File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.102: download - view: text, annotated - select for diffs
Mon Dec 19 21:11:03 2016 UTC (4 years, 4 months ago) by gdt
Branches: MAIN
CVS tags: HEAD
Adjust version info

Add cautions about upgrade issues.

    1: Introduction
    2: ============
    4: [![[Xen
    5: screenshot]](](
    7: Xen is a hypervisor (or virtual machine monitor) for x86 hardware
    8: (i686-class or higher), which supports running multiple guest
    9: operating systems on a single physical machine.  Xen is a Type 1 or
   10: bare-metal hypervisor; one uses the Xen kernel to control the CPU,
   11: memory and console, a dom0 operating system which mediates access to
   12: other hardware (e.g., disks, network, USB), and one or more domU
   13: operating systems which operate in an unprivileged virtualized
   14: environment.  IO requests from the domU systems are forwarded by the
   15: hypervisor (Xen) to the dom0 to be fulfilled.
   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.
   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.
   30: Generally any amd64 machine will work with Xen and PV guests.  In
   31: theory i386 computers without amd64 support can be used for Xen <=
   32: 4.2, but we have no recent reports of this working (this is a hint).
   33: For HVM guests, the VT or VMX cpu feature (Intel) or SVM/HVM/VT
   34: (amd64) is needed; "cpuctl identify 0" will show this.  TODO: Clean up
   35: and check the above features.
   37: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
   38: The dom0 can start one or more domUs.  (Booting is explained in detail
   39: in the dom0 section.)
   41: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
   42: and that Xen kernels and tools are available in pkgsrc.  This HOWTO
   43: attempts to address both the case of running a NetBSD dom0 on hardware
   44: and running domUs under it (NetBSD and other), and also running NetBSD
   45: as a domU in a VPS.
   47: Some versions of Xen support "PCI passthrough", which means that
   48: specific PCI devices can be made available to a specific domU instead
   49: of the dom0.  This can be useful to let a domU run X11, or access some
   50: network interface or other peripheral.
   52: NetBSD used to support Xen2; this has been removed.
   54: Prerequisites
   55: -------------
   57: Installing NetBSD/Xen is not extremely difficult, but it is more
   58: complex than a normal installation of NetBSD.
   59: In general, this HOWTO is occasionally overly restrictive about how
   60: things must be done, guiding the reader to stay on the established
   61: path when there are no known good reasons to stray.
   63: This HOWTO presumes a basic familiarity with the Xen system
   64: architecture.  This HOWTO presumes familiarity with installing NetBSD
   65: on i386/amd64 hardware and installing software from pkgsrc.
   66: See also the [Xen website](
   68: Versions of Xen and NetBSD
   69: ==========================
   71: Most of the installation concepts and instructions are independent
   72: of Xen version and NetBSD version.  This section gives advice on
   73: which version to choose.  Versions not in pkgsrc and older unsupported
   74: versions of NetBSD are intentionally ignored.
   76: Xen
   77: ---
   79: In NetBSD, xen is provided in pkgsrc, via matching pairs of packages
   80: xenkernel and xentools.  We will refer only to the kernel versions,
   81: but note that both packages must be installed together and must have
   82: matching versions.
   84: xenkernel3 provides Xen 3.1.  This no longer receives security patches
   85: and should not be used.  It supports PCI passthrough, which is why
   86: people use it anyway. Xen 3.1 supports non-PAE on i386.
   88: xenkernel33 provides Xen 3.3.  This no longer receives security
   89: patches and should not be used.  Xen 3.3 supports non-PAE on i386.
   91: xenkernel41 provides Xen 4.1.  This is no longer maintained by Xen,
   92: but as of 2014-12 receives backported security patches.  There are no
   93: good reasons to run this version.  reasonable although trailing-edge
   94: choice.
   96: xenkernel42 provides Xen 4.2. This is no longer maintained by Xen, but
   97: as of 2014-12 receives backported security patches.  The only reason
   98: to run this is if you need to use xm instead of xl.
  100: xenkernel45 provides Xen 4.5.  This is new to pkgsrc as of 2015-01 and
  101: recommended for use as a conservative choice.
  103: xenkernel46 provides Xen 4.6.  TODO: Probably this is the recommended
  104: version.
  106: See also the [Xen Security Advisory page](
  108: Ideally newer versions of Xen will be added to pkgsrc.
  110: Note that NetBSD support is called XEN3.  It works with Xen 3 and Xen
  111: 4 because the hypercall interface has been stable.
  113: Xen command program
  114: -------------------
  116: Early Xen used a program called xm to manipulate the system from the
  117: dom0.  Starting in 4.1, a replacement program with similar behavior
  118: called xl is provided, but it does not work well in 4.1.  In 4.2, both
  119: xm and xl work fine.  4.4 is the last version that has xm.  You must
  120: choose one or the other, because it affects which daemons you run.
  122: NetBSD
  123: ------
  125: The netbsd-5, netbsd-6, netbsd-7, and -current branches are all
  126: reasonable choices, with more or less the same considerations for
  127: non-Xen use.  Therefore, netbsd-6 is recommended as the stable version
  128: of the most recent release for production use.  For those wanting to
  129: learn Xen or without production stability concerns, netbsd-7 is likely
  130: most appropriate.
  132: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
  133: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
  134: need SMP; the lack of support is really a problem when using a dom0 as
  135: a normal computer.)
  137: Architecture
  138: ------------
  140: Xen itself can run on i386 or amd64 machines.  (Practically, almost
  141: any computer where one would want to run Xen today supports amd64.)
  143: Xen, the dom0 kernel, and each domU kernel can be either i386 or
  144: amd64.  When building a xenkernel package, one obtains i386 on an i386
  145: host, and amd64 on an amd64 host.  If the xen kernel is i386, then the
  146: dom0 kernel and all domU kernels must be i386.  With an amd64 xen
  147: kernel, an amd64 dom0 kernel is known to work, and an i386 dom0 kernel
  148: should in theory work.  An amd64 xen/dom0 is known to support both
  149: i386 and amd64 domUs.
  151: i386 dom0 and domU kernels must be PAE (except for Xen 3.1); these are
  152: built by default.  (Note that emacs (at least) fails if run on i386
  153: with PAE when built without, and vice versa, presumably due to bugs in
  154: the undump code.)
  156: Because of the above, the standard approach is to use amd64 for the
  157: dom0.
  159: Xen 4.2 is the last version to support i386 as a host.  TODO: Clarify
  160: if this is about the CPU, the xen kernel, or the dom0 kernel having to
  161: be amd64.
  164: Stability
  165: ---------
  167: Mostly, NetBSD as a dom0 or domU is quite stable.
  168: However, there are some open PRs indicating problems.
  170:  - [PR 48125](
  171:  - [PR 47720](
  173: Note also that there are issues with sparse vnd(4) instances, but
  174: these are not about Xen.
  176: Recommendation
  177: --------------
  179: Therefore, this HOWTO recommends running xenkernel42 (and xentools42),
  180: xl, the NetBSD 6 stable branch, and to use an amd64 kernel as the
  181: dom0.  Either the i386 or amd64 of NetBSD may be used as domUs.
  183: Build problems
  184: --------------
  186: Ideally, all versions of Xen in pkgsrc would build on all versions of
  187: NetBSD on both i386 and amd64.  However, that isn't the case.  Besides
  188: aging code and aging compilers, qemu (included in xentools for HVM
  189: support) is difficult to build.  The following are known to work or FAIL:
  191:         xenkernel3 netbsd-5 amd64
  192:         xentools3 netbsd-5 amd64
  193:         xentools3=hvm netbsd-5 amd64 ????
  194:         xenkernel33 netbsd-5 amd64
  195:         xentools33 netbsd-5 amd64
  196:         xenkernel41 netbsd-5 amd64
  197:         xentools41 netbsd-5 amd64
  198:         xenkernel42 netbsd-5 amd64
  199:         xentools42 netbsd-5 amd64
  201:         xenkernel3 netbsd-6 i386 FAIL
  202:         xentools3 netbsd-6 i386
  203:         xentools3-hvm netbsd-6 i386 FAIL (dependencies fail)
  204:         xenkernel33 netbsd-6 i386
  205:         xentools33 netbsd-6 i386
  206:         xenkernel41 netbsd-6 i386
  207:         xentools41 netbsd-6 i386
  208:         xenkernel42 netbsd-6 i386
  209:         xentools42 netbsd-6 i386 *MIXED
  211: 	(all 3 and 33 seem to FAIL)
  212:         xenkernel41 netbsd-7 i386
  213:         xentools41 netbsd-7 i386
  214:         xenkernel42 netbsd-7 i386
  215:         xentools42 netbsd-7 i386 ??FAIL
  217: (*On netbsd-6 i386, there is a xentools42 in the 2014Q3 official builds,
  218: but it does not build for gdt.)
  220: NetBSD as a dom0
  221: ================
  223: NetBSD can be used as a dom0 and works very well.  The following
  224: sections address installation, updating NetBSD, and updating Xen.
  225: Note that it doesn't make sense to talk about installing a dom0 OS
  226: without also installing Xen itself.  We first address installing
  227: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
  228: NetBSD install to a dom0 install by just changing the kernel and boot
  229: configuration.
  231: For experimenting with Xen, a machine with as little as 1G of RAM and
  232: 100G of disk can work.  For running many domUs in productions, far
  233: more will be needed.
  235: Styles of dom0 operation
  236: ------------------------
  238: There are two basic ways to use Xen.  The traditional method is for
  239: the dom0 to do absolutely nothing other than providing support to some
  240: number of domUs.  Such a system was probably installed for the sole
  241: purpose of hosting domUs, and sits in a server room on a UPS.
  243: The other way is to put Xen under a normal-usage computer, so that the
  244: dom0 is what the computer would have been without Xen, perhaps a
  245: desktop or laptop.  Then, one can run domUs at will.  Purists will
  246: deride this as less secure than the previous approach, and for a
  247: computer whose purpose is to run domUs, they are right.  But Xen and a
  248: dom0 (without domUs) is not meaningfully less secure than the same
  249: things running without Xen.  One can boot Xen or boot regular NetBSD
  250: alternately with little problems, simply refraining from starting the
  251: Xen daemons when not running Xen.
  253: Note that NetBSD as dom0 does not support multiple CPUs.  This will
  254: limit the performance of the Xen/dom0 workstation approach.  In theory
  255: the only issue is that the "backend drivers" are not yet MPSAFE:
  258: Installation of NetBSD
  259: ----------------------
  261: First,
  262: [install NetBSD/amd64](/guide/inst/)
  263: just as you would if you were not using Xen.
  264: However, the partitioning approach is very important.
  266: If you want to use RAIDframe for the dom0, there are no special issues
  267: for Xen.  Typically one provides RAID storage for the dom0, and the
  268: domU systems are unaware of RAID.  The 2nd-stage loader bootxx_* skips
  269: over a RAID1 header to find /boot from a filesystem within a RAID
  270: partition; this is no different when booting Xen.
  272: There are 4 styles of providing backing storage for the virtual disks
  273: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN.
  275: With raw partitions, one has a disklabel (or gpt) partition sized for
  276: each virtual disk to be used by the domU.  (If you are able to predict
  277: how domU usage will evolve, please add an explanation to the HOWTO.
  278: Seriously, needs tend to change over time.)
  280: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
  281: for domU disks.  This is almost as efficient as raw disk partitions
  282: and more flexible.  Hence raw disk partitions should typically not
  283: be used.
  285: One can use files in the dom0 filesystem, typically created by dd'ing
  286: /dev/zero to create a specific size.  This is somewhat less efficient,
  287: but very convenient, as one can cp the files for backup, or move them
  288: between dom0 hosts.
  290: Finally, in theory one can place the files backing the domU disks in a
  291: SAN.  (This is an invitation for someone who has done this to add a
  292: HOWTO page.)
  294: Installation of Xen
  295: -------------------
  297: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
  298: pkgsrc (or another matching pair).
  299: See [the pkgsrc
  300: documentation]( for help with pkgsrc.
  302: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm.  More
  303: recent versions have HVM support integrated in the main xentools
  304: package.  It is entirely reasonable to run only PV guests.
  306: Next you need to install the selected Xen kernel itself, which is
  307: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz".  Copy it to /.
  308: For debugging, one may copy xen-debug.gz; this is conceptually similar
  309: to DIAGNOSTIC and DEBUG in NetBSD.  xen-debug.gz is basically only
  310: useful with a serial console.  Then, place a NetBSD XEN3_DOM0 kernel
  311: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
  312: of a NetBSD build.  If using i386, use
  313: releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz.  (If using Xen
  314: 3.1 and i386, you may use XEN3_DOM0 with the non-PAE Xen.  But you
  315: should not use Xen 3.1.)  Both xen and the NetBSD kernel may be (and
  316: typically are) left compressed.
  318: In a dom0 kernel, kernfs is mandatory for xend to comunicate with the
  319: kernel, so ensure that /kern is in fstab.  TODO: Say this is default,
  320: or file a PR and give a reference.
  322: Because you already installed NetBSD, you have a working boot setup
  323: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
  324: beginning of your root filesystem, /boot present, and likely
  325: /boot.cfg.  (If not, fix before continuing!)
  327: Add a line to to /boot.cfg to boot Xen.  See boot.cfg(5) for an
  328: example.  The basic line is
  330:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
  332: which specifies that the dom0 should have 256M, leaving the rest to be
  333: allocated for domUs.  To use a serial console, use
  335:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=256M console=com1 com1=9600,8n1
  337: which will use the first serial port for Xen (which counts starting
  338: from 1), forcing speed/parity, and also for NetBSD (which counts
  339: starting at 0).  In an attempt to add performance, one can also add
  341:         dom0_max_vcpus=1 dom0_vcpus_pin
  343: to force only one vcpu to be provided (since NetBSD dom0 can't use
  344: more) and to pin that vcpu to a physical cpu.  TODO: benchmark this.
  346: Xen has [many boot
  347: options](,
  348: and other tham dom0 memory and max_vcpus, they are generally not
  349: necessary.
  351: As with non-Xen systems, you should have a line to boot /netbsd (a
  352: kernel that works without Xen) and fallback versions of the non-Xen
  353: kernel, Xen, and the dom0 kernel.
  355: Now, reboot so that you are running a DOM0 kernel under Xen, rather
  356: than GENERIC without Xen.
  358: Using grub (historic)
  359: ---------------------
  361: Before NetBSD's native bootloader could support Xen, the use of
  362: grub was recommended.  If necessary, see the
  363: [old grub information](/ports/xen/howto-grub/).
  365: The [HowTo on Installing into
  366: RAID-1](
  367: explains how to set up booting a dom0 with Xen using grub with
  368: NetBSD's RAIDframe.  (This is obsolete with the use of NetBSD's native
  369: boot.)
  371: Configuring Xen
  372: ---------------
  374: Xen logs will be in /var/log/xen.
  376: Now, you have a system that will boot Xen and the dom0 kernel, but not
  377: do anything else special.  Make sure that you have rebooted into Xen.
  378: There will be no domUs, and none can be started because you still have
  379: to configure the dom0 daemons.
  381: The daemons which should be run vary with Xen version and with whether
  382: one is using xm or xl.  The Xen 3.1 and 3.3 packages use xm.  Xen 4.1
  383: and higher packages use xl.  While is is possible to use xm with some
  384: 4.x versions (TODO: 4.1 and 4.2?), the pkgsrc-provided rc.d scripts do
  385: not support this as of 2014-12-26, and thus the HOWTO does not support
  386: it either.  (Make sure your packages are reasonably recent.)
  388: For "xm" (3.1 and 3.3), you should enable xend and xenbackendd (but
  389: note that you should be using 4.x):
  391:         xend=YES
  392:         xenbackendd=YES
  394: For "xl" (4.x), you should enabled xend and xencommons (xenstored).
  395: Trying to boot 4.x without xencommons=YES will result in a hang; it is
  396: necessary to hig ^C on the console to let the machine finish booting.
  397: TODO: explain why xend is installed by the package.
  399:         xencommons=YES
  401: The installation of NetBSD should already have created devices for xen
  402: (xencons, xenevt), but if they are not present, create them:
  404:         cd /dev && sh MAKEDEV xen
  406: TODO: Recommend for/against xen-watchdog.
  408: After you have configured the daemons and either started them (in the
  409: order given) or rebooted, use xm or xl to inspect Xen's boot messages,
  410: available resources, and running domains.  An example with xl follows:
  412:         # xl dmesg
  413: 	[xen's boot info]
  414:         # xl info
  415: 	[available memory, etc.]
  416:         # xl list
  417:         Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  418:         Domain-0           0       64    0  r----     58.1
  420: ### Issues with xencommons
  422: xencommons starts xenstored, which stores data on behalf of dom0 and
  423: domUs.  It does not currently work to stop and start xenstored.
  424: Certainly all domUs should be shutdown first, following the sort order
  425: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
  426: and is not notified when xenstored exits, leading to not recreating
  427: the state when the new xenstored starts.  Until there's a mechanism to
  428: make this work, one should not expect to be able to restart xenstored
  429: (and thus xencommons).  There is currently no reason to expect that
  430: this will get fixed any time soon.
  432: anita (for testing NetBSD)
  433: --------------------------
  435: With the setup so far (assuming 4.2/xl), one should be able to run
  436: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
  437: root, because anita must create a domU):
  439:         anita --vmm=xl test file:///usr/obj/i386/
  441: Alternatively, one can use --vmm=xm to use xm-based domU creation
  442: instead (and must, on Xen <= 4.1).   TODO: confirm that anita xl really works.
  444: Xen-specific NetBSD issues
  445: --------------------------
  447: There are (at least) two additional things different about NetBSD as a
  448: dom0 kernel compared to hardware.
  450: One is that modules are not usable in DOM0 kernels, so one must
  451: compile in what's needed.  It's not really that modules cannot work,
  452: but that modules must be built for XEN3_DOM0 because some of the
  453: defines change and the normal module builds don't do this.  Basically,
  454: enabling Xen changes the kernel ABI, and the module build system
  455: doesn't cope with this.
  457: The other difference is that XEN3_DOM0 does not have exactly the same
  458: options as GENERIC.  While it is debatable whether or not this is a
  459: bug, users should be aware of this and can simply add missing config
  460: items if desired.
  462: Updating NetBSD in a dom0
  463: -------------------------
  465: This is just like updating NetBSD on bare hardware, assuming the new
  466: version supports the version of Xen you are running.  Generally, one
  467: replaces the kernel and reboots, and then overlays userland binaries
  468: and adjusts /etc.
  470: Note that one must update both the non-Xen kernel typically used for
  471: rescue purposes and the DOM0 kernel used with Xen.
  473: Converting from grub to /boot
  474: -----------------------------
  476: These instructions were [TODO: will be] used to convert a system from
  477: grub to /boot.  The system was originally installed in February of
  478: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
  479: over time.  Before these commands, it was running NetBSD 6 i386, Xen
  480: 4.1 and grub, much like the message linked earlier in the grub
  481: section.
  483:         # Install mbr bootblocks on both disks. 
  484:         fdisk -i /dev/rwd0d
  485:         fdisk -i /dev/rwd1d
  486:         # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
  487:         installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
  488:         installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
  489:         # Install secondary boot loader
  490:         cp -p /usr/mdec/boot /
  491:         # Create boog.cfg following earlier guidance:
  492:         menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
  493:         menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=256M
  494:         menu=GENERIC:boot
  495:         menu=GENERIC single-user:boot -s
  496:         menu=GENERIC.ok:boot netbsd.ok
  497:         menu=GENERIC.ok single-user:boot netbsd.ok -s
  498:         menu=Drop to boot prompt:prompt
  499:         default=1
  500:         timeout=30
  502: TODO: actually do this and fix it if necessary.
  504: Upgrading Xen versions
  505: ---------------------
  507: Updating Xen is conceptually not difficult, but can run into all the
  508: issues found when installing Xen.  Assuming migration from 4.1 to 4.2,
  509: remove the xenkernel41 and xentools41 packages and install the
  510: xenkernel42 and xentools42 packages.  Copy the 4.2 xen.gz to /.
  512: Ensure that the contents of /etc/rc.d/xen* are correct.  Specifically,
  513: they must match the package you just installed and not be left over
  514: from some previous installation.
  516: Enable the correct set of daemons; see the configuring section above.
  517: (Upgrading from 3.x to 4.x without doing this will result in a hang.)
  519: Ensure that the domU config files are valid for the new version.
  520: Specifically: remove autorestart=True, and ensure that disks are
  521: specified with numbers as the second argument, as the examples above
  522: show, and not NetBSD device names.
  524: Hardware known to work
  525: ----------------------
  527: Arguably, this section is misplaced, and there should be a page of
  528: hardware that runs NetBSD/amd64 well, with the mostly-well-founded
  529: assumption that NetBSD/xen runs fine on any modern hardware that
  530: NetBSD/amd64 runs well on.  Until then, we give motherboard/CPU/RAM
  531: triples to aid those choosing a motherboard.  Note that Xen systems
  532: usually do not run X, so a listing here does not imply that X works at
  533: all.
  535:         Supermicro X9SRL-F, Xeon E5-1650 v2, 96 GiB ECC
  536:         Supermicro ??, Atom C2758 (8 core), 32 GiB ECC
  537:         ASUS M5A78L-M/USB3 AM3+ microATX, AMD Piledriver X8 4000MHz, 16 GiB ECC
  539: Older hardware:
  541:         Intel D915GEV, Pentium4 CPU 3.40GHz, 4GB 533MHz Synchronous DDR2
  543: Running Xen under qemu
  544: ----------------------
  546: The astute reader will note that this section is somewhat twisted.
  547: However, it can be useful to run Xen under qemu either because the
  548: version of NetBSD as a dom0 does not run on the hardware in use, or to
  549: generate automated test cases involving Xen.
  551: In 2015-01, the following combination was reported to mostly work:
  553:         host OS: NetBSD/amd64 6.1.4
  554:         qemu: 2.2.0 from pkgsrc
  555:         Xen kernel: xenkernel42-4.2.5nb1 from pkgsrc
  556:         dom0 kernel: NetBSD/amd64 6.1.5
  557:         Xen tools: xentools42-4.2.5 from pkgsrc
  559: See [PR 47720]( for a problem with dom0
  560: shutdown.
  562: Unprivileged domains (domU)
  563: ===========================
  565: This section describes general concepts about domUs.  It does not
  566: address specific domU operating systems or how to install them.  The
  567: config files for domUs are typically in /usr/pkg/etc/xen, and are
  568: typically named so that the file name, domU name and the domU's host
  569: name match.
  571: The domU is provided with cpu and memory by Xen, configured by the
  572: dom0.  The domU is provided with disk and network by the dom0,
  573: mediated by Xen, and configured in the dom0.
  575: Entropy in domUs can be an issue; physical disks and network are on
  576: the dom0.  NetBSD's /dev/random system works, but is often challenged.
  578: Config files
  579: ------------
  581: There is no good order to present config files and the concepts
  582: surrounding what is being configured.  We first show an example config
  583: file, and then in the various sections give details.
  585: See (at least in xentools41) /usr/pkg/share/examples/xen/xmexample*,
  586: for a large number of well-commented examples, mostly for running
  587: GNU/Linux.
  589: The following is an example minimal domain configuration file
  590: "/usr/pkg/etc/xen/foo".  It is (with only a name change) an actual
  591: known working config file on Xen 4.1 (NetBSD 5 amd64 dom0 and NetBSD 5
  592: i386 domU).  The domU serves as a network file server.
  594:         # -*- mode: python; -*-
  596:         kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
  597:         memory = 1024
  598:         vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
  599:         disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
  600:                  'file:/n0/xen/foo-wd1,0x1,w' ]
  602: The domain will have the same name as the file.  The kernel has the
  603: host/domU name in it, so that on the dom0 one can update the various
  604: domUs independently.  The vif line causes an interface to be provided,
  605: with a specific mac address (do not reuse MAC addresses!), in bridge
  606: mode.  Two disks are provided, and they are both writable; the bits
  607: are stored in files and Xen attaches them to a vnd(4) device in the
  608: dom0 on domain creation.  The system treates xbd0 as the boot device
  609: without needing explicit configuration.
  611: By default xm looks for domain config files in /usr/pkg/etc/xen.  Note
  612: that "xm create" takes the name of a config file, while other commands
  613: take the name of a domain.  To create the domain, connect to the
  614: console, create the domain while attaching the console, shutdown the
  615: domain, and see if it has finished stopping, do (or xl with Xen >=
  616: 4.2):
  618:         xm create foo
  619:         xm console foo
  620:         xm create -c foo
  621:         xm shutdown foo
  622:         xm list
  624: Typing ^] will exit the console session.  Shutting down a domain is
  625: equivalent to pushing the power button; a NetBSD domU will receive a
  626: power-press event and do a clean shutdown.  Shutting down the dom0
  627: will trigger controlled shutdowns of all configured domUs.
  629: domU kernels
  630: ------------
  632: On a physical computer, the BIOS reads sector 0, and a chain of boot
  633: loaders finds and loads a kernel.  Normally this comes from the root
  634: filesystem.  With Xen domUs, the process is totally different.  The
  635: normal path is for the domU kernel to be a file in the dom0's
  636: filesystem.  At the request of the dom0, Xen loads that kernel into a
  637: new domU instance and starts execution.  While domU kernels can be
  638: anyplace, reasonable places to store domU kernels on the dom0 are in /
  639: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
  640: config files), or in /u0/xen (where the vdisks are).
  642: Note that loading the domU kernel from the dom0 implies that boot
  643: blocks, /boot, /boot.cfg, and so on are all ignored in the domU.
  644: See the VPS section near the end for discussion of alternate ways to
  645: obtain domU kernels.
  647: CPU and memory
  648: --------------
  650: A domain is provided with some number of vcpus, less than the number
  651: of cpus seen by the hypervisor.  (For a dom0, this is controlled by
  652: the boot argument "dom0_max_vcpus=1".)  For a domU, it is controlled
  653: from the config file by the "vcpus = N" directive.
  655: A domain is provided with memory; this is controlled in the config
  656: file by "memory = N" (in megabytes).  In the straightforward case, the
  657: sum of the the memory allocated to the dom0 and all domUs must be less
  658: than the available memory.
  660: Xen also provides a "balloon" driver, which can be used to let domains
  661: use more memory temporarily.  TODO: Explain better, and explain how
  662: well it works with NetBSD.
  664: Virtual disks
  665: -------------
  667: With the file/vnd style, typically one creates a directory,
  668: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
  669: domUs.  Then, for each domU disk, one writes zeros to a file that then
  670: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
  671: for the first virtual disk for the domU called foo.  Writing zeros to
  672: the file serves two purposes.  One is that preallocating the contents
  673: improves performance.  The other is that vnd on sparse files has
  674: failed to work.  TODO: give working/notworking NetBSD versions for
  675: sparse vnd.  Note that the use of file/vnd for Xen is not really
  676: different than creating a file-backed virtual disk for some other
  677: purpose, except that xentools handles the vnconfig commands.  To
  678: create an empty 4G virtual disk, simply do
  680:         dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
  682: Do not use qemu-img-xen, because this will create sparse file.  There
  683: have been recent (2015) reports of sparse vnd(4) devices causing
  684: lockups, but there is apparently no PR.
  686: With the lvm style, one creates logical devices.  They are then used
  687: similarly to vnds.  TODO: Add an example with lvm.
  689: In domU config files, the disks are defined as a sequence of 3-tuples.
  690: The first element is "method:/path/to/disk".  Common methods are
  691: "file:" for file-backed vnd. and "phy:" for something that is already
  692: a (TODO: character or block) device.
  694: The second element is an artifact of how virtual disks are passed to
  695: Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
  696: are given a device name to associate with the disk, and values like
  697: "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
  698: as xbd0, the second as xbd1, and so on.  However, xm/xl demand a
  699: second argument.  The name given is converted to a major/minor by
  700: calling stat(2) on the name in /dev and this is passed to the domU.
  701: In the general case, the dom0 and domU can be different operating
  702: systems, and it is an unwarranted assumption that they have consistent
  703: numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
  704: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
  705: for the second works fine and avoids this issue.  For a GNU/Linux
  706: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
  707: /dev/hda1.
  709: The third element is "w" for writable disks, and "r" for read-only
  710: disks.
  712: Virtual Networking
  713: ------------------
  715: Xen provides virtual ethernets, each of which connects the dom0 and a
  716: domU.  For each virtual network, there is an interface "xvifN.M" in
  717: the dom0, and in domU index N, a matching interface xennetM (NetBSD
  718: name).  The interfaces behave as if there is an Ethernet with two
  719: adaptors connected.  From this primitive, one can construct various
  720: configurations.  We focus on two common and useful cases for which
  721: there are existing scripts: bridging and NAT.
  723: With bridging (in the example above), the domU perceives itself to be
  724: on the same network as the dom0.  For server virtualization, this is
  725: usually best.  Bridging is accomplished by creating a bridge(4) device
  726: and adding the dom0's physical interface and the various xvifN.0
  727: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
  728: config file.  The bridge must be set up already in the dom0; an
  729: example /etc/ifconfig.bridge0 is:
  731:         create
  732:         up
  733:         !brconfig bridge0 add wm0
  735: With NAT, the domU perceives itself to be behind a NAT running on the
  736: dom0.  This is often appropriate when running Xen on a workstation.
  737: TODO: NAT appears to be configured by "vif = [ '' ]".
  739: The MAC address specified is the one used for the interface in the new
  740: domain.  The interface in dom0 will use this address XOR'd with
  741: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
  743: Sizing domains
  744: --------------
  746: Modern x86 hardware has vast amounts of resources.  However, many
  747: virtual servers can function just fine on far less.  A system with
  748: 256M of RAM and a 4G disk can be a reasonable choice.  Note that it is
  749: far easier to adjust virtual resources than physical ones.  For
  750: memory, it's just a config file edit and a reboot.  For disk, one can
  751: create a new file and vnconfig it (or lvm), and then dump/restore,
  752: just like updating physical disks, but without having to be there and
  753: without those pesky connectors.
  755: Starting domains automatically
  756: ------------------------------
  758: To start domains foo at bar at boot and shut them down cleanly on dom0
  759: shutdown, in rc.conf add:
  761:         xendomains="foo bar"
  763: Note that earlier versions of the xentools41 xendomains rc.d scripth
  764: usd xl, when one should use xm with 4.1.
  766: Creating specific unprivileged domains (domU)
  767: =============================================
  769: Creating domUs is almost entirely independent of operating system.  We
  770: have already presented the basics of config files.  Note that you must
  771: have already completed the dom0 setup so that "xl list" (or "xm list")
  772: works.
  774: Creating an unprivileged NetBSD domain (domU)
  775: ---------------------------------------------
  777: See the earlier config file, and adjust memory.  Decide on how much
  778: storage you will provide, and prepare it (file or lvm).
  780: While the kernel will be obtained from the dom0 filesystem, the same
  781: file should be present in the domU as /netbsd so that tools like
  782: savecore(8) can work.   (This is helpful but not necessary.)
  784: The kernel must be specifically for Xen and for use as a domU.  The
  785: i386 and amd64 provide the following kernels:
  787:         i386 XEN3_DOMU
  788:         i386 XEN3PAE_DOMU
  789:         amd64 XEN3_DOMU
  791: Unless using Xen 3.1 (and you shouldn't) with i386-mode Xen, you must
  792: use the PAE version of the i386 kernel.
  794: This will boot NetBSD, but this is not that useful if the disk is
  795: empty.  One approach is to unpack sets onto the disk outside of xen
  796: (by mounting it, just as you would prepare a physical disk for a
  797: system you can't run the installer on).
  799: A second approach is to run an INSTALL kernel, which has a miniroot
  800: and can load sets from the network.  To do this, copy the INSTALL
  801: kernel to / and change the kernel line in the config file to:
  803:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  805: Then, start the domain as "xl create -c configname".
  807: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
  808: line should be used in the config file.
  810:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  812: After booting the domain, the option to install via CDROM may be
  813: selected.  The CDROM device should be changed to `xbd1d`.
  815: Once done installing, "halt -p" the new domain (don't reboot or halt,
  816: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
  817: config file), switch the config file back to the XEN3_DOMU kernel,
  818: and start the new domain again. Now it should be able to use "root on
  819: xbd0a" and you should have a, functional NetBSD domU.
  821: TODO: check if this is still accurate.
  822: When the new domain is booting you'll see some warnings about *wscons*
  823: and the pseudo-terminals. These can be fixed by editing the files
  824: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  825: `/etc/ttys`, except *console*, like this:
  827:     console "/usr/libexec/getty Pc"         vt100   on secure
  828:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  829:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  830:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  831:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  833: Finally, all screens must be commented out from `/etc/wscons.conf`.
  835: It is also desirable to add
  837:         powerd=YES
  839: in rc.conf. This way, the domain will be properly shut down if
  840: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
  842: It is not strictly necessary to have a kernel (as /netbsd) in the domU
  843: filesystem.  However, various programs (e.g. netstat) will use that
  844: kernel to look up symbols to read from kernel virtual memory.  If
  845: /netbsd is not the running kernel, those lookups will fail.  (This is
  846: not really a Xen-specific issue, but because the domU kernel is
  847: obtained from the dom0, it is far more likely to be out of sync or
  848: missing with Xen.)
  850: Creating an unprivileged Linux domain (domU)
  851: --------------------------------------------
  853: Creating unprivileged Linux domains isn't much different from
  854: unprivileged NetBSD domains, but there are some details to know.
  856: First, the second parameter passed to the disk declaration (the '0x1' in
  857: the example below)
  859:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  861: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  862: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
  863: So, hda1 which has major 3 and minor 1 on a Linux system will have
  864: device number 0x301.  Alternatively, devices names can be used (hda,
  865: hdb, ...)  as xentools has a table to map these names to devices
  866: numbers.  To export a partition to a Linux guest we can use:
  868:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
  869:         root = "/dev/hda1 ro"
  871: and it will appear as /dev/hda on the Linux system, and be used as root
  872: partition.
  874: To install the Linux system on the partition to be exported to the
  875: guest domain, the following method can be used: install
  876: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
  877: that will be the root partition of your Linux domain, and mount it.
  878: Then copy the files from a working Linux system, make adjustments in
  879: `/etc` (fstab, network config).  It should also be possible to extract
  880: binary packages such as .rpm or .deb directly to the mounted partition
  881: using the appropriate tool, possibly running under NetBSD's Linux
  882: emulation.  Once the filesystem has been populated, umount it.  If
  883: desirable, the filesystem can be converted to ext3 using tune2fs -j.
  884: It should now be possible to boot the Linux guest domain, using one of
  885: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
  887: To get the linux console right, you need to add:
  889:     extra = "xencons=tty1"
  891: to your configuration since not all linux distributions auto-attach a
  892: tty to the xen console.
  894: Creating an unprivileged Solaris domain (domU)
  895: ----------------------------------------------
  897: See possibly outdated
  898: [Solaris domU instructions](/ports/xen/howto-solaris/).
  901: PCI passthrough: Using PCI devices in guest domains
  902: ---------------------------------------------------
  904: The dom0 can give other domains access to selected PCI
  905: devices. This can allow, for example, a non-privileged domain to have
  906: access to a physical network interface or disk controller.  However,
  907: keep in mind that giving a domain access to a PCI device most likely
  908: will give the domain read/write access to the whole physical memory,
  909: as PCs don't have an IOMMU to restrict memory access to DMA-capable
  910: device.  Also, it's not possible to export ISA devices to non-dom0
  911: domains, which means that the primary VGA adapter can't be exported.
  912: A guest domain trying to access the VGA registers will panic.
  914: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
  915: not been ported to later versions at this time.
  917: For a PCI device to be exported to a domU, is has to be attached to
  918: the "pciback" driver in dom0.  Devices passed to the dom0 via the
  919: pciback.hide boot parameter will attach to "pciback" instead of the
  920: usual driver.  The list of devices is specified as "(bus:dev.func)",
  921: where bus and dev are 2-digit hexadecimal numbers, and func a
  922: single-digit number:
  924:         pciback.hide=(00:0a.0)(00:06.0)
  926: pciback devices should show up in the dom0's boot messages, and the
  927: devices should be listed in the `/kern/xen/pci` directory.
  929: PCI devices to be exported to a domU are listed in the "pci" array of
  930: the domU's config file, with the format "0000:bus:dev.func".
  932:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
  934: In the domU an "xpci" device will show up, to which one or more pci
  935: busses will attach.  Then the PCI drivers will attach to PCI busses as
  936: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
  937: or any PCI drivers built in by default; you have to build your own
  938: kernel to use PCI devices in a domU.  Here's a kernel config example;
  939: note that only the "xpci" lines are unusual.
  941:         include         "arch/i386/conf/XEN3_DOMU"
  943:         # Add support for PCI busses to the XEN3_DOMU kernel
  944:         xpci* at xenbus ?
  945:         pci* at xpci ?
  947:         # PCI USB controllers
  948:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
  950:         # USB bus support
  951:         usb*    at uhci?
  953:         # USB Hubs
  954:         uhub*   at usb?
  955:         uhub*   at uhub? port ? configuration ? interface ?
  957:         # USB Mass Storage
  958:         umass*  at uhub? port ? configuration ? interface ?
  959:         wd*     at umass?
  960:         # SCSI controllers
  961:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
  963:         # SCSI bus support (for both ahc and umass)
  964:         scsibus* at scsi?
  966:         # SCSI devices
  967:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
  968:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
  971: NetBSD as a domU in a VPS
  972: =========================
  974: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
  975: hardware.  This section explains how to deal with Xen in a domU as a
  976: virtual private server where you do not control or have access to the
  977: dom0.  This is not intended to be an exhaustive list of VPS providers;
  978: only a few are mentioned that specifically support NetBSD.
  980: VPS operators provide varying degrees of access and mechanisms for
  981: configuration.  The big issue is usually how one controls which kernel
  982: is booted, because the kernel is nominally in the dom0 filesystem (to
  983: which VPS users do not normally have acesss).  A second issue is how
  984: to install NetBSD.
  985: A VPS user may want to compile a kernel for security updates, to run
  986: npf, run IPsec, or any other reason why someone would want to change
  987: their kernel.
  989: One approach is to have an adminstrative interface to upload a kernel,
  990: or to select from a prepopulated list.  Other approaches are pygrub
  991: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
  992: kernel from the domU filesystem.  This is closer to a regular physical
  993: computer, where someone who controls a machine can replace the kernel.
  995: A second issue is multiple CPUs.  With NetBSD 6, domUs support
  996: multiple vcpus, and it is typical for VPS providers to enable multiple
  997: CPUs for NetBSD domUs.
  999: pygrub
 1000: -------
 1002: pygrub runs in the dom0 and looks into the domU filesystem.  This
 1003: implies that the domU must have a kernel in a filesystem in a format
 1004: known to pygrub.  As of 2014, pygrub seems to be of mostly historical
 1005: interest.
 1007: pvgrub
 1008: ------
 1010: pvgrub is a version of grub that uses PV operations instead of BIOS
 1011: calls.  It is booted from the dom0 as the domU kernel, and then reads
 1012: /grub/menu.lst and loads a kernel from the domU filesystem.
 1014: [Panix]( lets users use pvgrub.  Panix reports
 1015: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
 1016: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
 1017: page](, which describes only
 1018: Linux but should be updated to cover NetBSD :-).
 1020: []( also lets users with pvgrub to boot
 1021: their own kernel.  See then [ NetBSD
 1022: HOWTO](
 1023: (which is in need of updating).
 1025: It appears that [grub's FFS
 1026: code](
 1027: does not support all aspects of modern FFS, but there are also reports
 1028: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
 1029: partition for the kernel with the intent that grub can understand it,
 1030: which leads to /netbsd not being the actual kernel.  One must remember
 1031: to update the special boot partiion.
 1033: Amazon
 1034: ------
 1036: See the [Amazon EC2 page](../amazon_ec2/).
 1038: Using npf
 1039: ---------
 1041: In standard kernels, npf is a module, and thus cannot be loaded in a
 1042: DOMU kernel.
 1044: TODO: Explain how to compile npf into a custom kernel, answering (but
 1045: note that the problem was caused by not booting the right kernel)
 1046: [this email to
 1047: netbsd-users](
 1049: TODO items for improving NetBSD/xen
 1050: ===================================
 1052: * Make the NetBSD dom0 kernel work with SMP.
 1053: * Test the Xen 4.5 packages adequately to be able to recommend them as
 1054:   the standard approach.
 1055: * Get PCI passthrough working on Xen 4.5
 1056: * Get pvgrub into pkgsrc, either via xentools or separately.
 1057: * grub
 1058:   * Check/add support to pkgsrc grub2 for UFS2 and arbitrary
 1059:     fragsize/blocksize (UFS2 support may be present; the point is to
 1060:     make it so that with any UFS1/UFS2 filesystem setup that works
 1061:     with NetBSD grub will also work).
 1062:     See [pkg/40258](
 1063:   * Push patches upstream.
 1064:   * Get UFS2 patches into pvgrub.
 1065: * Add support for PV ops to a version of /boot, and make it usable as
 1066:   a kernel in Xen, similar to pvgrub.
 1067: * Solve somehow the issue with modules for GENERIC not being loadable
 1068:   in a Xen dom0 or domU kernel.
 1070: Random pointers
 1071: ===============
 1073: TODO: This section contains links from elsewhere not yet integrated
 1074: into the HOWTO.
 1076: *
 1077: *

CVSweb for NetBSD wikisrc <> software: FreeBSD-CVSweb