Annotation of wikisrc/ports/xen/howto.mdwn, revision 1.18

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: 
                     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
1.13      gdt        40: -------------
1.12      gdt        41: 
                     42: Installing NetBSD/Xen is not extremely difficult, but it is more
                     43: complex than a normal installation of NetBSD.
1.15      gdt        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.
1.12      gdt        47: 
                     48: This HOWTO presumes a basic familiarity with the Xen system
1.16      gdt        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/).
1.1       mspo       52: 
1.15      gdt        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: 
1.18    ! gdt        94: Architecture
        !            95: ------------
        !            96: 
        !            97: Xen is basically amd64 only at this point.  One can either run i386
        !            98: domains or amd64 domains.  If running i386, PAE versions are required,
        !            99: for both dom0 and domU.  These versions are built by default in NetBSD
        !           100: releases.  While i386 dom0 works fine, amd64 is recommended as more
        !           101: normal.  (Note that emacs (at least) fails if run on i386 with PAE when
        !           102: built without, and vice versa, presumably due to bugs in the undump
        !           103: code.)
        !           104: 
1.15      gdt       105: Recommendation
                    106: --------------
                    107: 
1.18    ! gdt       108: Therefore, this HOWTO recommends running xenkernel42 (and xentools42),
        !           109: the NetBSD 6 stable branch, and to use amd64 as the dom0.  Either the
        !           110: i386 or amd64 of NetBSD may be used as domUs.
1.15      gdt       111: 
                    112: NetBSD as a dom0
                    113: ================
                    114: 
                    115: NetBSD can be used as a dom0 and works very well.  The following
                    116: sections address installation, updating NetBSD, and updating Xen.
                    117: 
                    118: Styles of dom0 operation
                    119: ------------------------
                    120: 
                    121: There are two basic ways to use Xen.  The traditional method is for
                    122: the dom0 to do absolutely nothing other than providing support to some
                    123: number of domUs.  Such a system was probably installed for the sole
                    124: purpose of hosting domUs, and sits in a server room on a UPS.
                    125: 
                    126: The other way is to put Xen under a normal-usage computer, so that the
                    127: dom0 is what the computer would have been without Xen, perhaps a
                    128: desktop or laptop.  Then, one can run domUs at will.  Purists will
                    129: deride this as less secure than the previous approach, and for a
                    130: computer whose purpose is to run domUs, they are right.  But Xen and a
                    131: dom0 (without domUs) is not meaingfully less secure than the same
                    132: things running without Xen.  One can boot Xen or boot regular NetBSD
                    133: alternately with little problems, simply refraining from starting the
                    134: Xen daemons when not running Xen.
                    135: 
                    136: Note that NetBSD as dom0 does not support multiple CPUs.  This will
                    137: limit the performance of the Xen/dom0 workstation approach.
                    138: 
                    139: Installation of NetBSD and Xen
                    140: ------------------------------
1.13      gdt       141: 
                    142: Note that it doesn't make sense to talk about installing a dom0 OS
                    143: without also installing Xen itself.
1.1       mspo      144: 
                    145: First do a NetBSD/i386 or NetBSD/amd64
                    146: [installation](../../docs/guide/en/chap-inst.html) of the 5.1 release
                    147: (or newer) as you usually do on x86 hardware. The binary releases are
1.5       mspo      148: available from [](ftp://ftp.NetBSD.org/pub/NetBSD/). Binary snapshots
                    149: for current and the stable branches are available on daily autobuilds.
                    150: If you plan to use the `grub` boot loader, when partitioning the disk
                    151: you have to make the root partition smaller than 512Mb, and formatted as
1.1       mspo      152: FFSv1 with 8k block/1k fragments. If the partition is larger than this,
                    153: uses FFSv2 or has different block/fragment sizes, grub may fail to load
                    154: some files. Also keep in mind that you'll probably want to provide
                    155: virtual disks to other domains, so reserve some partitions for these
                    156: virtual disks. Alternatively, you can create large files in the file
                    157: system, map them to vnd(4) devices and export theses vnd devices to
                    158: other domains.
                    159: 
                    160: Next step is to install the Xen packages via pkgsrc or from binary
                    161: packages. See [the pkgsrc
                    162: documentation](http://www.NetBSD.org/docs/pkgsrc/) if you are unfamiliar
                    163: with pkgsrc and/or handling of binary packages. Xen 3.1, 3.3, 4.1 and
                    164: 4.2 are available. 3.1 supports PCI pass-through while other versions do
1.5       mspo      165: not. You'll need either `sysutils/xentools3` and `sysutils/xenkernel3`
                    166: for Xen 3.1, `sysutils/xentools33` and `sysutils/xenkernel33` for Xen
                    167: 3.3, `sysutils/xentools41` and `sysutils/xenkernel41` for Xen 4.1. or
                    168: `sysutils/xentools42` and `sysutils/xenkernel42` for Xen 4.2. You'll
                    169: also need `sysutils/grub` if you plan do use the grub boot loader. If
                    170: using Xen 3.1, you may also want to install `sysutils/xentools3-hvm`
1.1       mspo      171: which contains the utilities to run unmodified guests OSes using the
                    172: *HVM* support (for later versions this is included in
1.5       mspo      173: `sysutils/xentools`). Note that your CPU needs to support this. Intel
                    174: CPUs must have the 'VT' instruction, AMD CPUs the 'SVM' instruction. You
                    175: can easily find out if your CPU support HVM by using NetBSD's cpuctl
                    176: command:
1.1       mspo      177: 
1.3       mspo      178:     # cpuctl identify 0
                    179:     cpu0: Intel Core 2 (Merom) (686-class), id 0x6f6
                    180:     cpu0: features 0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR>
                    181:     cpu0: features 0xbfebfbff<PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX>
                    182:     cpu0: features 0xbfebfbff<FXSR,SSE,SSE2,SS,HTT,TM,SBF>
1.5       mspo      183:     cpu0: features2 0x4e33d<SSE3,DTES64,MONITOR,DS-CPL,,TM2,SSSE3,CX16,xTPR,PDCM,DCA>
1.3       mspo      184:     cpu0: features3 0x20100800<SYSCALL/SYSRET,XD,EM64T>
                    185:     cpu0: "Intel(R) Xeon(R) CPU            5130  @ 2.00GHz"
                    186:     cpu0: I-cache 32KB 64B/line 8-way, D-cache 32KB 64B/line 8-way
                    187:     cpu0: L2 cache 4MB 64B/line 16-way
                    188:     cpu0: ITLB 128 4KB entries 4-way
                    189:     cpu0: DTLB 256 4KB entries 4-way, 32 4MB entries 4-way
                    190:     cpu0: Initial APIC ID 0
                    191:     cpu0: Cluster/Package ID 0
                    192:     cpu0: Core ID 0
                    193:     cpu0: family 06 model 0f extfamily 00 extmodel 00
1.1       mspo      194: 
                    195: Depending on your CPU, the feature you are looking for is called HVM,
                    196: SVM or VMX.
                    197: 
                    198: Next you need to copy the selected Xen kernel itself. pkgsrc installed
1.5       mspo      199: them under `/usr/pkg/xen*-kernel/`. The file you're looking for is
                    200: `xen.gz`. Copy it to your root file system. `xen-debug.gz` is a kernel
                    201: with more consistency checks and more details printed on the serial
                    202: console. It is useful for debugging crashing guests if you use a serial
                    203: console. It is not useful with a VGA console.
1.1       mspo      204: 
                    205: You'll then need a NetBSD/Xen kernel for *domain0* on your root file
                    206: system. The XEN3PAE\_DOM0 kernel or XEN3\_DOM0 provided as part of the
                    207: i386 or amd64 binaries is suitable for this, but you may want to
                    208: customize it. Keep your native kernel around, as it can be useful for
1.5       mspo      209: recovery. *Note:* the *domain0* kernel must support KERNFS and `/kern`
                    210: must be mounted because *xend* needs access to `/kern/xen/privcmd`.
                    211: 
                    212: Next you need to get a bootloader to load the `xen.gz` kernel, and the
                    213: NetBSD *domain0* kernel as a module. This can be `grub` or NetBSD's boot
                    214: loader. Below is a detailled example for grub, see the boot.cfg(5)
                    215: manual page for an example using the latter.
1.1       mspo      216: 
                    217: This is also where you'll specify the memory allocated to *domain0*, the
                    218: console to use, etc ...
                    219: 
1.5       mspo      220: Here is a commented `/grub/menu.lst` file:
1.1       mspo      221: 
1.5       mspo      222:     #Grub config file for NetBSD/xen. Copy as /grub/menu.lst and run
1.3       mspo      223:     # grub-install /dev/rwd0d (assuming your boot device is wd0).
                    224:     #
                    225:     # The default entry to load will be the first one
                    226:     default=0
1.5       mspo      227: 
1.3       mspo      228:     # boot the default entry after 10s if the user didn't hit keyboard
                    229:     timeout=10
1.5       mspo      230: 
1.3       mspo      231:     # Configure serial port to use as console. Ignore if you'll use VGA only
                    232:     serial --unit=0 --speed=115200 --word=8 --parity=no --stop=1
1.5       mspo      233: 
1.3       mspo      234:     # Let the user select which console to use (serial or VGA), default
                    235:     # to serial after 10s
                    236:     terminal --timeout=10 serial console
1.5       mspo      237: 
1.3       mspo      238:     # An entry for NetBSD/xen, using /netbsd as the domain0 kernel, and serial
                    239:     # console. Domain0 will have 64MB RAM allocated.
                    240:     # Assume NetBSD is installed in the first MBR partition.
                    241:     title Xen 3 / NetBSD (hda0, serial)
                    242:       root(hd0,0)
                    243:       kernel (hd0,a)/xen.gz dom0_mem=65536 com1=115200,8n1
                    244:       module (hd0,a)/netbsd bootdev=wd0a ro console=ttyS0
1.5       mspo      245: 
1.3       mspo      246:     # Same as above, but using VGA console
                    247:     # We can use console=tty0 (Linux syntax) or console=pc (NetBSD syntax)
                    248:     title Xen 3 / NetBSD (hda0, vga)
                    249:       root(hd0,0)
                    250:       kernel (hd0,a)/xen.gz dom0_mem=65536
                    251:       module (hd0,a)/netbsd bootdev=wd0a ro console=tty0
1.5       mspo      252: 
1.3       mspo      253:     # NetBSD/xen using a backup domain0 kernel (in case you installed a
                    254:     # nonworking kernel as /netbsd
                    255:     title Xen 3 / NetBSD (hda0, backup, serial)
                    256:       root(hd0,0)
                    257:       kernel (hd0,a)/xen.gz dom0_mem=65536 com1=115200,8n1
                    258:       module (hd0,a)/netbsd.backup bootdev=wd0a ro console=ttyS0
                    259:     title Xen 3 / NetBSD (hda0, backup, VGA)
                    260:       root(hd0,0)
                    261:       kernel (hd0,a)/xen.gz dom0_mem=65536
                    262:       module (hd0,a)/netbsd.backup bootdev=wd0a ro console=tty0
1.5       mspo      263: 
1.3       mspo      264:     #Load a regular NetBSD/i386 kernel. Can be useful if you end up with a
                    265:     #nonworking /xen.gz
                    266:     title NetBSD 5.1
                    267:       root (hd0,a)
                    268:       kernel --type=netbsd /netbsd-GENERIC
1.5       mspo      269: 
1.3       mspo      270:     #Load the NetBSD bootloader, letting it load the NetBSD/i386 kernel.
                    271:     #May be better than the above, as grub can't pass all required infos
                    272:     #to the NetBSD/i386 kernel (e.g. console, root device, ...)
                    273:     title NetBSD chain
                    274:       root        (hd0,0)
                    275:       chainloader +1
1.5       mspo      276: 
1.3       mspo      277:     ## end of grub config file.
1.5       mspo      278:           
                    279: 
1.1       mspo      280: Install grub with the following command:
                    281: 
1.3       mspo      282:     # grub --no-floppy
1.5       mspo      283: 
1.3       mspo      284:     grub> root (hd0,a)
                    285:      Filesystem type is ffs, partition type 0xa9
1.5       mspo      286: 
1.3       mspo      287:     grub> setup (hd0)
                    288:      Checking if "/boot/grub/stage1" exists... no
                    289:      Checking if "/grub/stage1" exists... yes
                    290:      Checking if "/grub/stage2" exists... yes
                    291:      Checking if "/grub/ffs_stage1_5" exists... yes
                    292:      Running "embed /grub/ffs_stage1_5 (hd0)"...  14 sectors are embedded.
                    293:     succeeded
                    294:      Running "install /grub/stage1 (hd0) (hd0)1+14 p (hd0,0,a)/grub/stage2 /grub/menu.lst"...
                    295:      succeeded
                    296:     Done.
1.5       mspo      297:           
1.1       mspo      298: 
1.15      gdt       299: Updating NetBSD in a dom0
                    300: -------------------------
                    301: 
                    302: This is just like updating NetBSD on bare hardware, assuming the new
                    303: version supports the version of Xen you are running.  Generally, one
                    304: replaces the kernel and reboots, and then overlays userland binaries
                    305: and adjusts /etc.
                    306: 
                    307: Note that one must update both the non-Xen kernel typically used for
                    308: rescue purposes and the DOM0 kernel used with Xen.
                    309: 
                    310: Updating Xen versions
                    311: ---------------------
                    312: 
                    313: TODO: write
                    314: 
1.14      gdt       315: Creating unprivileged domains (domU)
                    316: ====================================
                    317: 
                    318: Creating domUs is almost entirely independent of operating system.  We
                    319: first explain NetBSD, and then differences for Linux and Solaris.
                    320: 
                    321: Creating an unprivileged NetBSD domain (domU)
                    322: ---------------------------------------------
1.1       mspo      323: 
                    324: Once you have *domain0* running, you need to start the xen tool daemon
1.5       mspo      325: (`/usr/pkg/share/examples/rc.d/xend start`) and the xen backend daemon
                    326: (`/usr/pkg/share/examples/rc.d/xenbackendd start` for Xen3\*,
                    327: `/usr/pkg/share/examples/rc.d/xencommons start` for Xen4.\*). Make sure
                    328: that `/dev/xencons` and `/dev/xenevt` exist before starting `xend`. You
                    329: can create them with this command:
1.1       mspo      330: 
1.3       mspo      331:     # cd /dev && sh MAKEDEV xen
1.1       mspo      332: 
1.5       mspo      333: xend will write logs to `/var/log/xend.log` and
                    334: `/var/log/xend-debug.log`. You can then control xen with the xm tool.
                    335: 'xm list' will show something like:
1.1       mspo      336: 
1.3       mspo      337:     # xm list
                    338:     Name              Id  Mem(MB)  CPU  State  Time(s)  Console
                    339:     Domain-0           0       64    0  r----     58.1
1.1       mspo      340: 
                    341: 'xm create' allows you to create a new domain. It uses a config file in
                    342: PKG\_SYSCONFDIR for its parameters. By default, this file will be in
1.5       mspo      343: `/usr/pkg/etc/xen/`. On creation, a kernel has to be specified, which
                    344: will be executed in the new domain (this kernel is in the *domain0* file
                    345: system, not on the new domain virtual disk; but please note, you should
                    346: install the same kernel into *domainU* as `/netbsd` in order to make
                    347: your system tools, like MAN.SAVECORE.8, work). A suitable kernel is
                    348: provided as part of the i386 and amd64 binary sets: XEN3\_DOMU.
1.1       mspo      349: 
                    350: Here is an /usr/pkg/etc/xen/nbsd example config file:
                    351: 
1.3       mspo      352:     #  -*- mode: python; -*-
                    353:     #============================================================================
                    354:     # Python defaults setup for 'xm create'.
                    355:     # Edit this file to reflect the configuration of your system.
                    356:     #============================================================================
1.5       mspo      357: 
1.3       mspo      358:     #----------------------------------------------------------------------------
                    359:     # Kernel image file. This kernel will be loaded in the new domain.
                    360:     kernel = "/home/bouyer/netbsd-XEN3_DOMU"
                    361:     #kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
1.5       mspo      362: 
1.3       mspo      363:     # Memory allocation (in megabytes) for the new domain.
                    364:     memory = 128
1.5       mspo      365: 
1.3       mspo      366:     # A handy name for your new domain. This will appear in 'xm list',
                    367:     # and you can use this as parameters for xm in place of the domain
                    368:     # number. All domains must have different names.
                    369:     #
                    370:     name = "nbsd"
1.5       mspo      371: 
1.3       mspo      372:     # The number of virtual CPUs this domain has.
                    373:     #
                    374:     vcpus = 1
1.5       mspo      375: 
1.3       mspo      376:     #----------------------------------------------------------------------------
                    377:     # Define network interfaces for the new domain.
1.5       mspo      378: 
1.3       mspo      379:     # Number of network interfaces (must be at least 1). Default is 1.
                    380:     nics = 1
1.5       mspo      381: 
1.3       mspo      382:     # Define MAC and/or bridge for the network interfaces.
                    383:     #
                    384:     # The MAC address specified in ``mac'' is the one used for the interface
                    385:     # in the new domain. The interface in domain0 will use this address XOR'd
                    386:     # with 00:00:00:01:00:00 (i.e. aa:00:00:51:02:f0 in our example). Random
                    387:     # MACs are assigned if not given.
                    388:     #
                    389:     # ``bridge'' is a required parameter, which will be passed to the
                    390:     # vif-script called by xend(8) when a new domain is created to configure
                    391:     # the new xvif interface in domain0.
                    392:     #
                    393:     # In this example, the xvif is added to bridge0, which should have been
                    394:     # set up prior to the new domain being created -- either in the
                    395:     # ``network'' script or using a /etc/ifconfig.bridge0 file.
                    396:     #
                    397:     vif = [ 'mac=aa:00:00:50:02:f0, bridge=bridge0' ]
1.5       mspo      398: 
1.3       mspo      399:     #----------------------------------------------------------------------------
                    400:     # Define the disk devices you want the domain to have access to, and
                    401:     # what you want them accessible as.
                    402:     #
                    403:     # Each disk entry is of the form:
                    404:     #
1.5       mspo      405:     #   phy:DEV,VDEV,MODE
1.3       mspo      406:     #
                    407:     # where DEV is the device, VDEV is the device name the domain will see,
                    408:     # and MODE is r for read-only, w for read-write.  You can also create
                    409:     # file-backed domains using disk entries of the form:
                    410:     #
1.5       mspo      411:     #   file:PATH,VDEV,MODE
1.3       mspo      412:     #
                    413:     # where PATH is the path to the file used as the virtual disk, and VDEV
                    414:     # and MODE have the same meaning as for ``phy'' devices.
                    415:     #
                    416:     # VDEV doesn't really matter for a NetBSD guest OS (it's just used as an index),
                    417:     # but it does for Linux.
                    418:     # Worse, the device has to exist in /dev/ of domain0, because xm will
                    419:     # try to stat() it. This means that in order to load a Linux guest OS
                    420:     # from a NetBSD domain0, you'll have to create /dev/hda1, /dev/hda2, ...
                    421:     # on domain0, with the major/minor from Linux :(
                    422:     # Alternatively it's possible to specify the device number in hex,
                    423:     # e.g. 0x301 for /dev/hda1, 0x302 for /dev/hda2, etc ...
1.5       mspo      424: 
1.3       mspo      425:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
                    426:     #disk = [ 'file:/var/xen/nbsd-disk,0x01,w' ]
                    427:     #disk = [ 'file:/var/xen/nbsd-disk,0x301,w' ]
1.5       mspo      428: 
1.3       mspo      429:     #----------------------------------------------------------------------------
                    430:     # Set the kernel command line for the new domain.
1.5       mspo      431: 
1.3       mspo      432:     # Set root device. This one does matter for NetBSD
                    433:     root = "xbd0"
                    434:     # extra parameters passed to the kernel
                    435:     # this is where you can set boot flags like -s, -a, etc ...
                    436:     #extra = ""
1.5       mspo      437: 
1.3       mspo      438:     #----------------------------------------------------------------------------
                    439:     # Set according to whether you want the domain restarted when it exits.
                    440:     # The default is False.
                    441:     #autorestart = True
1.5       mspo      442: 
1.3       mspo      443:     # end of nbsd config file ====================================================
1.1       mspo      444: 
                    445: When a new domain is created, xen calls the
1.5       mspo      446: `/usr/pkg/etc/xen/vif-bridge` script for each virtual network interface
                    447: created in *domain0*. This can be used to automatically configure the
                    448: xvif?.? interfaces in *domain0*. In our example, these will be bridged
                    449: with the bridge0 device in *domain0*, but the bridge has to exist first.
                    450: To do this, create the file `/etc/ifconfig.bridge0` and make it look
                    451: like this:
1.1       mspo      452: 
1.3       mspo      453:     create
                    454:     !brconfig $int add ex0 up
1.1       mspo      455: 
1.5       mspo      456: (replace `ex0` with the name of your physical interface). Then bridge0
                    457: will be created on boot. See the MAN.BRIDGE.4 man page for details.
1.1       mspo      458: 
1.5       mspo      459: So, here is a suitable `/usr/pkg/etc/xen/vif-bridge` for xvif?.? (a
                    460: working vif-bridge is also provided with xentools20) configuring:
1.1       mspo      461: 
1.5       mspo      462:     #!/bin/sh
1.3       mspo      463:     #============================================================================
1.18    ! gdt       464:     # $NetBSD: howto.mdwn,v 1.17 2014/12/24 00:06:31 gdt Exp $
1.3       mspo      465:     #
                    466:     # /usr/pkg/etc/xen/vif-bridge
                    467:     #
                    468:     # Script for configuring a vif in bridged mode with a dom0 interface.
                    469:     # The xend(8) daemon calls a vif script when bringing a vif up or down.
                    470:     # The script name to use is defined in /usr/pkg/etc/xen/xend-config.sxp
                    471:     # in the ``vif-script'' field.
                    472:     #
                    473:     # Usage: vif-bridge up|down [var=value ...]
                    474:     #
                    475:     # Actions:
1.5       mspo      476:     #    up     Adds the vif interface to the bridge.
                    477:     #    down   Removes the vif interface from the bridge.
1.3       mspo      478:     #
                    479:     # Variables:
1.5       mspo      480:     #    domain name of the domain the interface is on (required).
                    481:     #    vifq   vif interface name (required).
                    482:     #    mac    vif MAC address (required).
                    483:     #    bridge bridge to add the vif to (required).
1.3       mspo      484:     #
                    485:     # Example invocation:
                    486:     #
                    487:     # vif-bridge up domain=VM1 vif=xvif1.0 mac="ee:14:01:d0:ec:af" bridge=bridge0
                    488:     #
                    489:     #============================================================================
1.5       mspo      490: 
1.3       mspo      491:     # Exit if anything goes wrong
                    492:     set -e
1.5       mspo      493: 
1.3       mspo      494:     echo "vif-bridge $*"
1.5       mspo      495: 
1.3       mspo      496:     # Operation name.
                    497:     OP=$1; shift
1.5       mspo      498: 
1.3       mspo      499:     # Pull variables in args into environment
                    500:     for arg ; do export "${arg}" ; done
1.5       mspo      501: 
1.3       mspo      502:     # Required parameters. Fail if not set.
                    503:     domain=${domain:?}
                    504:     vif=${vif:?}
                    505:     mac=${mac:?}
                    506:     bridge=${bridge:?}
1.5       mspo      507: 
1.3       mspo      508:     # Optional parameters. Set defaults.
                    509:     ip=${ip:-''}   # default to null (do nothing)
1.5       mspo      510: 
1.3       mspo      511:     # Are we going up or down?
                    512:     case $OP in
1.5       mspo      513:     up) brcmd='add' ;;
1.3       mspo      514:     down)   brcmd='delete' ;;
                    515:     *)
1.5       mspo      516:         echo 'Invalid command: ' $OP
                    517:         echo 'Valid commands are: up, down'
                    518:         exit 1
                    519:         ;;
1.3       mspo      520:     esac
1.5       mspo      521: 
1.3       mspo      522:     # Don't do anything if the bridge is "null".
                    523:     if [ "${bridge}" = "null" ] ; then
1.5       mspo      524:         exit
1.3       mspo      525:     fi
1.5       mspo      526: 
1.3       mspo      527:     # Don't do anything if the bridge doesn't exist.
                    528:     if ! ifconfig -l | grep "${bridge}" >/dev/null; then
1.5       mspo      529:         exit
1.3       mspo      530:     fi
1.5       mspo      531: 
1.3       mspo      532:     # Add/remove vif to/from bridge.
                    533:     ifconfig x${vif} $OP
                    534:     brconfig ${bridge} ${brcmd} x${vif}
1.1       mspo      535: 
                    536: Now, running
                    537: 
1.3       mspo      538:     xm create -c /usr/pkg/etc/xen/nbsd
1.1       mspo      539: 
1.5       mspo      540: should create a domain and load a NetBSD kernel in it. (Note: `-c`
                    541: causes xm to connect to the domain's console once created.) The kernel
                    542: will try to find its root file system on xbd0 (i.e., wd0e) which hasn't
                    543: been created yet. wd0e will be seen as a disk device in the new domain,
                    544: so it will be 'sub-partitioned'. We could attach a ccd to wd0e in
                    545: *domain0* and partition it, newfs and extract the NetBSD/i386 or amd64
                    546: tarballs there, but there's an easier way: load the
                    547: `netbsd-INSTALL_XEN3_DOMU` kernel provided in the NetBSD binary sets.
                    548: Like other install kernels, it contains a ramdisk with sysinst, so you
                    549: can install NetBSD using sysinst on your new domain.
1.1       mspo      550: 
                    551: If you want to install NetBSD/Xen with a CDROM image, the following line
1.5       mspo      552: should be used in the `/usr/pkg/etc/xen/nbsd` file:
1.1       mspo      553: 
1.3       mspo      554:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
1.1       mspo      555: 
                    556: After booting the domain, the option to install via CDROM may be
1.5       mspo      557: selected. The CDROM device should be changed to `xbd1d`.
1.1       mspo      558: 
1.5       mspo      559: Once done installing, `halt -p` the new domain (don't reboot or halt, it
                    560: would reload the INSTALL\_XEN3\_DOMU kernel even if you changed the
1.1       mspo      561: config file), switch the config file back to the XEN3\_DOMU kernel, and
1.5       mspo      562: start the new domain again. Now it should be able to use `root on xbd0a`
                    563: and you should have a second, functional NetBSD system on your xen
                    564: installation.
1.1       mspo      565: 
                    566: When the new domain is booting you'll see some warnings about *wscons*
                    567: and the pseudo-terminals. These can be fixed by editing the files
1.5       mspo      568: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
                    569: `/etc/ttys`, except *console*, like this:
1.1       mspo      570: 
1.3       mspo      571:     console "/usr/libexec/getty Pc"         vt100   on secure
                    572:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
                    573:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
                    574:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
                    575:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
1.1       mspo      576: 
1.5       mspo      577: Finally, all screens must be commented out from `/etc/wscons.conf`.
1.1       mspo      578: 
                    579: It is also desirable to add
                    580: 
1.3       mspo      581:     powerd=YES
1.1       mspo      582: 
1.5       mspo      583: in rc.conf. This way, the domain will be properly shut down if
                    584: `xm shutdown -R` or `xm shutdown -H` is used on the domain0.
1.1       mspo      585: 
                    586: Your domain should be now ready to work, enjoy.
                    587: 
1.14      gdt       588: Creating an unprivileged Linux domain (domU)
1.5       mspo      589: --------------------------------------------
1.1       mspo      590: 
                    591: Creating unprivileged Linux domains isn't much different from
                    592: unprivileged NetBSD domains, but there are some details to know.
                    593: 
                    594: First, the second parameter passed to the disk declaration (the '0x1' in
                    595: the example below)
                    596: 
1.3       mspo      597:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
1.1       mspo      598: 
                    599: does matter to Linux. It wants a Linux device number here (e.g. 0x300
                    600: for hda). Linux builds device numbers as: (major \<\< 8 + minor). So,
                    601: hda1 which has major 3 and minor 1 on a Linux system will have device
                    602: number 0x301. Alternatively, devices names can be used (hda, hdb, ...)
                    603: as xentools has a table to map these names to devices numbers. To export
                    604: a partition to a Linux guest we can use:
                    605: 
1.3       mspo      606:     disk = [ 'phy:/dev/wd0e,0x300,w' ]
                    607:     root = "/dev/hda1 ro"
1.1       mspo      608: 
                    609: and it will appear as /dev/hda on the Linux system, and be used as root
                    610: partition.
                    611: 
                    612: To install the Linux system on the partition to be exported to the guest
                    613: domain, the following method can be used: install sysutils/e2fsprogs
                    614: from pkgsrc. Use mke2fs to format the partition that will be the root
                    615: partition of your Linux domain, and mount it. Then copy the files from a
1.5       mspo      616: working Linux system, make adjustments in `/etc` (fstab, network
                    617: config). It should also be possible to extract binary packages such as
                    618: .rpm or .deb directly to the mounted partition using the appropriate
                    619: tool, possibly running under NetBSD's Linux emulation. Once the
                    620: filesystem has been populated, umount it. If desirable, the filesystem
                    621: can be converted to ext3 using tune2fs -j. It should now be possible to
                    622: boot the Linux guest domain, using one of the vmlinuz-\*-xenU kernels
                    623: available in the Xen binary distribution.
1.1       mspo      624: 
                    625: To get the linux console right, you need to add:
                    626: 
1.3       mspo      627:     extra = "xencons=tty1"
1.1       mspo      628: 
                    629: to your configuration since not all linux distributions auto-attach a
                    630: tty to the xen console.
                    631: 
1.14      gdt       632: Creating an unprivileged Solaris domain (domU)
1.5       mspo      633: ----------------------------------------------
1.1       mspo      634: 
                    635: Download an Opensolaris [release](http://opensolaris.org/os/downloads/)
                    636: or [development snapshot](http://genunix.org/) DVD image. Attach the DVD
1.5       mspo      637: image to a MAN.VND.4 device. Copy the kernel and ramdisk filesystem
                    638: image to your dom0 filesystem.
1.1       mspo      639: 
1.3       mspo      640:     dom0# mkdir /root/solaris
                    641:     dom0# vnconfig vnd0 osol-1002-124-x86.iso
                    642:     dom0# mount /dev/vnd0a /mnt
1.5       mspo      643: 
1.3       mspo      644:     ## for a 64-bit guest
                    645:     dom0# cp /mnt/boot/amd64/x86.microroot /root/solaris
                    646:     dom0# cp /mnt/platform/i86xpv/kernel/amd64/unix /root/solaris
1.5       mspo      647: 
1.3       mspo      648:     ## for a 32-bit guest
                    649:     dom0# cp /mnt/boot/x86.microroot /root/solaris
                    650:     dom0# cp /mnt/platform/i86xpv/kernel/unix /root/solaris
1.5       mspo      651: 
1.3       mspo      652:     dom0# umount /mnt
1.5       mspo      653:           
                    654: 
                    655: Keep the MAN.VND.4 configured. For some reason the boot process stalls
                    656: unless the DVD image is attached to the guest as a "phy" device. Create
                    657: an initial configuration file with the following contents. Substitute
                    658: */dev/wd0k* with an empty partition at least 8 GB large.
1.1       mspo      659: 
1.4       mspo      660:     memory = 640
                    661:     name = 'solaris'
                    662:     disk = [ 'phy:/dev/wd0k,0,w' ]
                    663:     disk += [ 'phy:/dev/vnd0d,6:cdrom,r' ]
                    664:     vif = [ 'bridge=bridge0' ]
                    665:     kernel = '/root/solaris/unix'
                    666:     ramdisk = '/root/solaris/x86.microroot'
                    667:     # for a 64-bit guest
                    668:     extra = '/platform/i86xpv/kernel/amd64/unix - nowin -B install_media=cdrom'
                    669:     # for a 32-bit guest
                    670:     #extra = '/platform/i86xpv/kernel/unix - nowin -B install_media=cdrom'
1.5       mspo      671:           
                    672: 
1.1       mspo      673: Start the guest.
                    674: 
1.4       mspo      675:     dom0# xm create -c solaris.cfg
                    676:     Started domain solaris
                    677:                           v3.3.2 chgset 'unavailable'
                    678:     SunOS Release 5.11 Version snv_124 64-bit
                    679:     Copyright 1983-2009 Sun Microsystems, Inc.  All rights reserved.
                    680:     Use is subject to license terms.
                    681:     Hostname: opensolaris
                    682:     Remounting root read/write
                    683:     Probing for device nodes ...
                    684:     WARNING: emlxs: ddi_modopen drv/fct failed: err 2
                    685:     Preparing live image for use
                    686:     Done mounting Live image
1.5       mspo      687:           
1.1       mspo      688: 
                    689: Make sure the network is configured. Note that it can take a minute for
                    690: the xnf0 interface to appear.
                    691: 
1.4       mspo      692:     opensolaris console login: jack
                    693:     Password: jack
                    694:     Sun Microsystems Inc.   SunOS 5.11      snv_124 November 2008
                    695:     jack@opensolaris:~$ pfexec sh
                    696:     sh-3.2# ifconfig -a
                    697:     sh-3.2# exit
1.5       mspo      698:           
1.1       mspo      699: 
                    700: Set a password for VNC and start the VNC server which provides the X11
                    701: display where the installation program runs.
                    702: 
1.4       mspo      703:     jack@opensolaris:~$ vncpasswd
                    704:     Password: solaris
                    705:     Verify: solaris
                    706:     jack@opensolaris:~$ cp .Xclients .vnc/xstartup
                    707:     jack@opensolaris:~$ vncserver :1
1.5       mspo      708:           
1.1       mspo      709: 
1.5       mspo      710: From a remote machine connect to the VNC server. Use `ifconfig xnf0` on
                    711: the guest to find the correct IP address to use.
1.1       mspo      712: 
1.4       mspo      713:     remote$ vncviewer 172.18.2.99:1
1.5       mspo      714:           
1.1       mspo      715: 
                    716: It is also possible to launch the installation on a remote X11 display.
                    717: 
1.4       mspo      718:     jack@opensolaris:~$ export DISPLAY=172.18.1.1:0
                    719:     jack@opensolaris:~$ pfexec gui-install
1.5       mspo      720:            
1.1       mspo      721: 
                    722: After the GUI installation is complete you will be asked to reboot.
                    723: Before that you need to determine the ZFS ID for the new boot filesystem
                    724: and update the configuration file accordingly. Return to the guest
                    725: console.
                    726: 
1.4       mspo      727:     jack@opensolaris:~$ pfexec zdb -vvv rpool | grep bootfs
                    728:                     bootfs = 43
                    729:     ^C
                    730:     jack@opensolaris:~$
1.5       mspo      731:            
1.1       mspo      732: 
                    733: The final configuration file should look like this. Note in particular
                    734: the last line.
                    735: 
1.4       mspo      736:     memory = 640
                    737:     name = 'solaris'
                    738:     disk = [ 'phy:/dev/wd0k,0,w' ]
                    739:     vif = [ 'bridge=bridge0' ]
                    740:     kernel = '/root/solaris/unix'
                    741:     ramdisk = '/root/solaris/x86.microroot'
                    742:     extra = '/platform/i86xpv/kernel/amd64/unix -B zfs-bootfs=rpool/43,bootpath="/xpvd/xdf@0:a"'
1.5       mspo      743:            
1.1       mspo      744: 
                    745: Restart the guest to verify it works correctly.
                    746: 
1.4       mspo      747:     dom0# xm destroy solaris
                    748:     dom0# xm create -c solaris.cfg
                    749:     Using config file "./solaris.cfg".
                    750:     v3.3.2 chgset 'unavailable'
                    751:     Started domain solaris
                    752:     SunOS Release 5.11 Version snv_124 64-bit
                    753:     Copyright 1983-2009 Sun Microsystems, Inc.  All rights reserved.
                    754:     Use is subject to license terms.
                    755:     WARNING: emlxs: ddi_modopen drv/fct failed: err 2
                    756:     Hostname: osol
                    757:     Configuring devices.
                    758:     Loading smf(5) service descriptions: 160/160
                    759:     svccfg import warnings. See /var/svc/log/system-manifest-import:default.log .
                    760:     Reading ZFS config: done.
                    761:     Mounting ZFS filesystems: (6/6)
                    762:     Creating new rsa public/private host key pair
                    763:     Creating new dsa public/private host key pair
1.5       mspo      764: 
1.4       mspo      765:     osol console login:
1.5       mspo      766:            
1.1       mspo      767: 
                    768: Using PCI devices in guest domains
1.14      gdt       769: ----------------------------------
1.1       mspo      770: 
                    771: The domain0 can give other domains access to selected PCI devices. This
                    772: can allow, for example, a non-privileged domain to have access to a
                    773: physical network interface or disk controller. However, keep in mind
                    774: that giving a domain access to a PCI device most likely will give the
                    775: domain read/write access to the whole physical memory, as PCs don't have
                    776: an IOMMU to restrict memory access to DMA-capable device. Also, it's not
                    777: possible to export ISA devices to non-domain0 domains (which means that
                    778: the primary VGA adapter can't be exported. A guest domain trying to
                    779: access the VGA registers will panic).
                    780: 
                    781: This functionality is only available in NetBSD-5.1 (and later) domain0
                    782: and domU. If the domain0 is NetBSD, it has to be running Xen 3.1, as
                    783: support has not been ported to later versions at this time.
                    784: 
                    785: For a PCI device to be exported to a domU, is has to be attached to the
1.5       mspo      786: `pciback` driver in domain0. Devices passed to the domain0 via the
                    787: pciback.hide boot parameter will attach to `pciback` instead of the
                    788: usual driver. The list of devices is specified as `(bus:dev.func)`,
                    789: where bus and dev are 2-digit hexadecimal numbers, and func a
                    790: single-digit number:
1.1       mspo      791: 
1.4       mspo      792:     pciback.hide=(00:0a.0)(00:06.0)
1.1       mspo      793: 
                    794: pciback devices should show up in the domain0's boot messages, and the
1.5       mspo      795: devices should be listed in the `/kern/xen/pci` directory.
1.1       mspo      796: 
1.5       mspo      797: PCI devices to be exported to a domU are listed in the `pci` array of
                    798: the domU's config file, with the format `'0000:bus:dev.func'`
1.1       mspo      799: 
1.4       mspo      800:     pci = [ '0000:00:06.0', '0000:00:0a.0' ]
1.1       mspo      801: 
1.5       mspo      802: In the domU an `xpci` device will show up, to which one or more pci
                    803: busses will attach. Then the PCI drivers will attach to PCI busses as
                    804: usual. Note that the default NetBSD DOMU kernels do not have `xpci` or
                    805: any PCI drivers built in by default; you have to build your own kernel
                    806: to use PCI devices in a domU. Here's a kernel config example:
1.1       mspo      807: 
1.4       mspo      808:     include         "arch/i386/conf/XEN3_DOMU"
                    809:     #include         "arch/i386/conf/XENU"           # in NetBSD 3.0
1.5       mspo      810: 
1.4       mspo      811:     # Add support for PCI busses to the XEN3_DOMU kernel
                    812:     xpci* at xenbus ?
                    813:     pci* at xpci ?
1.5       mspo      814: 
1.4       mspo      815:     # Now add PCI and related devices to be used by this domain
                    816:     # USB Controller and Devices
1.5       mspo      817: 
1.4       mspo      818:     # PCI USB controllers
                    819:     uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
1.5       mspo      820: 
1.4       mspo      821:     # USB bus support
                    822:     usb*    at uhci?
1.5       mspo      823: 
1.4       mspo      824:     # USB Hubs
                    825:     uhub*   at usb?
                    826:     uhub*   at uhub? port ? configuration ? interface ?
1.5       mspo      827: 
1.4       mspo      828:     # USB Mass Storage
                    829:     umass*  at uhub? port ? configuration ? interface ?
                    830:     wd*     at umass?
                    831:     # SCSI controllers
                    832:     ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
1.5       mspo      833: 
1.4       mspo      834:     # SCSI bus support (for both ahc and umass)
                    835:     scsibus* at scsi?
1.5       mspo      836: 
1.4       mspo      837:     # SCSI devices
                    838:     sd*     at scsibus? target ? lun ?      # SCSI disk drives
                    839:     cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
1.1       mspo      840: 
                    841: Links and further information
1.5       mspo      842: =============================
1.1       mspo      843: 
1.9       gdt       844: -   The [HowTo on Installing into RAID-1](http://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
1.8       gdt       845:     explains how to set up booting a dom0 with Xen using grub 
                    846:     with NetBSD's RAIDframe.  (This is obsolete with the use of
                    847:     NetBSD's native boot.)
1.1       mspo      848: -   An example of how to use NetBSD's native bootloader to load
1.9       gdt       849:     NetBSD/Xen instead of Grub can be found in the i386/amd64 boot(8)
                    850:     and boot.cfg(5) manpages.

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