114: The other way is to put Xen under a normal-usage computer, so that the
  115: dom0 is what the computer would have been without Xen, perhaps a
  116: desktop or laptop.  Then, one can run domUs at will.  Purists will
  117: deride this as less secure than the previous approach, and for a
  118: computer whose purpose is to run domUs, they are right.  But Xen and a
  119: dom0 (without domUs) is not meaingfully less secure than the same
  120: things running without Xen.  One can boot Xen or boot regular NetBSD
  121: alternately with little problems, simply refraining from starting the
  122: Xen daemons when not running Xen.
  123: 
  124: Note that NetBSD as dom0 does not support multiple CPUs.  This will
  125: limit the performance of the Xen/dom0 workstation approach.
  126: 
  127: Installation of NetBSD and Xen
  128: ------------------------------
  129: 
  130: Note that it doesn't make sense to talk about installing a dom0 OS
  131: without also installing Xen itself.
  132: 
  133: First do a NetBSD/i386 or NetBSD/amd64
  134: [installation](../../docs/guide/en/chap-inst.html) of the 5.1 release
  135: (or newer) as you usually do on x86 hardware. The binary releases are
  136: available from [](ftp://ftp.NetBSD.org/pub/NetBSD/). Binary snapshots
  137: for current and the stable branches are available on daily autobuilds.
  138: If you plan to use the `grub` boot loader, when partitioning the disk
  139: you have to make the root partition smaller than 512Mb, and formatted as
  140: FFSv1 with 8k block/1k fragments. If the partition is larger than this,
  141: uses FFSv2 or has different block/fragment sizes, grub may fail to load
  142: some files. Also keep in mind that you'll probably want to provide
  143: virtual disks to other domains, so reserve some partitions for these
  144: virtual disks. Alternatively, you can create large files in the file
  145: system, map them to vnd(4) devices and export theses vnd devices to
  146: other domains.
  147: 
  148: Next step is to install the Xen packages via pkgsrc or from binary
  149: packages. See [the pkgsrc
  150: documentation](http://www.NetBSD.org/docs/pkgsrc/) if you are unfamiliar
  151: with pkgsrc and/or handling of binary packages. Xen 3.1, 3.3, 4.1 and
  152: 4.2 are available. 3.1 supports PCI pass-through while other versions do
  153: not. You'll need either `sysutils/xentools3` and `sysutils/xenkernel3`
  154: for Xen 3.1, `sysutils/xentools33` and `sysutils/xenkernel33` for Xen
  155: 3.3, `sysutils/xentools41` and `sysutils/xenkernel41` for Xen 4.1. or
  156: `sysutils/xentools42` and `sysutils/xenkernel42` for Xen 4.2. You'll
  157: also need `sysutils/grub` if you plan do use the grub boot loader. If
  158: using Xen 3.1, you may also want to install `sysutils/xentools3-hvm`
  159: which contains the utilities to run unmodified guests OSes using the
  160: *HVM* support (for later versions this is included in
  161: `sysutils/xentools`). Note that your CPU needs to support this. Intel
  162: CPUs must have the 'VT' instruction, AMD CPUs the 'SVM' instruction. You
  163: can easily find out if your CPU support HVM by using NetBSD's cpuctl
  164: command:
  165: 
  166:     # cpuctl identify 0
  167:     cpu0: Intel Core 2 (Merom) (686-class), id 0x6f6
  168:     cpu0: features 0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR>
  169:     cpu0: features 0xbfebfbff<PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX>
  170:     cpu0: features 0xbfebfbff<FXSR,SSE,SSE2,SS,HTT,TM,SBF>
  171:     cpu0: features2 0x4e33d<SSE3,DTES64,MONITOR,DS-CPL,,TM2,SSSE3,CX16,xTPR,PDCM,DCA>
  172:     cpu0: features3 0x20100800<SYSCALL/SYSRET,XD,EM64T>
  173:     cpu0: "Intel(R) Xeon(R) CPU            5130  @ 2.00GHz"
  174:     cpu0: I-cache 32KB 64B/line 8-way, D-cache 32KB 64B/line 8-way
  175:     cpu0: L2 cache 4MB 64B/line 16-way
  176:     cpu0: ITLB 128 4KB entries 4-way
  177:     cpu0: DTLB 256 4KB entries 4-way, 32 4MB entries 4-way
  178:     cpu0: Initial APIC ID 0
  179:     cpu0: Cluster/Package ID 0
  180:     cpu0: Core ID 0
  181:     cpu0: family 06 model 0f extfamily 00 extmodel 00
  182: 
  183: Depending on your CPU, the feature you are looking for is called HVM,
  184: SVM or VMX.
  185: 
  186: Next you need to copy the selected Xen kernel itself. pkgsrc installed
  187: them under `/usr/pkg/xen*-kernel/`. The file you're looking for is
  188: `xen.gz`. Copy it to your root file system. `xen-debug.gz` is a kernel
  189: with more consistency checks and more details printed on the serial
  190: console. It is useful for debugging crashing guests if you use a serial
  191: console. It is not useful with a VGA console.
  192: 
  193: You'll then need a NetBSD/Xen kernel for *domain0* on your root file
  194: system. The XEN3PAE\_DOM0 kernel or XEN3\_DOM0 provided as part of the
  195: i386 or amd64 binaries is suitable for this, but you may want to
  196: customize it. Keep your native kernel around, as it can be useful for
  197: recovery. *Note:* the *domain0* kernel must support KERNFS and `/kern`
  198: must be mounted because *xend* needs access to `/kern/xen/privcmd`.
  199: 
  200: Next you need to get a bootloader to load the `xen.gz` kernel, and the
  201: NetBSD *domain0* kernel as a module. This can be `grub` or NetBSD's boot
  202: loader. Below is a detailled example for grub, see the boot.cfg(5)
  203: manual page for an example using the latter.
  204: 
  205: This is also where you'll specify the memory allocated to *domain0*, the
  206: console to use, etc ...
  207: 
  208: Here is a commented `/grub/menu.lst` file:
  209: 
  210:     #Grub config file for NetBSD/xen. Copy as /grub/menu.lst and run
  211:     # grub-install /dev/rwd0d (assuming your boot device is wd0).
  212:     #
  213:     # The default entry to load will be the first one
  214:     default=0
  215: 
  216:     # boot the default entry after 10s if the user didn't hit keyboard
  217:     timeout=10
  218: 
  219:     # Configure serial port to use as console. Ignore if you'll use VGA only
  220:     serial --unit=0 --speed=115200 --word=8 --parity=no --stop=1
  221: 
  222:     # Let the user select which console to use (serial or VGA), default
  223:     # to serial after 10s
  224:     terminal --timeout=10 serial console
  225: 
  226:     # An entry for NetBSD/xen, using /netbsd as the domain0 kernel, and serial
  227:     # console. Domain0 will have 64MB RAM allocated.
  228:     # Assume NetBSD is installed in the first MBR partition.
  229:     title Xen 3 / NetBSD (hda0, serial)
  230:       root(hd0,0)
  231:       kernel (hd0,a)/xen.gz dom0_mem=65536 com1=115200,8n1
  232:       module (hd0,a)/netbsd bootdev=wd0a ro console=ttyS0
  233: 
  234:     # Same as above, but using VGA console
  235:     # We can use console=tty0 (Linux syntax) or console=pc (NetBSD syntax)
  236:     title Xen 3 / NetBSD (hda0, vga)
  237:       root(hd0,0)
  238:       kernel (hd0,a)/xen.gz dom0_mem=65536
  239:       module (hd0,a)/netbsd bootdev=wd0a ro console=tty0
  240: 
  241:     # NetBSD/xen using a backup domain0 kernel (in case you installed a
  242:     # nonworking kernel as /netbsd
  243:     title Xen 3 / NetBSD (hda0, backup, serial)
  244:       root(hd0,0)
  245:       kernel (hd0,a)/xen.gz dom0_mem=65536 com1=115200,8n1
  246:       module (hd0,a)/netbsd.backup bootdev=wd0a ro console=ttyS0
  247:     title Xen 3 / NetBSD (hda0, backup, VGA)
  248:       root(hd0,0)
  249:       kernel (hd0,a)/xen.gz dom0_mem=65536
  250:       module (hd0,a)/netbsd.backup bootdev=wd0a ro console=tty0
  251: 
  252:     #Load a regular NetBSD/i386 kernel. Can be useful if you end up with a
  253:     #nonworking /xen.gz
  254:     title NetBSD 5.1
  255:       root (hd0,a)
  256:       kernel --type=netbsd /netbsd-GENERIC
  257: 
  258:     #Load the NetBSD bootloader, letting it load the NetBSD/i386 kernel.
  259:     #May be better than the above, as grub can't pass all required infos
  260:     #to the NetBSD/i386 kernel (e.g. console, root device, ...)
  261:     title NetBSD chain
  262:       root        (hd0,0)
  263:       chainloader +1
  264: 
  265:     ## end of grub config file.
  266:           
  267: 
  268: Install grub with the following command:
  269: 
  270:     # grub --no-floppy
  271: 
  272:     grub> root (hd0,a)
  273:      Filesystem type is ffs, partition type 0xa9
  274: 
  275:     grub> setup (hd0)
  276:      Checking if "/boot/grub/stage1" exists... no
  277:      Checking if "/grub/stage1" exists... yes
  278:      Checking if "/grub/stage2" exists... yes
  279:      Checking if "/grub/ffs_stage1_5" exists... yes
  280:      Running "embed /grub/ffs_stage1_5 (hd0)"...  14 sectors are embedded.
  281:     succeeded
  282:      Running "install /grub/stage1 (hd0) (hd0)1+14 p (hd0,0,a)/grub/stage2 /grub/menu.lst"...
  283:      succeeded
  284:     Done.
  285:           
  286: 
  287: Updating NetBSD in a dom0
  288: -------------------------
  289: 
  290: This is just like updating NetBSD on bare hardware, assuming the new
  291: version supports the version of Xen you are running.  Generally, one
  292: replaces the kernel and reboots, and then overlays userland binaries
  293: and adjusts /etc.
  294: 
  295: Note that one must update both the non-Xen kernel typically used for
  296: rescue purposes and the DOM0 kernel used with Xen.
  297: 
  298: Updating Xen versions
  299: ---------------------
  300: 
  301: TODO: write
  302: 
  303: Creating unprivileged domains (domU)
  304: ====================================
  305: 
  306: Creating domUs is almost entirely independent of operating system.  We
  307: first explain NetBSD, and then differences for Linux and Solaris.
  308: 
  309: Creating an unprivileged NetBSD domain (domU)
  310: ---------------------------------------------
  311: 
  312: Once you have *domain0* running, you need to start the xen tool daemon
  313: (`/usr/pkg/share/examples/rc.d/xend start`) and the xen backend daemon
  314: (`/usr/pkg/share/examples/rc.d/xenbackendd start` for Xen3\*,
  315: `/usr/pkg/share/examples/rc.d/xencommons start` for Xen4.\*). Make sure
  316: that `/dev/xencons` and `/dev/xenevt` exist before starting `xend`. You
  317: can create them with this command:
  318: 
  319:     # cd /dev && sh MAKEDEV xen
  320: 
  321: xend will write logs to `/var/log/xend.log` and
  322: `/var/log/xend-debug.log`. You can then control xen with the xm tool.
  323: 'xm list' will show something like:
  324: 
  325:     # xm list
  326:     Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  327:     Domain-0           0       64    0  r----     58.1
  328: 
  329: 'xm create' allows you to create a new domain. It uses a config file in
  330: PKG\_SYSCONFDIR for its parameters. By default, this file will be in
  331: `/usr/pkg/etc/xen/`. On creation, a kernel has to be specified, which
  332: will be executed in the new domain (this kernel is in the *domain0* file
  333: system, not on the new domain virtual disk; but please note, you should
  334: install the same kernel into *domainU* as `/netbsd` in order to make
  335: your system tools, like MAN.SAVECORE.8, work). A suitable kernel is
  336: provided as part of the i386 and amd64 binary sets: XEN3\_DOMU.
  337: 
  338: Here is an /usr/pkg/etc/xen/nbsd example config file:
  339: 
  340:     #  -*- mode: python; -*-
  341:     #============================================================================
  342:     # Python defaults setup for 'xm create'.
  343:     # Edit this file to reflect the configuration of your system.
  344:     #============================================================================
  345: 
  346:     #----------------------------------------------------------------------------
  347:     # Kernel image file. This kernel will be loaded in the new domain.
  348:     kernel = "/home/bouyer/netbsd-XEN3_DOMU"
  349:     #kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  350: 
  351:     # Memory allocation (in megabytes) for the new domain.
  352:     memory = 128
  353: 
  354:     # A handy name for your new domain. This will appear in 'xm list',
  355:     # and you can use this as parameters for xm in place of the domain
  356:     # number. All domains must have different names.
  357:     #
  358:     name = "nbsd"
  359: 
  360:     # The number of virtual CPUs this domain has.
  361:     #
  362:     vcpus = 1
  363: 
  364:     #----------------------------------------------------------------------------
  365:     # Define network interfaces for the new domain.
  366: 
  367:     # Number of network interfaces (must be at least 1). Default is 1.
  368:     nics = 1
  369: 
  370:     # Define MAC and/or bridge for the network interfaces.
  371:     #
  372:     # The MAC address specified in ``mac'' is the one used for the interface
  373:     # in the new domain. The interface in domain0 will use this address XOR'd
  374:     # with 00:00:00:01:00:00 (i.e. aa:00:00:51:02:f0 in our example). Random
  375:     # MACs are assigned if not given.
  376:     #
  377:     # ``bridge'' is a required parameter, which will be passed to the
  378:     # vif-script called by xend(8) when a new domain is created to configure
  379:     # the new xvif interface in domain0.
  380:     #
  381:     # In this example, the xvif is added to bridge0, which should have been
  382:     # set up prior to the new domain being created -- either in the
  383:     # ``network'' script or using a /etc/ifconfig.bridge0 file.
  384:     #
  385:     vif = [ 'mac=aa:00:00:50:02:f0, bridge=bridge0' ]
  386: 
  387:     #----------------------------------------------------------------------------
  388:     # Define the disk devices you want the domain to have access to, and
  389:     # what you want them accessible as.
  390:     #
  391:     # Each disk entry is of the form:
  392:     #
  393:     #   phy:DEV,VDEV,MODE
  394:     #
  395:     # where DEV is the device, VDEV is the device name the domain will see,
  396:     # and MODE is r for read-only, w for read-write.  You can also create
  397:     # file-backed domains using disk entries of the form:
  398:     #
  399:     #   file:PATH,VDEV,MODE
  400:     #
  401:     # where PATH is the path to the file used as the virtual disk, and VDEV
  402:     # and MODE have the same meaning as for ``phy'' devices.
  403:     #
  404:     # VDEV doesn't really matter for a NetBSD guest OS (it's just used as an index),
  405:     # but it does for Linux.
  406:     # Worse, the device has to exist in /dev/ of domain0, because xm will
  407:     # try to stat() it. This means that in order to load a Linux guest OS
  408:     # from a NetBSD domain0, you'll have to create /dev/hda1, /dev/hda2, ...
  409:     # on domain0, with the major/minor from Linux :(
  410:     # Alternatively it's possible to specify the device number in hex,
  411:     # e.g. 0x301 for /dev/hda1, 0x302 for /dev/hda2, etc ...
  412: 
  413:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  414:     #disk = [ 'file:/var/xen/nbsd-disk,0x01,w' ]
  415:     #disk = [ 'file:/var/xen/nbsd-disk,0x301,w' ]
  416: 
  417:     #----------------------------------------------------------------------------
  418:     # Set the kernel command line for the new domain.
  419: 
  420:     # Set root device. This one does matter for NetBSD
  421:     root = "xbd0"
  422:     # extra parameters passed to the kernel
  423:     # this is where you can set boot flags like -s, -a, etc ...
  424:     #extra = ""
  425: 
  426:     #----------------------------------------------------------------------------
  427:     # Set according to whether you want the domain restarted when it exits.
  428:     # The default is False.
  429:     #autorestart = True
  430: 
  431:     # end of nbsd config file ====================================================
  432: 
  433: When a new domain is created, xen calls the
  434: `/usr/pkg/etc/xen/vif-bridge` script for each virtual network interface
  435: created in *domain0*. This can be used to automatically configure the
  436: xvif?.? interfaces in *domain0*. In our example, these will be bridged
  437: with the bridge0 device in *domain0*, but the bridge has to exist first.
  438: To do this, create the file `/etc/ifconfig.bridge0` and make it look
  439: like this:
  440: 
  441:     create
  442:     !brconfig $int add ex0 up
  443: 
  444: (replace `ex0` with the name of your physical interface). Then bridge0
  445: will be created on boot. See the MAN.BRIDGE.4 man page for details.
  446: 
  447: So, here is a suitable `/usr/pkg/etc/xen/vif-bridge` for xvif?.? (a
  448: working vif-bridge is also provided with xentools20) configuring:
  449: 
  450:     #!/bin/sh
  451:     #============================================================================
  452:     # $NetBSD: howto.mdwn,v 1.17 2014/12/24 00:06:31 gdt Exp $
  453:     #
  454:     # /usr/pkg/etc/xen/vif-bridge
  455:     #
  456:     # Script for configuring a vif in bridged mode with a dom0 interface.
  457:     # The xend(8) daemon calls a vif script when bringing a vif up or down.
  458:     # The script name to use is defined in /usr/pkg/etc/xen/xend-config.sxp
  459:     # in the ``vif-script'' field.
  460:     #
  461:     # Usage: vif-bridge up|down [var=value ...]
  462:     #
  463:     # Actions:
  464:     #    up     Adds the vif interface to the bridge.
  465:     #    down   Removes the vif interface from the bridge.
  466:     #
  467:     # Variables:
  468:     #    domain name of the domain the interface is on (required).
  469:     #    vifq   vif interface name (required).
  470:     #    mac    vif MAC address (required).
  471:     #    bridge bridge to add the vif to (required).
  472:     #
  473:     # Example invocation:
  474:     #
  475:     # vif-bridge up domain=VM1 vif=xvif1.0 mac="ee:14:01:d0:ec:af" bridge=bridge0
  476:     #
  477:     #============================================================================
  478: 
  479:     # Exit if anything goes wrong
  480:     set -e
  481: 
  482:     echo "vif-bridge $*"
  483: 
  484:     # Operation name.
  485:     OP=$1; shift
  486: 
  487:     # Pull variables in args into environment
  488:     for arg ; do export "${arg}" ; done
  489: 
  490:     # Required parameters. Fail if not set.
  491:     domain=${domain:?}
  492:     vif=${vif:?}
  493:     mac=${mac:?}
  494:     bridge=${bridge:?}
  495: 
  496:     # Optional parameters. Set defaults.
  497:     ip=${ip:-''}   # default to null (do nothing)
  498: 
  499:     # Are we going up or down?
  500:     case $OP in
  501:     up) brcmd='add' ;;
  502:     down)   brcmd='delete' ;;
  503:     *)
  504:         echo 'Invalid command: ' $OP
  505:         echo 'Valid commands are: up, down'
  506:         exit 1
  507:         ;;
  508:     esac
  509: 
  510:     # Don't do anything if the bridge is "null".
  511:     if [ "${bridge}" = "null" ] ; then
  512:         exit
  513:     fi
  514: 
  515:     # Don't do anything if the bridge doesn't exist.
  516:     if ! ifconfig -l | grep "${bridge}" >/dev/null; then
  517:         exit
  518:     fi
  519: 
  520:     # Add/remove vif to/from bridge.
  521:     ifconfig x${vif} $OP
  522:     brconfig ${bridge} ${brcmd} x${vif}
  523: 
  524: Now, running
  525: 
  526:     xm create -c /usr/pkg/etc/xen/nbsd
  527: 
  528: should create a domain and load a NetBSD kernel in it. (Note: `-c`
  529: causes xm to connect to the domain's console once created.) The kernel
  530: will try to find its root file system on xbd0 (i.e., wd0e) which hasn't
  531: been created yet. wd0e will be seen as a disk device in the new domain,
  532: so it will be 'sub-partitioned'. We could attach a ccd to wd0e in
  533: *domain0* and partition it, newfs and extract the NetBSD/i386 or amd64
  534: tarballs there, but there's an easier way: load the
  535: `netbsd-INSTALL_XEN3_DOMU` kernel provided in the NetBSD binary sets.
  536: Like other install kernels, it contains a ramdisk with sysinst, so you
  537: can install NetBSD using sysinst on your new domain.
  538: 
  539: If you want to install NetBSD/Xen with a CDROM image, the following line
  540: should be used in the `/usr/pkg/etc/xen/nbsd` file:
  541: 
  542:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  543: 
  544: After booting the domain, the option to install via CDROM may be
  545: selected. The CDROM device should be changed to `xbd1d`.
  546: 
  547: Once done installing, `halt -p` the new domain (don't reboot or halt, it
  548: would reload the INSTALL\_XEN3\_DOMU kernel even if you changed the
  549: config file), switch the config file back to the XEN3\_DOMU kernel, and
  550: start the new domain again. Now it should be able to use `root on xbd0a`
  551: and you should have a second, functional NetBSD system on your xen
  552: installation.
  553: 
  554: When the new domain is booting you'll see some warnings about *wscons*
  555: and the pseudo-terminals. These can be fixed by editing the files
  556: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  557: `/etc/ttys`, except *console*, like this:
  558: 
  559:     console "/usr/libexec/getty Pc"         vt100   on secure
  560:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  561:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  562:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  563:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  564: 
  565: Finally, all screens must be commented out from `/etc/wscons.conf`.
  566: 
  567: It is also desirable to add
  568: 
  569:     powerd=YES
  570: 
  571: in rc.conf. This way, the domain will be properly shut down if
  572: `xm shutdown -R` or `xm shutdown -H` is used on the domain0.
  573: 
  574: Your domain should be now ready to work, enjoy.
  575: 
  576: Creating an unprivileged Linux domain (domU)
  577: --------------------------------------------
  578: 
  579: Creating unprivileged Linux domains isn't much different from
  580: unprivileged NetBSD domains, but there are some details to know.
  581: 
  582: First, the second parameter passed to the disk declaration (the '0x1' in
  583: the example below)
  584: 
  585:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  586: 
  587: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  588: for hda). Linux builds device numbers as: (major \<\< 8 + minor). So,
  589: hda1 which has major 3 and minor 1 on a Linux system will have device
  590: number 0x301. Alternatively, devices names can be used (hda, hdb, ...)
  591: as xentools has a table to map these names to devices numbers. To export
  592: a partition to a Linux guest we can use:
  593: 
  594:     disk = [ 'phy:/dev/wd0e,0x300,w' ]
  595:     root = "/dev/hda1 ro"
  596: 
  597: and it will appear as /dev/hda on the Linux system, and be used as root
  598: partition.
  599: 
  600: To install the Linux system on the partition to be exported to the guest
  601: domain, the following method can be used: install sysutils/e2fsprogs
  602: from pkgsrc. Use mke2fs to format the partition that will be the root
  603: partition of your Linux domain, and mount it. Then copy the files from a
  604: working Linux system, make adjustments in `/etc` (fstab, network
  605: config). It should also be possible to extract binary packages such as
  606: .rpm or .deb directly to the mounted partition using the appropriate
  607: tool, possibly running under NetBSD's Linux emulation. Once the
  608: filesystem has been populated, umount it. If desirable, the filesystem
  609: can be converted to ext3 using tune2fs -j. It should now be possible to
  610: boot the Linux guest domain, using one of the vmlinuz-\*-xenU kernels
  611: available in the Xen binary distribution.
  612: 
  613: To get the linux console right, you need to add:
  614: 
  615:     extra = "xencons=tty1"
  616: 
  617: to your configuration since not all linux distributions auto-attach a
  618: tty to the xen console.
  619: 
  620: Creating an unprivileged Solaris domain (domU)
  621: ----------------------------------------------
  622: 
  623: Download an Opensolaris [release](http://opensolaris.org/os/downloads/)
  624: or [development snapshot](http://genunix.org/) DVD image. Attach the DVD
  625: image to a MAN.VND.4 device. Copy the kernel and ramdisk filesystem
  626: image to your dom0 filesystem.
  627: 
  628:     dom0# mkdir /root/solaris
  629:     dom0# vnconfig vnd0 osol-1002-124-x86.iso
  630:     dom0# mount /dev/vnd0a /mnt
  631: 
  632:     ## for a 64-bit guest
  633:     dom0# cp /mnt/boot/amd64/x86.microroot /root/solaris
  634:     dom0# cp /mnt/platform/i86xpv/kernel/amd64/unix /root/solaris
  635: 
  636:     ## for a 32-bit guest
  637:     dom0# cp /mnt/boot/x86.microroot /root/solaris
  638:     dom0# cp /mnt/platform/i86xpv/kernel/unix /root/solaris
  639: 
  640:     dom0# umount /mnt
  641:           
  642: 
  643: Keep the MAN.VND.4 configured. For some reason the boot process stalls
  644: unless the DVD image is attached to the guest as a "phy" device. Create
  645: an initial configuration file with the following contents. Substitute
  646: */dev/wd0k* with an empty partition at least 8 GB large.
  647: 
  648:     memory = 640
  649:     name = 'solaris'
  650:     disk = [ 'phy:/dev/wd0k,0,w' ]
  651:     disk += [ 'phy:/dev/vnd0d,6:cdrom,r' ]
  652:     vif = [ 'bridge=bridge0' ]
  653:     kernel = '/root/solaris/unix'
  654:     ramdisk = '/root/solaris/x86.microroot'
  655:     # for a 64-bit guest
  656:     extra = '/platform/i86xpv/kernel/amd64/unix - nowin -B install_media=cdrom'
  657:     # for a 32-bit guest
  658:     #extra = '/platform/i86xpv/kernel/unix - nowin -B install_media=cdrom'
  659:           
  660: 
  661: Start the guest.
  662: 
  663:     dom0# xm create -c solaris.cfg
  664:     Started domain solaris
  665:                           v3.3.2 chgset 'unavailable'
  666:     SunOS Release 5.11 Version snv_124 64-bit
  667:     Copyright 1983-2009 Sun Microsystems, Inc.  All rights reserved.
  668:     Use is subject to license terms.
  669:     Hostname: opensolaris
  670:     Remounting root read/write
  671:     Probing for device nodes ...
  672:     WARNING: emlxs: ddi_modopen drv/fct failed: err 2
  673:     Preparing live image for use
  674:     Done mounting Live image
  675:           
  676: 
  677: Make sure the network is configured. Note that it can take a minute for
  678: the xnf0 interface to appear.
  679: 
  680:     opensolaris console login: jack
  681:     Password: jack
  682:     Sun Microsystems Inc.   SunOS 5.11      snv_124 November 2008
  683:     jack@opensolaris:~$ pfexec sh
  684:     sh-3.2# ifconfig -a
  685:     sh-3.2# exit
  686:           
  687: 
  688: Set a password for VNC and start the VNC server which provides the X11
  689: display where the installation program runs.
  690: 
  691:     jack@opensolaris:~$ vncpasswd
  692:     Password: solaris
  693:     Verify: solaris
  694:     jack@opensolaris:~$ cp .Xclients .vnc/xstartup
  695:     jack@opensolaris:~$ vncserver :1
  696:           
  697: 
  698: From a remote machine connect to the VNC server. Use `ifconfig xnf0` on
  699: the guest to find the correct IP address to use.
  700: 
  701:     remote$ vncviewer 172.18.2.99:1
  702:           
  703: 
  704: It is also possible to launch the installation on a remote X11 display.
  705: 
  706:     jack@opensolaris:~$ export DISPLAY=172.18.1.1:0
  707:     jack@opensolaris:~$ pfexec gui-install
  708:            
  709: 
  710: After the GUI installation is complete you will be asked to reboot.
  711: Before that you need to determine the ZFS ID for the new boot filesystem
  712: and update the configuration file accordingly. Return to the guest
  713: console.
  714: 
  715:     jack@opensolaris:~$ pfexec zdb -vvv rpool | grep bootfs
  716:                     bootfs = 43
  717:     ^C
  718:     jack@opensolaris:~$
  719:            
  720: 
  721: The final configuration file should look like this. Note in particular
  722: the last line.
  723: 
  724:     memory = 640
  725:     name = 'solaris'
  726:     disk = [ 'phy:/dev/wd0k,0,w' ]
  727:     vif = [ 'bridge=bridge0' ]
  728:     kernel = '/root/solaris/unix'
  729:     ramdisk = '/root/solaris/x86.microroot'
  730:     extra = '/platform/i86xpv/kernel/amd64/unix -B zfs-bootfs=rpool/43,bootpath="/xpvd/xdf@0:a"'
  731:            
  732: 
  733: Restart the guest to verify it works correctly.
  734: 
  735:     dom0# xm destroy solaris
  736:     dom0# xm create -c solaris.cfg
  737:     Using config file "./solaris.cfg".
  738:     v3.3.2 chgset 'unavailable'
  739:     Started domain solaris
  740:     SunOS Release 5.11 Version snv_124 64-bit
  741:     Copyright 1983-2009 Sun Microsystems, Inc.  All rights reserved.
  742:     Use is subject to license terms.
  743:     WARNING: emlxs: ddi_modopen drv/fct failed: err 2
  744:     Hostname: osol
  745:     Configuring devices.
  746:     Loading smf(5) service descriptions: 160/160
  747:     svccfg import warnings. See /var/svc/log/system-manifest-import:default.log .
  748:     Reading ZFS config: done.
  749:     Mounting ZFS filesystems: (6/6)
  750:     Creating new rsa public/private host key pair
  751:     Creating new dsa public/private host key pair
  752: 
  753:     osol console login:
  754:            
  755: 
  756: Using PCI devices in guest domains
  757: ----------------------------------
  758: 
  759: The domain0 can give other domains access to selected PCI devices. This
  760: can allow, for example, a non-privileged domain to have access to a
  761: physical network interface or disk controller. However, keep in mind
  762: that giving a domain access to a PCI device most likely will give the
  763: domain read/write access to the whole physical memory, as PCs don't have
  764: an IOMMU to restrict memory access to DMA-capable device. Also, it's not
  765: possible to export ISA devices to non-domain0 domains (which means that
  766: the primary VGA adapter can't be exported. A guest domain trying to
  767: access the VGA registers will panic).
  768: 
  769: This functionality is only available in NetBSD-5.1 (and later) domain0
  770: and domU. If the domain0 is NetBSD, it has to be running Xen 3.1, as
  771: support has not been ported to later versions at this time.
  772: 
  773: For a PCI device to be exported to a domU, is has to be attached to the
  774: `pciback` driver in domain0. Devices passed to the domain0 via the
  775: pciback.hide boot parameter will attach to `pciback` instead of the
  776: usual driver. The list of devices is specified as `(bus:dev.func)`,
  777: where bus and dev are 2-digit hexadecimal numbers, and func a
  778: single-digit number:
  779: 
  780:     pciback.hide=(00:0a.0)(00:06.0)
  781: 
  782: pciback devices should show up in the domain0's boot messages, and the
  783: devices should be listed in the `/kern/xen/pci` directory.
  784: 
  785: PCI devices to be exported to a domU are listed in the `pci` array of
  786: the domU's config file, with the format `'0000:bus:dev.func'`
  787: 
  788:     pci = [ '0000:00:06.0', '0000:00:0a.0' ]
  789: 
  790: In the domU an `xpci` device will show up, to which one or more pci
  791: busses will attach. Then the PCI drivers will attach to PCI busses as
  792: usual. Note that the default NetBSD DOMU kernels do not have `xpci` or
  793: any PCI drivers built in by default; you have to build your own kernel
  794: to use PCI devices in a domU. Here's a kernel config example:
  795: 
  796:     include         "arch/i386/conf/XEN3_DOMU"
  797:     #include         "arch/i386/conf/XENU"           # in NetBSD 3.0
  798: 
  799:     # Add support for PCI busses to the XEN3_DOMU kernel
  800:     xpci* at xenbus ?
  801:     pci* at xpci ?
  802: 
  803:     # Now add PCI and related devices to be used by this domain
  804:     # USB Controller and Devices
  805: 
  806:     # PCI USB controllers
  807:     uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
  808: 
  809:     # USB bus support
  810:     usb*    at uhci?
  811: 
  812:     # USB Hubs
  813:     uhub*   at usb?
  814:     uhub*   at uhub? port ? configuration ? interface ?
  815: 
  816:     # USB Mass Storage
  817:     umass*  at uhub? port ? configuration ? interface ?
  818:     wd*     at umass?
  819:     # SCSI controllers
  820:     ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
  821: 
  822:     # SCSI bus support (for both ahc and umass)
  823:     scsibus* at scsi?
  824: 
  825:     # SCSI devices
  826:     sd*     at scsibus? target ? lun ?      # SCSI disk drives
  827:     cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
  828: 
  829: Links and further information
  830: =============================
  831: 
  832: -   The [HowTo on Installing into RAID-1](http://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
  833:     explains how to set up booting a dom0 with Xen using grub 
  834:     with NetBSD's RAIDframe.  (This is obsolete with the use of
  835:     NetBSD's native boot.)
  836: -   An example of how to use NetBSD's native bootloader to load
  837:     NetBSD/Xen instead of Grub can be found in the i386/amd64 boot(8)
  838:     and boot.cfg(5) manpages.

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