Annotation of wikisrc/ports/xen/howto.mdwn, revision 1.148

1.144     maxv        1: [[!meta title="Xen HowTo"]]
                      2: 
1.5       mspo        3: Introduction
1.13      gdt         4: ============
1.1       mspo        5: 
                      6: [![[Xen
1.140     khorben     7: screenshot]](https://www.netbsd.org/gallery/in-Action/hubertf-xens.png)](https://www.netbsd.org/gallery/in-Action/hubertf-xen.png)
1.1       mspo        8: 
1.58      gdt         9: Xen is a hypervisor (or virtual machine monitor) for x86 hardware
1.12      gdt        10: (i686-class or higher), which supports running multiple guest
1.58      gdt        11: operating systems on a single physical machine.  Xen is a Type 1 or
                     12: bare-metal hypervisor; one uses the Xen kernel to control the CPU,
                     13: memory and console, a dom0 operating system which mediates access to
                     14: other hardware (e.g., disks, network, USB), and one or more domU
                     15: operating systems which operate in an unprivileged virtualized
                     16: environment.  IO requests from the domU systems are forwarded by the
                     17: hypervisor (Xen) to the dom0 to be fulfilled.
1.12      gdt        18: 
                     19: Xen supports two styles of guests.  The original is Para-Virtualized
                     20: (PV) which means that the guest OS does not attempt to access hardware
                     21: directly, but instead makes hypercalls to the hypervisor.  This is
                     22: analogous to a user-space program making system calls.  (The dom0
                     23: operating system uses PV calls for some functions, such as updating
                     24: memory mapping page tables, but has direct hardware access for disk
                     25: and network.)   PV guests must be specifically coded for Xen.
                     26: 
                     27: The more recent style is HVM, which means that the guest does not have
                     28: code for Xen and need not be aware that it is running under Xen.
                     29: Attempts to access hardware registers are trapped and emulated.  This
                     30: style is less efficient but can run unmodified guests.
                     31: 
1.123     gdt        32: Generally any machine that runs NetBSD/amd64 will work with Xen and PV
                     33: guests.  In theory i386 computers (without x86_64/amd64 support) can
                     34: be used for Xen <= 4.2, but we have no recent reports of this working
                     35: (this is a hint).  For HVM guests, hardware support is needed, but it
                     36: is common on recent machines.  For Intel CPUs, one needs the VT-x
                     37: extension, shown in "cpuctl identify 0" as VMX.  For AMD CPUs, one
                     38: needs the AMD-V extensions, shown in "cpuctl identify 0" as SVM.
                     39: There are further features for IOMMU virtualization, Intel's VT-d and
                     40: AMD's AMD-Vi.  TODO: Explain whether Xen on NetBSD makes use of these
                     41: features.  TODO: Review by someone who really understands this.
1.19      gdt        42: 
1.123     gdt        43: Note that a FreeBSD dom0 requires VT-x and VT-d (or equivalent); this
                     44: is because the FreeBSD dom0 does not run in PV mode.
1.118     gdt        45: 
1.27      jnemeth    46: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
1.12      gdt        47: The dom0 can start one or more domUs.  (Booting is explained in detail
                     48: in the dom0 section.)
                     49: 
                     50: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
                     51: and that Xen kernels and tools are available in pkgsrc.  This HOWTO
                     52: attempts to address both the case of running a NetBSD dom0 on hardware
1.24      gdt        53: and running domUs under it (NetBSD and other), and also running NetBSD
                     54: as a domU in a VPS.
1.12      gdt        55: 
1.142     gdt        56: Xen 3.1 in pkgsrc used to support "PCI passthrough", which means that
1.20      gdt        57: specific PCI devices can be made available to a specific domU instead
                     58: of the dom0.  This can be useful to let a domU run X11, or access some
                     59: network interface or other peripheral.
                     60: 
1.111     wiz        61: NetBSD 6 and earlier supported Xen 2; support was removed from NetBSD
1.104     gdt        62: 7.  Xen 2 has been removed from pkgsrc.
1.54      gdt        63: 
1.12      gdt        64: Prerequisites
1.13      gdt        65: -------------
1.12      gdt        66: 
                     67: Installing NetBSD/Xen is not extremely difficult, but it is more
                     68: complex than a normal installation of NetBSD.
1.15      gdt        69: In general, this HOWTO is occasionally overly restrictive about how
                     70: things must be done, guiding the reader to stay on the established
                     71: path when there are no known good reasons to stray.
1.12      gdt        72: 
                     73: This HOWTO presumes a basic familiarity with the Xen system
1.123     gdt        74: architecture, with installing NetBSD on i386/amd64 hardware, and with
                     75: installing software from pkgsrc.  See also the [Xen
                     76: website](http://www.xenproject.org/).
1.1       mspo       77: 
1.15      gdt        78: Versions of Xen and NetBSD
                     79: ==========================
                     80: 
1.27      jnemeth    81: Most of the installation concepts and instructions are independent
                     82: of Xen version and NetBSD version.  This section gives advice on
                     83: which version to choose.  Versions not in pkgsrc and older unsupported
                     84: versions of NetBSD are intentionally ignored.
1.15      gdt        85: 
1.133     gdt        86: The term "amd64" is used to refer to both the NetBSD port and to the
                     87: hardware architecture on which it runs.  (Such hardware is made by
                     88: both Intel and AMD, and in 2016 a normal PC has this CPU
                     89: architecture.)
                     90: 
1.146     maxv       91: Xen versions
                     92: ------------
1.15      gdt        93: 
1.111     wiz        94: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
1.15      gdt        95: xenkernel and xentools.  We will refer only to the kernel versions,
                     96: but note that both packages must be installed together and must have
                     97: matching versions.
                     98: 
1.145     maxv       99: Versions available in pkgsrc:
1.85      gdt       100: 
1.145     maxv      101: [[!table data="""
                    102: Xen Version    |Package Name   |Xen CPU Support        |EOL'ed By Upstream
                    103: 4.2            |xenkernel42    |32bit, 64bit           |Yes
                    104: 4.5            |xenkernel45    |64bit                  |Yes
                    105: 4.6            |xenkernel46    |64bit                  |Partially
                    106: 4.8            |xenkernel48    |64bit                  |No
                    107: 4.11           |xenkernel411   |64bit                  |No
                    108: """]]
1.113     gdt       109: 
1.96      gdt       110: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
                    111: 
1.145     maxv      112: Note: Xen 4.2 was the last version to support 32bit CPUs.
                    113: 
1.19      gdt       114: Xen command program
                    115: -------------------
                    116: 
1.79      gdt       117: Early Xen used a program called xm to manipulate the system from the
1.19      gdt       118: dom0.  Starting in 4.1, a replacement program with similar behavior
1.79      gdt       119: called xl is provided, but it does not work well in 4.1.  In 4.2, both
1.127     gdt       120: xm and xl work fine.  4.4 is the last version that has xm.
                    121: 
                    122: You must make a global choice to use xm or xl, because it affects not
                    123: only which command you use, but the command used by rc.d scripts
                    124: (specifically xendomains) and which daemons should be run.  The
1.146     maxv      125: xentools packages provide xl for 4.2 and up.
1.127     gdt       126: 
                    127: In 4.2, you can choose to use xm by simply changing the ctl_command
1.135     gdt       128: variable and setting xend=YES in rc.conf.
1.127     gdt       129: 
                    130: With xl, virtual devices are configured in parallel, which can cause
                    131: problems if they are written assuming serial operation (e.g., updating
1.130     gdt       132: firewall rules without explicit locking).  There is now locking for
                    133: the provided scripts, which works for normal casses (e.g, file-backed
                    134: xbd, where a vnd must be allocated).  But, as of 201612, it has not
                    135: been adequately tested for a complex custom setup with a large number
                    136: of interfaces.
1.19      gdt       137: 
1.147     maxv      138: NetBSD versions
                    139: ---------------
1.15      gdt       140: 
1.142     gdt       141: The netbsd-7, netbsd-8, and -current branches are all reasonable
1.105     gdt       142: choices, with more or less the same considerations for non-Xen use.
1.147     maxv      143: NetBSD 8 is recommended as the stable version of the most recent
                    144: release for production use.
1.146     maxv      145: 
                    146: For developing Xen, netbsd-current may be appropriate.
1.15      gdt       147: 
                    148: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
                    149: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
1.105     gdt       150: need SMP for dom0 functions; the lack of support is really a problem
                    151: when using a dom0 as a normal computer.)
1.15      gdt       152: 
1.147     maxv      153: Note: NetBSD support is called XEN3. However, it does support Xen 4,
                    154: because the hypercall interface has remained identical.
                    155: 
1.18      gdt       156: Architecture
                    157: ------------
                    158: 
1.133     gdt       159: Xen itself can run on i386 (Xen < 4.2) or amd64 hardware (all Xen
1.105     gdt       160: versions).  (Practically, almost any computer where one would want to
                    161: run Xen today supports amd64.)
1.99      gdt       162: 
1.133     gdt       163: Xen, the dom0 system, and each domU system can be either i386 or
                    164: amd64.  When building a xenkernel package, one obtains an i386 Xen
1.134     wiz       165: kernel on an i386 host, and an amd64 Xen kernel on an amd64 host.  If
1.133     gdt       166: the Xen kernel is i386, then the dom0 kernel and all domU kernels must
                    167: be i386.  With an amd64 Xen kernel, an amd64 dom0 kernel is known to
                    168: work, and an i386 dom0 kernel should in theory work.  An amd64
                    169: Xen/dom0 is known to support both i386 and amd64 domUs.
                    170: 
                    171: i386 dom0 and domU kernels must be PAE (except for an i386 Xen 3.1
1.135     gdt       172: kernel, where one can use non-PAE for dom0 and all domUs); PAE kernels
                    173: are included in the NetBSD default build.  (Note that emacs (at least)
                    174: fails if run on i386 with PAE when built without, and vice versa,
                    175: presumably due to bugs in the undump code.)
                    176: 
                    177: Because of the above, the standard approach is to use an amd64 Xen
                    178: kernel and NetBSD/amd64 for the dom0.  For domUs, NetBSD/i386 (with
                    179: the PAE kernel) and NetBSD/amd64 are in widespread use, and there is
                    180: little to no Xen-specific reason to prefer one over the other.
1.133     gdt       181: 
                    182: Note that to use an i386 dom0 with Xen 4.5 or higher, one must build
1.135     gdt       183: (or obtain from pre-built packages) an amd64 Xen kernel and install
                    184: that on the system.  (One must also use a PAE i386 kernel, but this is
                    185: also required with an i386 Xen kernel.).  Almost no one in the
                    186: NetBSD/Xen community does this, and the standard, well-tested,
                    187: approach is to use an amd64 dom0.
                    188: 
                    189: A [posting on
                    190: xen-devel](https://lists.xen.org/archives/html/xen-devel/2012-07/msg00085.html)
                    191: explained that PV system call overhead was higher on amd64, and thus
                    192: there is some notion that i386 guests are faster.  It goes on to
                    193: caution that the total situation is complex and not entirely
                    194: understood. On top of that caution, the post is about Linux, not
                    195: NetBSD.  TODO: Include link to benchmarks, if someone posts them.
1.29      gdt       196: 
1.15      gdt       197: Recommendation
                    198: --------------
                    199: 
1.142     gdt       200: Therefore, this HOWTO recommends running xenkernel46, xl, the NetBSD 7
                    201: stable branch, and therefore to use an amd64 kernel as the dom0.
                    202: Either the i386PAE or amd64 version of NetBSD may be used as domUs.
                    203: 
                    204: A tentative replacement recommendation is xenkernel48, xl, and NetBSD
                    205: 8.
1.15      gdt       206: 
1.136     gdt       207: Because bugs are fixed quite often, and because of Xen security
                    208: advisories, it is good to stay up to date with NetBSD (tracking a
                    209: stable branch), with the Xen kernel (tracking a Xen version via
                    210: pkgsrc), and with the Xen tools.  Specifically, NetBSD (-7 and
                    211: -current) got an important fix affecting dom0/domU timesharing in
                    212: November, 2015, and xentools46 got a fix to enable Ubuntu guests to
                    213: boot in December, 2016.
                    214: 
1.15      gdt       215: NetBSD as a dom0
                    216: ================
                    217: 
                    218: NetBSD can be used as a dom0 and works very well.  The following
                    219: sections address installation, updating NetBSD, and updating Xen.
1.19      gdt       220: Note that it doesn't make sense to talk about installing a dom0 OS
                    221: without also installing Xen itself.  We first address installing
                    222: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
                    223: NetBSD install to a dom0 install by just changing the kernel and boot
                    224: configuration.
1.15      gdt       225: 
1.45      gdt       226: For experimenting with Xen, a machine with as little as 1G of RAM and
                    227: 100G of disk can work.  For running many domUs in productions, far
1.135     gdt       228: more will be needed; e.g. 4-8G and 1T of disk is reasonable for a
                    229: half-dozen domUs of 512M and 32G each.  Basically, the RAM and disk
                    230: have to be bigger than the sum of the RAM/disk needs of the dom0 and
                    231: all the domUs.
1.45      gdt       232: 
1.142     gdt       233: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
                    234: 512M it worked reliably.  This does not make sense, but if you see
                    235: "not ELF" after Xen boots, try increasing dom0 RAM.
                    236: 
1.15      gdt       237: Styles of dom0 operation
                    238: ------------------------
                    239: 
                    240: There are two basic ways to use Xen.  The traditional method is for
                    241: the dom0 to do absolutely nothing other than providing support to some
                    242: number of domUs.  Such a system was probably installed for the sole
                    243: purpose of hosting domUs, and sits in a server room on a UPS.
                    244: 
                    245: The other way is to put Xen under a normal-usage computer, so that the
                    246: dom0 is what the computer would have been without Xen, perhaps a
                    247: desktop or laptop.  Then, one can run domUs at will.  Purists will
                    248: deride this as less secure than the previous approach, and for a
                    249: computer whose purpose is to run domUs, they are right.  But Xen and a
1.93      gdt       250: dom0 (without domUs) is not meaningfully less secure than the same
1.15      gdt       251: things running without Xen.  One can boot Xen or boot regular NetBSD
                    252: alternately with little problems, simply refraining from starting the
                    253: Xen daemons when not running Xen.
                    254: 
                    255: Note that NetBSD as dom0 does not support multiple CPUs.  This will
1.51      gdt       256: limit the performance of the Xen/dom0 workstation approach.  In theory
                    257: the only issue is that the "backend drivers" are not yet MPSAFE:
1.140     khorben   258:   https://mail-index.netbsd.org/netbsd-users/2014/08/29/msg015195.html
1.15      gdt       259: 
1.19      gdt       260: Installation of NetBSD
                    261: ----------------------
1.13      gdt       262: 
1.19      gdt       263: First,
1.27      jnemeth   264: [install NetBSD/amd64](/guide/inst/)
1.19      gdt       265: just as you would if you were not using Xen.
                    266: However, the partitioning approach is very important.
                    267: 
                    268: If you want to use RAIDframe for the dom0, there are no special issues
                    269: for Xen.  Typically one provides RAID storage for the dom0, and the
1.22      gdt       270: domU systems are unaware of RAID.  The 2nd-stage loader bootxx_* skips
1.111     wiz       271: over a RAID1 header to find /boot from a file system within a RAID
1.22      gdt       272: partition; this is no different when booting Xen.
1.19      gdt       273: 
                    274: There are 4 styles of providing backing storage for the virtual disks
1.93      gdt       275: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN.
1.19      gdt       276: 
                    277: With raw partitions, one has a disklabel (or gpt) partition sized for
                    278: each virtual disk to be used by the domU.  (If you are able to predict
                    279: how domU usage will evolve, please add an explanation to the HOWTO.
                    280: Seriously, needs tend to change over time.)
                    281: 
1.27      jnemeth   282: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
                    283: for domU disks.  This is almost as efficient as raw disk partitions
                    284: and more flexible.  Hence raw disk partitions should typically not
                    285: be used.
1.19      gdt       286: 
1.111     wiz       287: One can use files in the dom0 file system, typically created by dd'ing
1.19      gdt       288: /dev/zero to create a specific size.  This is somewhat less efficient,
                    289: but very convenient, as one can cp the files for backup, or move them
                    290: between dom0 hosts.
                    291: 
                    292: Finally, in theory one can place the files backing the domU disks in a
                    293: SAN.  (This is an invitation for someone who has done this to add a
                    294: HOWTO page.)
1.1       mspo      295: 
1.19      gdt       296: Installation of Xen
                    297: -------------------
1.1       mspo      298: 
1.20      gdt       299: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
1.127     gdt       300: pkgsrc (or another matching pair).  See [the pkgsrc
1.140     khorben   301: documentation](https://www.NetBSD.org/docs/pkgsrc/) for help with
1.127     gdt       302: pkgsrc.  Ensure that your packages are recent; the HOWTO does not
                    303: contemplate old builds.
                    304: 
1.20      gdt       305: 
                    306: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm.  More
                    307: recent versions have HVM support integrated in the main xentools
                    308: package.  It is entirely reasonable to run only PV guests.
                    309: 
                    310: Next you need to install the selected Xen kernel itself, which is
                    311: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz".  Copy it to /.
                    312: For debugging, one may copy xen-debug.gz; this is conceptually similar
                    313: to DIAGNOSTIC and DEBUG in NetBSD.  xen-debug.gz is basically only
                    314: useful with a serial console.  Then, place a NetBSD XEN3_DOM0 kernel
                    315: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
1.75      gdt       316: of a NetBSD build.  If using i386, use
                    317: releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz.  (If using Xen
                    318: 3.1 and i386, you may use XEN3_DOM0 with the non-PAE Xen.  But you
                    319: should not use Xen 3.1.)  Both xen and the NetBSD kernel may be (and
                    320: typically are) left compressed.
                    321: 
1.135     gdt       322: In a dom0, kernfs is mandatory for xend to communicate with the
                    323: kernel, so ensure that /kern is in fstab.  (A standard NetBSD install
                    324: should already mount /kern.)
1.20      gdt       325: 
                    326: Because you already installed NetBSD, you have a working boot setup
                    327: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
1.135     gdt       328: beginning of your root file system, have /boot, and likely also
1.20      gdt       329: /boot.cfg.  (If not, fix before continuing!)
                    330: 
1.76      gdt       331: Add a line to to /boot.cfg to boot Xen.  See boot.cfg(5) for an
                    332: example.  The basic line is
1.20      gdt       333: 
1.142     gdt       334:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
1.20      gdt       335: 
1.142     gdt       336: which specifies that the dom0 should have 512M, leaving the rest to be
1.77      gdt       337: allocated for domUs.  To use a serial console, use
                    338: 
1.142     gdt       339:         menu=Xen:load /netbsd-XEN3_DOM0.gz;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
1.77      gdt       340: 
                    341: which will use the first serial port for Xen (which counts starting
1.139     gson      342: from 1, unlike NetBSD which counts starting from 0), forcing
                    343: speed/parity.  Because the NetBSD command line lacks a
                    344: "console=pc" argument, it will use the default "xencons" console device,
                    345: which directs the console I/O through Xen to the same console device Xen
                    346: itself uses (in this case, the serial port).
                    347: 
                    348: In an attempt to add performance, one can also add
1.37      gdt       349: 
                    350:         dom0_max_vcpus=1 dom0_vcpus_pin
                    351: 
                    352: to force only one vcpu to be provided (since NetBSD dom0 can't use
1.111     wiz       353: more) and to pin that vcpu to a physical CPU.  TODO: benchmark this.
1.20      gdt       354: 
1.93      gdt       355: Xen has [many boot
                    356: options](http://xenbits.xenproject.org/docs/4.5-testing/misc/xen-command-line.html),
1.111     wiz       357: and other than dom0 memory and max_vcpus, they are generally not
1.93      gdt       358: necessary.
                    359: 
1.20      gdt       360: As with non-Xen systems, you should have a line to boot /netbsd (a
1.127     gdt       361: kernel that works without Xen).  Consider a line to boot /netbsd.ok (a
                    362: fallback version of the non-Xen kernel, updated manually when you are
                    363: sure /netbsd is ok).  Consider also a line to boot fallback versions
                    364: of Xen and the dom0 kernel, but note that non-Xen NetBSD can be used
                    365: to resolve Xen booting issues.
                    366: 
                    367: Probably you want a default=N line to choose Xen in the absence of
                    368: intervention.
1.1       mspo      369: 
1.76      gdt       370: Now, reboot so that you are running a DOM0 kernel under Xen, rather
                    371: than GENERIC without Xen.
                    372: 
1.54      gdt       373: Using grub (historic)
                    374: ---------------------
                    375: 
                    376: Before NetBSD's native bootloader could support Xen, the use of
                    377: grub was recommended.  If necessary, see the
1.135     gdt       378: [old grub information](/ports/xen/howto-grub).
1.54      gdt       379: 
1.28      gdt       380: The [HowTo on Installing into
1.140     khorben   381: RAID-1](https://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
1.28      gdt       382: explains how to set up booting a dom0 with Xen using grub with
                    383: NetBSD's RAIDframe.  (This is obsolete with the use of NetBSD's native
1.135     gdt       384: boot.  Now, just create a system with RAID-1, and alter /boot.cfg as
                    385: described above.)
1.28      gdt       386: 
1.21      gdt       387: Configuring Xen
                    388: ---------------
                    389: 
1.53      gdt       390: Xen logs will be in /var/log/xen.
                    391: 
1.76      gdt       392: Now, you have a system that will boot Xen and the dom0 kernel, but not
                    393: do anything else special.  Make sure that you have rebooted into Xen.
                    394: There will be no domUs, and none can be started because you still have
1.102     gdt       395: to configure the dom0 daemons.
1.21      gdt       396: 
1.102     gdt       397: The daemons which should be run vary with Xen version and with whether
1.127     gdt       398: one is using xm or xl.  The Xen 3.1, 3.3 and 4.1 packages use xm.  Xen
                    399: 4.2 and up packages use xl.  To use xm with 4.2, edit xendomains to
                    400: use xm instead.
                    401: 
1.132     gdt       402: For 3.1 and 3.3, you should enable xend and xenbackendd:
1.31      gdt       403: 
1.32      gdt       404:         xend=YES
                    405:         xenbackendd=YES
1.31      gdt       406: 
1.132     gdt       407: For 4.1 and up, you should enable xencommons.  Not enabling xencommons
                    408: will result in a hang; it is necessary to hit ^C on the console to let
                    409: the machine finish booting.  If you are using xm (default in 4.1, or
                    410: if you changed xendomains in 4.2), you should also enable xend:
1.31      gdt       411: 
1.132     gdt       412:         xend=YES # only if using xm, and only installed <= 4.2
1.53      gdt       413:         xencommons=YES
1.31      gdt       414: 
                    415: TODO: Recommend for/against xen-watchdog.
1.27      jnemeth   416: 
1.53      gdt       417: After you have configured the daemons and either started them (in the
1.79      gdt       418: order given) or rebooted, use xm or xl to inspect Xen's boot messages,
1.102     gdt       419: available resources, and running domains.  An example with xl follows:
1.34      gdt       420: 
1.102     gdt       421:         # xl dmesg
1.43      gdt       422:        [xen's boot info]
1.102     gdt       423:         # xl info
1.43      gdt       424:        [available memory, etc.]
1.102     gdt       425:         # xl list
1.43      gdt       426:         Name              Id  Mem(MB)  CPU  State  Time(s)  Console
                    427:         Domain-0           0       64    0  r----     58.1
1.33      gdt       428: 
1.88      gdt       429: ### Issues with xencommons
                    430: 
                    431: xencommons starts xenstored, which stores data on behalf of dom0 and
                    432: domUs.  It does not currently work to stop and start xenstored.
                    433: Certainly all domUs should be shutdown first, following the sort order
                    434: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
                    435: and is not notified when xenstored exits, leading to not recreating
                    436: the state when the new xenstored starts.  Until there's a mechanism to
                    437: make this work, one should not expect to be able to restart xenstored
                    438: (and thus xencommons).  There is currently no reason to expect that
                    439: this will get fixed any time soon.
                    440: 
1.127     gdt       441: ### No-longer needed advice about devices
                    442: 
                    443: The installation of NetBSD should already have created devices for xen
                    444: (xencons, xenevt, xsd_kva), but if they are not present, create them:
                    445: 
                    446:         cd /dev && sh MAKEDEV xen
                    447: 
1.41      gdt       448: anita (for testing NetBSD)
                    449: --------------------------
                    450: 
1.82      gdt       451: With the setup so far (assuming 4.2/xl), one should be able to run
                    452: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
                    453: root, because anita must create a domU):
                    454: 
                    455:         anita --vmm=xl test file:///usr/obj/i386/
                    456: 
                    457: Alternatively, one can use --vmm=xm to use xm-based domU creation
                    458: instead (and must, on Xen <= 4.1).   TODO: confirm that anita xl really works.
                    459:     
1.40      gdt       460: Xen-specific NetBSD issues
                    461: --------------------------
                    462: 
                    463: There are (at least) two additional things different about NetBSD as a
                    464: dom0 kernel compared to hardware.
                    465: 
1.111     wiz       466: One is that the module ABI is different because some of the #defines
1.109     gdt       467: change, so one must build modules for Xen.  As of netbsd-7, the build
                    468: system does this automatically.  TODO: check this.  (Before building
                    469: Xen modules was added, it was awkward to use modules to the point
                    470: where it was considered that it did not work.)
1.40      gdt       471: 
                    472: The other difference is that XEN3_DOM0 does not have exactly the same
                    473: options as GENERIC.  While it is debatable whether or not this is a
                    474: bug, users should be aware of this and can simply add missing config
                    475: items if desired.
                    476: 
1.15      gdt       477: Updating NetBSD in a dom0
                    478: -------------------------
                    479: 
                    480: This is just like updating NetBSD on bare hardware, assuming the new
                    481: version supports the version of Xen you are running.  Generally, one
                    482: replaces the kernel and reboots, and then overlays userland binaries
                    483: and adjusts /etc.
                    484: 
                    485: Note that one must update both the non-Xen kernel typically used for
                    486: rescue purposes and the DOM0 kernel used with Xen.
                    487: 
1.55      gdt       488: Converting from grub to /boot
                    489: -----------------------------
                    490: 
                    491: These instructions were [TODO: will be] used to convert a system from
                    492: grub to /boot.  The system was originally installed in February of
                    493: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
                    494: over time.  Before these commands, it was running NetBSD 6 i386, Xen
                    495: 4.1 and grub, much like the message linked earlier in the grub
                    496: section.
                    497: 
1.111     wiz       498:         # Install MBR bootblocks on both disks. 
1.55      gdt       499:         fdisk -i /dev/rwd0d
                    500:         fdisk -i /dev/rwd1d
                    501:         # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
                    502:         installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
                    503:         installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
                    504:         # Install secondary boot loader
                    505:         cp -p /usr/mdec/boot /
1.111     wiz       506:         # Create boot.cfg following earlier guidance:
1.142     gdt       507:         menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
                    508:         menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=512M
1.55      gdt       509:         menu=GENERIC:boot
                    510:         menu=GENERIC single-user:boot -s
                    511:         menu=GENERIC.ok:boot netbsd.ok
                    512:         menu=GENERIC.ok single-user:boot netbsd.ok -s
                    513:         menu=Drop to boot prompt:prompt
                    514:         default=1
                    515:         timeout=30
                    516: 
                    517: TODO: actually do this and fix it if necessary.
1.22      gdt       518: 
1.102     gdt       519: Upgrading Xen versions
1.15      gdt       520: ---------------------
                    521: 
1.110     gdt       522: Minor version upgrades are trivial.  Just rebuild/replace the
                    523: xenkernel version and copy the new xen.gz to / (where /boot.cfg
                    524: references it), and reboot.
                    525: 
                    526: Major version upgrades are conceptually not difficult, but can run
                    527: into all the issues found when installing Xen.  Assuming migration
                    528: from 4.1 to 4.2, remove the xenkernel41 and xentools41 packages and
                    529: install the xenkernel42 and xentools42 packages.  Copy the 4.2 xen.gz
                    530: to /.
1.21      gdt       531: 
1.102     gdt       532: Ensure that the contents of /etc/rc.d/xen* are correct.  Specifically,
                    533: they must match the package you just installed and not be left over
                    534: from some previous installation.
                    535: 
                    536: Enable the correct set of daemons; see the configuring section above.
                    537: (Upgrading from 3.x to 4.x without doing this will result in a hang.)
                    538: 
                    539: Ensure that the domU config files are valid for the new version.
1.110     gdt       540: Specifically, for 4.x remove autorestart=True, and ensure that disks
                    541: are specified with numbers as the second argument, as the examples
                    542: above show, and not NetBSD device names.
1.15      gdt       543: 
1.97      gdt       544: Hardware known to work
                    545: ----------------------
                    546: 
                    547: Arguably, this section is misplaced, and there should be a page of
                    548: hardware that runs NetBSD/amd64 well, with the mostly-well-founded
                    549: assumption that NetBSD/xen runs fine on any modern hardware that
1.129     gdt       550: NetBSD/amd64 runs well on.  Until then, we give motherboard/CPU (and
                    551: sometimes RAM) pairs/triples to aid those choosing a motherboard.
                    552: Note that Xen systems usually do not run X, so a listing here does not
                    553: imply that X works at all.
1.97      gdt       554: 
                    555:         Supermicro X9SRL-F, Xeon E5-1650 v2, 96 GiB ECC
                    556:         Supermicro ??, Atom C2758 (8 core), 32 GiB ECC
                    557:         ASUS M5A78L-M/USB3 AM3+ microATX, AMD Piledriver X8 4000MHz, 16 GiB ECC
                    558: 
                    559: Older hardware:
                    560: 
1.98      gdt       561:         Intel D915GEV, Pentium4 CPU 3.40GHz, 4GB 533MHz Synchronous DDR2
1.129     gdt       562:         INTEL DG33FB, "Intel(R) Core(TM)2 Duo CPU     E6850  @ 3.00GHz"
                    563:         INTEL DG33FB, "Intel(R) Core(TM)2 Duo CPU     E8400  @ 3.00GHz"
1.28      gdt       564: 
1.82      gdt       565: Running Xen under qemu
                    566: ----------------------
                    567: 
                    568: The astute reader will note that this section is somewhat twisted.
                    569: However, it can be useful to run Xen under qemu either because the
                    570: version of NetBSD as a dom0 does not run on the hardware in use, or to
                    571: generate automated test cases involving Xen.
                    572: 
1.84      gdt       573: In 2015-01, the following combination was reported to mostly work:
1.82      gdt       574: 
                    575:         host OS: NetBSD/amd64 6.1.4
                    576:         qemu: 2.2.0 from pkgsrc
                    577:         Xen kernel: xenkernel42-4.2.5nb1 from pkgsrc
                    578:         dom0 kernel: NetBSD/amd64 6.1.5
                    579:         Xen tools: xentools42-4.2.5 from pkgsrc
                    580: 
1.140     khorben   581: See [PR 47720](https://gnats.netbsd.org/47720) for a problem with dom0
1.91      gdt       582: shutdown.
1.84      gdt       583: 
1.28      gdt       584: Unprivileged domains (domU)
                    585: ===========================
                    586: 
                    587: This section describes general concepts about domUs.  It does not
1.33      gdt       588: address specific domU operating systems or how to install them.  The
                    589: config files for domUs are typically in /usr/pkg/etc/xen, and are
1.60      wiki      590: typically named so that the file name, domU name and the domU's host
1.33      gdt       591: name match.
                    592: 
1.111     wiz       593: The domU is provided with CPU and memory by Xen, configured by the
1.33      gdt       594: dom0.  The domU is provided with disk and network by the dom0,
                    595: mediated by Xen, and configured in the dom0.
                    596: 
                    597: Entropy in domUs can be an issue; physical disks and network are on
                    598: the dom0.  NetBSD's /dev/random system works, but is often challenged.
                    599: 
1.48      gdt       600: Config files
                    601: ------------
                    602: 
                    603: There is no good order to present config files and the concepts
                    604: surrounding what is being configured.  We first show an example config
                    605: file, and then in the various sections give details.
                    606: 
                    607: See (at least in xentools41) /usr/pkg/share/examples/xen/xmexample*,
                    608: for a large number of well-commented examples, mostly for running
                    609: GNU/Linux.
                    610: 
                    611: The following is an example minimal domain configuration file
                    612: "/usr/pkg/etc/xen/foo".  It is (with only a name change) an actual
                    613: known working config file on Xen 4.1 (NetBSD 5 amd64 dom0 and NetBSD 5
                    614: i386 domU).  The domU serves as a network file server.
                    615: 
                    616:         # -*- mode: python; -*-
                    617: 
                    618:         kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
                    619:         memory = 1024
                    620:         vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
                    621:         disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
                    622:                  'file:/n0/xen/foo-wd1,0x1,w' ]
                    623: 
                    624: The domain will have the same name as the file.  The kernel has the
                    625: host/domU name in it, so that on the dom0 one can update the various
                    626: domUs independently.  The vif line causes an interface to be provided,
                    627: with a specific mac address (do not reuse MAC addresses!), in bridge
                    628: mode.  Two disks are provided, and they are both writable; the bits
                    629: are stored in files and Xen attaches them to a vnd(4) device in the
1.111     wiz       630: dom0 on domain creation.  The system treats xbd0 as the boot device
1.48      gdt       631: without needing explicit configuration.
                    632: 
                    633: By default xm looks for domain config files in /usr/pkg/etc/xen.  Note
                    634: that "xm create" takes the name of a config file, while other commands
                    635: take the name of a domain.  To create the domain, connect to the
                    636: console, create the domain while attaching the console, shutdown the
                    637: domain, and see if it has finished stopping, do (or xl with Xen >=
                    638: 4.2):
                    639: 
                    640:         xm create foo
                    641:         xm console foo
                    642:         xm create -c foo
                    643:         xm shutdown foo
1.90      gdt       644:         xm list
1.48      gdt       645: 
                    646: Typing ^] will exit the console session.  Shutting down a domain is
                    647: equivalent to pushing the power button; a NetBSD domU will receive a
                    648: power-press event and do a clean shutdown.  Shutting down the dom0
                    649: will trigger controlled shutdowns of all configured domUs.
                    650: 
                    651: domU kernels
                    652: ------------
                    653: 
                    654: On a physical computer, the BIOS reads sector 0, and a chain of boot
                    655: loaders finds and loads a kernel.  Normally this comes from the root
1.111     wiz       656: file system.  With Xen domUs, the process is totally different.  The
1.48      gdt       657: normal path is for the domU kernel to be a file in the dom0's
1.111     wiz       658: file system.  At the request of the dom0, Xen loads that kernel into a
1.48      gdt       659: new domU instance and starts execution.  While domU kernels can be
                    660: anyplace, reasonable places to store domU kernels on the dom0 are in /
                    661: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
                    662: config files), or in /u0/xen (where the vdisks are).
                    663: 
1.59      gdt       664: Note that loading the domU kernel from the dom0 implies that boot
                    665: blocks, /boot, /boot.cfg, and so on are all ignored in the domU.
1.48      gdt       666: See the VPS section near the end for discussion of alternate ways to
                    667: obtain domU kernels.
                    668: 
1.33      gdt       669: CPU and memory
                    670: --------------
                    671: 
1.48      gdt       672: A domain is provided with some number of vcpus, less than the number
1.111     wiz       673: of CPUs seen by the hypervisor.  (For a dom0, this is controlled by
1.48      gdt       674: the boot argument "dom0_max_vcpus=1".)  For a domU, it is controlled
                    675: from the config file by the "vcpus = N" directive.
                    676: 
                    677: A domain is provided with memory; this is controlled in the config
                    678: file by "memory = N" (in megabytes).  In the straightforward case, the
                    679: sum of the the memory allocated to the dom0 and all domUs must be less
1.33      gdt       680: than the available memory.
                    681: 
                    682: Xen also provides a "balloon" driver, which can be used to let domains
                    683: use more memory temporarily.  TODO: Explain better, and explain how
                    684: well it works with NetBSD.
1.28      gdt       685: 
                    686: Virtual disks
                    687: -------------
                    688: 
1.33      gdt       689: With the file/vnd style, typically one creates a directory,
                    690: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
                    691: domUs.  Then, for each domU disk, one writes zeros to a file that then
                    692: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
                    693: for the first virtual disk for the domU called foo.  Writing zeros to
                    694: the file serves two purposes.  One is that preallocating the contents
                    695: improves performance.  The other is that vnd on sparse files has
                    696: failed to work.  TODO: give working/notworking NetBSD versions for
1.127     gdt       697: sparse vnd and gnats reference.  Note that the use of file/vnd for Xen
                    698: is not really different than creating a file-backed virtual disk for
                    699: some other purpose, except that xentools handles the vnconfig
                    700: commands.  To create an empty 4G virtual disk, simply do
1.39      gdt       701: 
                    702:         dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
1.33      gdt       703: 
1.89      gdt       704: Do not use qemu-img-xen, because this will create sparse file.  There
                    705: have been recent (2015) reports of sparse vnd(4) devices causing
                    706: lockups, but there is apparently no PR.
                    707: 
1.33      gdt       708: With the lvm style, one creates logical devices.  They are then used
1.48      gdt       709: similarly to vnds.  TODO: Add an example with lvm.
                    710: 
                    711: In domU config files, the disks are defined as a sequence of 3-tuples.
                    712: The first element is "method:/path/to/disk".  Common methods are
                    713: "file:" for file-backed vnd. and "phy:" for something that is already
                    714: a (TODO: character or block) device.
                    715: 
                    716: The second element is an artifact of how virtual disks are passed to
                    717: Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
                    718: are given a device name to associate with the disk, and values like
                    719: "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
                    720: as xbd0, the second as xbd1, and so on.  However, xm/xl demand a
                    721: second argument.  The name given is converted to a major/minor by
1.49      gdt       722: calling stat(2) on the name in /dev and this is passed to the domU.
                    723: In the general case, the dom0 and domU can be different operating
1.48      gdt       724: systems, and it is an unwarranted assumption that they have consistent
                    725: numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
                    726: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
1.49      gdt       727: for the second works fine and avoids this issue.  For a GNU/Linux
                    728: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
                    729: /dev/hda1.
1.48      gdt       730: 
                    731: The third element is "w" for writable disks, and "r" for read-only
                    732: disks.
1.28      gdt       733: 
1.127     gdt       734: Note that NetBSD by default creates only vnd[0123].  If you need more
                    735: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
                    736: dom0.
                    737: 
                    738: Note that NetBSD by default creates only xbd[0123].  If you need more
                    739: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
                    740: 
1.28      gdt       741: Virtual Networking
                    742: ------------------
                    743: 
1.111     wiz       744: Xen provides virtual Ethernets, each of which connects the dom0 and a
1.46      gdt       745: domU.  For each virtual network, there is an interface "xvifN.M" in
                    746: the dom0, and in domU index N, a matching interface xennetM (NetBSD
                    747: name).  The interfaces behave as if there is an Ethernet with two
1.111     wiz       748: adapters connected.  From this primitive, one can construct various
1.46      gdt       749: configurations.  We focus on two common and useful cases for which
                    750: there are existing scripts: bridging and NAT.
1.28      gdt       751: 
1.48      gdt       752: With bridging (in the example above), the domU perceives itself to be
                    753: on the same network as the dom0.  For server virtualization, this is
                    754: usually best.  Bridging is accomplished by creating a bridge(4) device
                    755: and adding the dom0's physical interface and the various xvifN.0
                    756: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
                    757: config file.  The bridge must be set up already in the dom0; an
                    758: example /etc/ifconfig.bridge0 is:
1.46      gdt       759: 
                    760:         create
                    761:         up
                    762:         !brconfig bridge0 add wm0
1.28      gdt       763: 
                    764: With NAT, the domU perceives itself to be behind a NAT running on the
                    765: dom0.  This is often appropriate when running Xen on a workstation.
1.48      gdt       766: TODO: NAT appears to be configured by "vif = [ '' ]".
1.28      gdt       767: 
1.49      gdt       768: The MAC address specified is the one used for the interface in the new
1.53      gdt       769: domain.  The interface in dom0 will use this address XOR'd with
1.49      gdt       770: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
                    771: 
1.33      gdt       772: Sizing domains
                    773: --------------
                    774: 
                    775: Modern x86 hardware has vast amounts of resources.  However, many
                    776: virtual servers can function just fine on far less.  A system with
1.142     gdt       777: 512M of RAM and a 4G disk can be a reasonable choice.  Note that it is
1.33      gdt       778: far easier to adjust virtual resources than physical ones.  For
                    779: memory, it's just a config file edit and a reboot.  For disk, one can
                    780: create a new file and vnconfig it (or lvm), and then dump/restore,
                    781: just like updating physical disks, but without having to be there and
                    782: without those pesky connectors.
                    783: 
1.48      gdt       784: Starting domains automatically
                    785: ------------------------------
1.28      gdt       786: 
1.48      gdt       787: To start domains foo at bar at boot and shut them down cleanly on dom0
                    788: shutdown, in rc.conf add:
1.28      gdt       789: 
1.48      gdt       790:         xendomains="foo bar"
1.28      gdt       791: 
1.111     wiz       792: Note that earlier versions of the xentools41 xendomains rc.d script
                    793: used xl, when one should use xm with 4.1.
1.28      gdt       794: 
                    795: Creating specific unprivileged domains (domU)
                    796: =============================================
1.14      gdt       797: 
                    798: Creating domUs is almost entirely independent of operating system.  We
1.49      gdt       799: have already presented the basics of config files.  Note that you must
                    800: have already completed the dom0 setup so that "xl list" (or "xm list")
                    801: works.
1.14      gdt       802: 
                    803: Creating an unprivileged NetBSD domain (domU)
                    804: ---------------------------------------------
1.1       mspo      805: 
1.49      gdt       806: See the earlier config file, and adjust memory.  Decide on how much
                    807: storage you will provide, and prepare it (file or lvm).
                    808: 
1.111     wiz       809: While the kernel will be obtained from the dom0 file system, the same
1.49      gdt       810: file should be present in the domU as /netbsd so that tools like
                    811: savecore(8) can work.   (This is helpful but not necessary.)
                    812: 
                    813: The kernel must be specifically for Xen and for use as a domU.  The
                    814: i386 and amd64 provide the following kernels:
                    815: 
                    816:         i386 XEN3_DOMU
                    817:         i386 XEN3PAE_DOMU
1.95      gdt       818:         amd64 XEN3_DOMU
1.5       mspo      819: 
1.49      gdt       820: Unless using Xen 3.1 (and you shouldn't) with i386-mode Xen, you must
                    821: use the PAE version of the i386 kernel.
                    822: 
                    823: This will boot NetBSD, but this is not that useful if the disk is
                    824: empty.  One approach is to unpack sets onto the disk outside of xen
                    825: (by mounting it, just as you would prepare a physical disk for a
                    826: system you can't run the installer on).
                    827: 
                    828: A second approach is to run an INSTALL kernel, which has a miniroot
                    829: and can load sets from the network.  To do this, copy the INSTALL
                    830: kernel to / and change the kernel line in the config file to:
1.5       mspo      831: 
1.49      gdt       832:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
1.5       mspo      833: 
1.49      gdt       834: Then, start the domain as "xl create -c configname".
1.1       mspo      835: 
1.49      gdt       836: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
                    837: line should be used in the config file.
1.1       mspo      838: 
1.3       mspo      839:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
1.1       mspo      840: 
                    841: After booting the domain, the option to install via CDROM may be
1.49      gdt       842: selected.  The CDROM device should be changed to `xbd1d`.
1.1       mspo      843: 
1.49      gdt       844: Once done installing, "halt -p" the new domain (don't reboot or halt,
                    845: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
                    846: config file), switch the config file back to the XEN3_DOMU kernel,
                    847: and start the new domain again. Now it should be able to use "root on
                    848: xbd0a" and you should have a, functional NetBSD domU.
1.1       mspo      849: 
1.49      gdt       850: TODO: check if this is still accurate.
1.1       mspo      851: When the new domain is booting you'll see some warnings about *wscons*
                    852: and the pseudo-terminals. These can be fixed by editing the files
1.5       mspo      853: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
                    854: `/etc/ttys`, except *console*, like this:
1.1       mspo      855: 
1.3       mspo      856:     console "/usr/libexec/getty Pc"         vt100   on secure
                    857:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
                    858:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
                    859:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
                    860:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
1.1       mspo      861: 
1.5       mspo      862: Finally, all screens must be commented out from `/etc/wscons.conf`.
1.1       mspo      863: 
                    864: It is also desirable to add
                    865: 
1.49      gdt       866:         powerd=YES
1.1       mspo      867: 
1.5       mspo      868: in rc.conf. This way, the domain will be properly shut down if
1.53      gdt       869: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
1.1       mspo      870: 
1.92      gdt       871: It is not strictly necessary to have a kernel (as /netbsd) in the domU
1.111     wiz       872: file system.  However, various programs (e.g. netstat) will use that
1.92      gdt       873: kernel to look up symbols to read from kernel virtual memory.  If
                    874: /netbsd is not the running kernel, those lookups will fail.  (This is
                    875: not really a Xen-specific issue, but because the domU kernel is
                    876: obtained from the dom0, it is far more likely to be out of sync or
                    877: missing with Xen.)
                    878: 
1.14      gdt       879: Creating an unprivileged Linux domain (domU)
1.5       mspo      880: --------------------------------------------
1.1       mspo      881: 
                    882: Creating unprivileged Linux domains isn't much different from
                    883: unprivileged NetBSD domains, but there are some details to know.
                    884: 
                    885: First, the second parameter passed to the disk declaration (the '0x1' in
                    886: the example below)
                    887: 
1.3       mspo      888:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
1.1       mspo      889: 
                    890: does matter to Linux. It wants a Linux device number here (e.g. 0x300
1.49      gdt       891: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
                    892: So, hda1 which has major 3 and minor 1 on a Linux system will have
                    893: device number 0x301.  Alternatively, devices names can be used (hda,
                    894: hdb, ...)  as xentools has a table to map these names to devices
                    895: numbers.  To export a partition to a Linux guest we can use:
1.1       mspo      896: 
1.49      gdt       897:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
                    898:         root = "/dev/hda1 ro"
1.1       mspo      899: 
                    900: and it will appear as /dev/hda on the Linux system, and be used as root
                    901: partition.
                    902: 
1.49      gdt       903: To install the Linux system on the partition to be exported to the
                    904: guest domain, the following method can be used: install
                    905: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
                    906: that will be the root partition of your Linux domain, and mount it.
                    907: Then copy the files from a working Linux system, make adjustments in
                    908: `/etc` (fstab, network config).  It should also be possible to extract
                    909: binary packages such as .rpm or .deb directly to the mounted partition
                    910: using the appropriate tool, possibly running under NetBSD's Linux
1.111     wiz       911: emulation.  Once the file system has been populated, umount it.  If
                    912: desirable, the file system can be converted to ext3 using tune2fs -j.
1.49      gdt       913: It should now be possible to boot the Linux guest domain, using one of
                    914: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
1.1       mspo      915: 
1.111     wiz       916: To get the Linux console right, you need to add:
1.1       mspo      917: 
1.3       mspo      918:     extra = "xencons=tty1"
1.1       mspo      919: 
1.111     wiz       920: to your configuration since not all Linux distributions auto-attach a
1.1       mspo      921: tty to the xen console.
                    922: 
1.14      gdt       923: Creating an unprivileged Solaris domain (domU)
1.5       mspo      924: ----------------------------------------------
1.1       mspo      925: 
1.50      gdt       926: See possibly outdated
                    927: [Solaris domU instructions](/ports/xen/howto-solaris/).
1.5       mspo      928: 
1.1       mspo      929: 
1.52      gdt       930: PCI passthrough: Using PCI devices in guest domains
                    931: ---------------------------------------------------
1.1       mspo      932: 
1.53      gdt       933: The dom0 can give other domains access to selected PCI
1.52      gdt       934: devices. This can allow, for example, a non-privileged domain to have
                    935: access to a physical network interface or disk controller.  However,
                    936: keep in mind that giving a domain access to a PCI device most likely
                    937: will give the domain read/write access to the whole physical memory,
                    938: as PCs don't have an IOMMU to restrict memory access to DMA-capable
1.53      gdt       939: device.  Also, it's not possible to export ISA devices to non-dom0
1.52      gdt       940: domains, which means that the primary VGA adapter can't be exported.
                    941: A guest domain trying to access the VGA registers will panic.
                    942: 
1.53      gdt       943: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
1.52      gdt       944: not been ported to later versions at this time.
                    945: 
                    946: For a PCI device to be exported to a domU, is has to be attached to
                    947: the "pciback" driver in dom0.  Devices passed to the dom0 via the
                    948: pciback.hide boot parameter will attach to "pciback" instead of the
                    949: usual driver.  The list of devices is specified as "(bus:dev.func)",
1.5       mspo      950: where bus and dev are 2-digit hexadecimal numbers, and func a
                    951: single-digit number:
1.1       mspo      952: 
1.52      gdt       953:         pciback.hide=(00:0a.0)(00:06.0)
1.1       mspo      954: 
1.52      gdt       955: pciback devices should show up in the dom0's boot messages, and the
1.5       mspo      956: devices should be listed in the `/kern/xen/pci` directory.
1.1       mspo      957: 
1.52      gdt       958: PCI devices to be exported to a domU are listed in the "pci" array of
                    959: the domU's config file, with the format "0000:bus:dev.func".
1.1       mspo      960: 
1.52      gdt       961:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
1.1       mspo      962: 
1.52      gdt       963: In the domU an "xpci" device will show up, to which one or more pci
1.111     wiz       964: buses will attach.  Then the PCI drivers will attach to PCI buses as
1.52      gdt       965: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
                    966: or any PCI drivers built in by default; you have to build your own
                    967: kernel to use PCI devices in a domU.  Here's a kernel config example;
                    968: note that only the "xpci" lines are unusual.
                    969: 
                    970:         include         "arch/i386/conf/XEN3_DOMU"
                    971: 
1.111     wiz       972:         # Add support for PCI buses to the XEN3_DOMU kernel
1.52      gdt       973:         xpci* at xenbus ?
                    974:         pci* at xpci ?
                    975: 
                    976:         # PCI USB controllers
                    977:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
                    978: 
                    979:         # USB bus support
                    980:         usb*    at uhci?
                    981: 
                    982:         # USB Hubs
                    983:         uhub*   at usb?
                    984:         uhub*   at uhub? port ? configuration ? interface ?
                    985: 
                    986:         # USB Mass Storage
                    987:         umass*  at uhub? port ? configuration ? interface ?
                    988:         wd*     at umass?
                    989:         # SCSI controllers
                    990:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
                    991: 
                    992:         # SCSI bus support (for both ahc and umass)
                    993:         scsibus* at scsi?
                    994: 
                    995:         # SCSI devices
                    996:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
                    997:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
1.1       mspo      998: 
                    999: 
1.28      gdt      1000: NetBSD as a domU in a VPS
                   1001: =========================
                   1002: 
                   1003: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
                   1004: hardware.  This section explains how to deal with Xen in a domU as a
                   1005: virtual private server where you do not control or have access to the
1.70      gdt      1006: dom0.  This is not intended to be an exhaustive list of VPS providers;
                   1007: only a few are mentioned that specifically support NetBSD.
1.28      gdt      1008: 
1.52      gdt      1009: VPS operators provide varying degrees of access and mechanisms for
                   1010: configuration.  The big issue is usually how one controls which kernel
1.111     wiz      1011: is booted, because the kernel is nominally in the dom0 file system (to
                   1012: which VPS users do not normally have access).  A second issue is how
1.70      gdt      1013: to install NetBSD.
1.52      gdt      1014: A VPS user may want to compile a kernel for security updates, to run
                   1015: npf, run IPsec, or any other reason why someone would want to change
                   1016: their kernel.
                   1017: 
1.111     wiz      1018: One approach is to have an administrative interface to upload a kernel,
1.68      gdt      1019: or to select from a prepopulated list.  Other approaches are pygrub
1.59      gdt      1020: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
1.111     wiz      1021: kernel from the domU file system.  This is closer to a regular physical
1.59      gdt      1022: computer, where someone who controls a machine can replace the kernel.
1.52      gdt      1023: 
1.74      gdt      1024: A second issue is multiple CPUs.  With NetBSD 6, domUs support
                   1025: multiple vcpus, and it is typical for VPS providers to enable multiple
                   1026: CPUs for NetBSD domUs.
                   1027: 
1.68      gdt      1028: pygrub
1.59      gdt      1029: -------
1.52      gdt      1030: 
1.111     wiz      1031: pygrub runs in the dom0 and looks into the domU file system.  This
                   1032: implies that the domU must have a kernel in a file system in a format
1.68      gdt      1033: known to pygrub.  As of 2014, pygrub seems to be of mostly historical
                   1034: interest.
1.52      gdt      1035: 
1.59      gdt      1036: pvgrub
                   1037: ------
                   1038: 
                   1039: pvgrub is a version of grub that uses PV operations instead of BIOS
                   1040: calls.  It is booted from the dom0 as the domU kernel, and then reads
1.111     wiz      1041: /grub/menu.lst and loads a kernel from the domU file system.
1.59      gdt      1042: 
1.70      gdt      1043: [Panix](http://www.panix.com/) lets users use pvgrub.  Panix reports
1.71      gdt      1044: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
                   1045: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
1.70      gdt      1046: page](http://www.panix.com/v-colo/grub.html), which describes only
1.74      gdt      1047: Linux but should be updated to cover NetBSD :-).
1.70      gdt      1048: 
                   1049: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
                   1050: their own kernel.  See then [prgmr.com NetBSD
1.74      gdt      1051: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
                   1052: (which is in need of updating).
1.59      gdt      1053: 
1.70      gdt      1054: It appears that [grub's FFS
                   1055: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
                   1056: does not support all aspects of modern FFS, but there are also reports
1.72      gdt      1057: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
1.70      gdt      1058: partition for the kernel with the intent that grub can understand it,
                   1059: which leads to /netbsd not being the actual kernel.  One must remember
1.111     wiz      1060: to update the special boot partition.
1.59      gdt      1061: 
                   1062: Amazon
                   1063: ------
                   1064: 
1.143     wiki     1065: See the [Amazon EC2 page](/amazon_ec2/).
1.44      gdt      1066: 
1.65      gdt      1067: TODO items for improving NetBSD/xen
                   1068: ===================================
                   1069: 
1.93      gdt      1070: * Make the NetBSD dom0 kernel work with SMP.
                   1071: * Test the Xen 4.5 packages adequately to be able to recommend them as
                   1072:   the standard approach.
                   1073: * Get PCI passthrough working on Xen 4.5
1.65      gdt      1074: * Get pvgrub into pkgsrc, either via xentools or separately.
                   1075: * grub
1.70      gdt      1076:   * Check/add support to pkgsrc grub2 for UFS2 and arbitrary
1.66      gdt      1077:     fragsize/blocksize (UFS2 support may be present; the point is to
1.111     wiz      1078:     make it so that with any UFS1/UFS2 file system setup that works
1.66      gdt      1079:     with NetBSD grub will also work).
1.140     khorben  1080:     See [pkg/40258](https://gnats.netbsd.org/40258).
1.65      gdt      1081:   * Push patches upstream.
                   1082:   * Get UFS2 patches into pvgrub.
                   1083: * Add support for PV ops to a version of /boot, and make it usable as
                   1084:   a kernel in Xen, similar to pvgrub.
1.93      gdt      1085: 
                   1086: Random pointers
                   1087: ===============
                   1088: 
1.117     gdt      1089: This section contains links from elsewhere not yet integrated into the
                   1090: HOWTO, and other guides.
1.93      gdt      1091: 
                   1092: * http://www.lumbercartel.ca/library/xen/
                   1093: * http://pbraun.nethence.com/doc/sysutils/xen_netbsd_dom0.html
1.117     gdt      1094: * https://gmplib.org/~tege/xen.html

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