File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.135: download - view: text, annotated - select for diffs
Wed Dec 21 16:57:10 2016 UTC (4 years, 4 months ago) by gdt
Branches: MAIN
CVS tags: HEAD
Adjust architecture section

Add link to "expensive syscalls on amd64" xen-devel message.
Make assorted minor improvements, including explaining how machine
size relates to sum of dom[0U] sizes.

    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 machine that runs NetBSD/amd64 will work with Xen and PV
   31: guests.  In theory i386 computers (without x86_64/amd64 support) can
   32: be used for Xen <= 4.2, but we have no recent reports of this working
   33: (this is a hint).  For HVM guests, hardware support is needed, but it
   34: is common on recent machines.  For Intel CPUs, one needs the VT-x
   35: extension, shown in "cpuctl identify 0" as VMX.  For AMD CPUs, one
   36: needs the AMD-V extensions, shown in "cpuctl identify 0" as SVM.
   37: There are further features for IOMMU virtualization, Intel's VT-d and
   38: AMD's AMD-Vi.  TODO: Explain whether Xen on NetBSD makes use of these
   39: features.  TODO: Review by someone who really understands this.
   41: Note that a FreeBSD dom0 requires VT-x and VT-d (or equivalent); this
   42: is because the FreeBSD dom0 does not run in PV mode.
   44: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
   45: The dom0 can start one or more domUs.  (Booting is explained in detail
   46: in the dom0 section.)
   48: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
   49: and that Xen kernels and tools are available in pkgsrc.  This HOWTO
   50: attempts to address both the case of running a NetBSD dom0 on hardware
   51: and running domUs under it (NetBSD and other), and also running NetBSD
   52: as a domU in a VPS.
   54: Xen 3.1 in pkgsrc supports "PCI passthrough", which means that
   55: specific PCI devices can be made available to a specific domU instead
   56: of the dom0.  This can be useful to let a domU run X11, or access some
   57: network interface or other peripheral.
   59: NetBSD 6 and earlier supported Xen 2; support was removed from NetBSD
   60: 7.  Xen 2 has been removed from pkgsrc.
   62: Prerequisites
   63: -------------
   65: Installing NetBSD/Xen is not extremely difficult, but it is more
   66: complex than a normal installation of NetBSD.
   67: In general, this HOWTO is occasionally overly restrictive about how
   68: things must be done, guiding the reader to stay on the established
   69: path when there are no known good reasons to stray.
   71: This HOWTO presumes a basic familiarity with the Xen system
   72: architecture, with installing NetBSD on i386/amd64 hardware, and with
   73: installing software from pkgsrc.  See also the [Xen
   74: website](
   76: Versions of Xen and NetBSD
   77: ==========================
   79: Most of the installation concepts and instructions are independent
   80: of Xen version and NetBSD version.  This section gives advice on
   81: which version to choose.  Versions not in pkgsrc and older unsupported
   82: versions of NetBSD are intentionally ignored.
   84: The term "amd64" is used to refer to both the NetBSD port and to the
   85: hardware architecture on which it runs.  (Such hardware is made by
   86: both Intel and AMD, and in 2016 a normal PC has this CPU
   87: architecture.)
   89: Xen
   90: ---
   92: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
   93: xenkernel and xentools.  We will refer only to the kernel versions,
   94: but note that both packages must be installed together and must have
   95: matching versions.
   97: xenkernel3 provides Xen 3.1.  It is no longer maintained by Xen, and
   98: the last applied security patch was in 2011. Thus, it should not be
   99: used.  It supports PCI passthrough, which is why people use it anyway.
  100: Xen 3.1 runs on i386 (both non-PAE and PAE) and amd64 hardware.
  102: xenkernel33 provides Xen 3.3.  It is no longer maintained by Xen, and
  103: the last applied security patch was in 2012.  Thus, it should not be
  104: used.  Xen 3.3 runs on i386 PAE and amd64 hardware.  There are no good
  105: reasons to run this version.
  107: xenkernel41 provides Xen 4.1.  It is no longer maintained by Xen, but
  108: as of 2016-12 received backported security patches.  Xen 4.1 runs on
  109: i386 PAE and amd64 hardware.  There are no good reasons to run this
  110: version.
  112: xenkernel42 provides Xen 4.2.  It is no longer maintained by Xen, but
  113: as of 2016-12 received backported security patches.  Xen 4.2 runs on
  114: i386 PAE and amd64 hardware.  The only reason to run this is if you
  115: need to use xm instead of xl, or if you need to run on hardware that
  116: supports i386 but not amd64.  (This might also be useful if you need
  117: an i386 dom0, if it turns out that an amd64 Xen kernel and an i386
  118: dom0 is problematic.)
  120: xenkernel45 provides Xen 4.5.  As of 2016-12, security patches were
  121: released by Xen and applied to pkgsrc.  Xen 4.5 runs on amd64 hardware
  122: only.  While slightly old, 4.5 has been tested and run by others, so
  123: it is the conservative choice.
  125: xenkernel46 provides Xen 4.6.  It is new to pkgsrc as of 2016-05.  As
  126: of 2016-12, security patches were released by Xen and applied to
  127: pkgsrc.  Xen 4.6 runs on amd64 hardware only For new installations,
  128: 4.6 is probably the appropriate choice and it will likely soon be the
  129: standard approach.
  131: Xen 4.7 (released 2016-06) and 4.8 (released 2016-12) are not yet in
  132: pkgsrc.
  134: See also the [Xen Security Advisory page](
  136: Note that NetBSD support is called XEN3.  It works with Xen 3 and Xen
  137: 4 because the hypercall interface has been stable.
  139: Xen command program
  140: -------------------
  142: Early Xen used a program called xm to manipulate the system from the
  143: dom0.  Starting in 4.1, a replacement program with similar behavior
  144: called xl is provided, but it does not work well in 4.1.  In 4.2, both
  145: xm and xl work fine.  4.4 is the last version that has xm.
  147: You must make a global choice to use xm or xl, because it affects not
  148: only which command you use, but the command used by rc.d scripts
  149: (specifically xendomains) and which daemons should be run.  The
  150: xentools packages provide xm for 3.1, 3.3 and 4.1 and xl for 4.2 and up.
  152: In 4.2, you can choose to use xm by simply changing the ctl_command
  153: variable and setting xend=YES in rc.conf.
  155: With xl, virtual devices are configured in parallel, which can cause
  156: problems if they are written assuming serial operation (e.g., updating
  157: firewall rules without explicit locking).  There is now locking for
  158: the provided scripts, which works for normal casses (e.g, file-backed
  159: xbd, where a vnd must be allocated).  But, as of 201612, it has not
  160: been adequately tested for a complex custom setup with a large number
  161: of interfaces.
  163: NetBSD
  164: ------
  166: The netbsd-6, netbsd-7, and -current branches are all reasonable
  167: choices, with more or less the same considerations for non-Xen use.
  168: Therefore, netbsd-7 is recommended as the stable version of the most
  169: recent release for production use.  In addition, netbsd-7 and -current
  170: have a important scheduler fix (in November of 2015) affecting
  171: contention between dom0 and domUs; see
  172: for a
  173: description.  For those wanting to learn Xen or without production
  174: stability concerns, netbsd-7 is still likely most appropriate, but
  175: -current is also a reasonable choice.  (Xen runs ok on netbsd-5, but
  176: the xentools packages are likely difficult to build, and netbsd-5 is
  177: not supported.)
  179: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
  180: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
  181: need SMP for dom0 functions; the lack of support is really a problem
  182: when using a dom0 as a normal computer.)
  184: Architecture
  185: ------------
  187: Xen itself can run on i386 (Xen < 4.2) or amd64 hardware (all Xen
  188: versions).  (Practically, almost any computer where one would want to
  189: run Xen today supports amd64.)
  191: Xen, the dom0 system, and each domU system can be either i386 or
  192: amd64.  When building a xenkernel package, one obtains an i386 Xen
  193: kernel on an i386 host, and an amd64 Xen kernel on an amd64 host.  If
  194: the Xen kernel is i386, then the dom0 kernel and all domU kernels must
  195: be i386.  With an amd64 Xen kernel, an amd64 dom0 kernel is known to
  196: work, and an i386 dom0 kernel should in theory work.  An amd64
  197: Xen/dom0 is known to support both i386 and amd64 domUs.
  199: i386 dom0 and domU kernels must be PAE (except for an i386 Xen 3.1
  200: kernel, where one can use non-PAE for dom0 and all domUs); PAE kernels
  201: are included in the NetBSD default build.  (Note that emacs (at least)
  202: fails if run on i386 with PAE when built without, and vice versa,
  203: presumably due to bugs in the undump code.)
  205: Because of the above, the standard approach is to use an amd64 Xen
  206: kernel and NetBSD/amd64 for the dom0.  For domUs, NetBSD/i386 (with
  207: the PAE kernel) and NetBSD/amd64 are in widespread use, and there is
  208: little to no Xen-specific reason to prefer one over the other.
  210: Note that to use an i386 dom0 with Xen 4.5 or higher, one must build
  211: (or obtain from pre-built packages) an amd64 Xen kernel and install
  212: that on the system.  (One must also use a PAE i386 kernel, but this is
  213: also required with an i386 Xen kernel.).  Almost no one in the
  214: NetBSD/Xen community does this, and the standard, well-tested,
  215: approach is to use an amd64 dom0.
  217: A [posting on
  218: xen-devel](
  219: explained that PV system call overhead was higher on amd64, and thus
  220: there is some notion that i386 guests are faster.  It goes on to
  221: caution that the total situation is complex and not entirely
  222: understood. On top of that caution, the post is about Linux, not
  223: NetBSD.  TODO: Include link to benchmarks, if someone posts them.
  225: Stability
  226: ---------
  228: Mostly, NetBSD as a dom0 or domU is quite stable.
  229: However, there are some open PRs indicating problems.
  231:  - [PR 48125](
  232:  - [PR 47720](
  234: Note also that there are issues with sparse vnd(4) instances, but
  235: these are not about Xen -- they just are noticed with sparse vnd(4)
  236: instances in support of virtual disks in a dom0.
  238: Recommendation
  239: --------------
  241: Therefore, this HOWTO recommends running xenkernel45 or xenkernel46,
  242: xl, the NetBSD 7 stable branch, and to use an amd64 kernel as the
  243: dom0.  Either the i386PAE or amd64 version of NetBSD may be used as
  244: domUs.
  246: Status
  247: ------
  249: Ideally, all versions of Xen in pkgsrc would build on all supported
  250: versions of NetBSD/amd64, to the point where this section would be
  251: silly.  However, that has not always been the case.  Besides aging
  252: code and aging compilers, qemu (included in xentools for HVM support)
  253: is difficult to build.  Note that there is intentionally no data for
  254: 4.5+ up for i386, and often omits xentools info if the corresponding
  255: kernel fails.
  257: The following table gives status, with the date last checked
  258: (generally on the most recent quarterly branch).  The first code is
  259: "builds" if it builds ok, and "FAIL" for a failure to build.  The
  260: second code/date only appears for xenkernel* and is "works" if it runs
  261: ok as a dom0 and can support a domU, and "FAIL" if it won't boot or
  262: run a domU.
  264: 	xenkernel3 netbsd-6 i386 FAIL 201612
  265: 	xenkernel33 netbsd-6 i386 FAIL 201612
  266: 	xenkernel41 netbsd-6 i386 builds 201612
  267: 	xenkernel42 netbsd-6 i386 builds 201612
  268: 	xentools3 netbsd-6 i386 FAIL 201612
  269: 	xentools33 netbsd-6 i386 FAIL 201612
  270: 	xentools41 netbsd-6 i386 builds 201612
  271: 	xentools42 netbsd-6 i386 FAIL 201612
  273: 	xenkernel3 netbsd-7 i386 FAIL 201412
  274: 	xenkernel33 netbsd-7 i386 FAIL 201412
  275: 	xenkernel41 netbsd-7 i386 builds 201412
  276: 	xenkernel42 netbsd-7 i386 builds 201412
  277: 	xentools41 netbsd-7 i386 builds 201412
  278: 	xentools42 netbsd-7 i386 ??FAIL 201412
  280: 	xenkernel3 netbsd-6 amd64 FAIL 201612
  281: 	xenkernel33 netbsd-6 amd64 FAIL 201612
  282: 	xenkernel41 netbsd-6 amd64 builds 201612 works 201612
  283: 	xenkernel42 netbsd-6 amd64 builds 201612 works 201612
  284: 	xenkernel45 netbsd-6 amd64 builds 201612
  285: 	xenkernel46 netbsd-6 amd64 builds 201612
  286: 	xentools41 netbsd-6 amd64 builds 201612
  287: 	xentools42 netbsd-6 amd64 builds 201612
  288: 	xentools45 netbsd-6 amd64 builds 201612
  289: 	xentools46 netbsd-6 amd64 FAIL 201612
  291: 	xenkernel3 netbsd-7 amd64 builds 201612
  292: 	xenkernel33 netbsd-7 amd64 builds 201612
  293: 	xenkernel41 netbsd-7 amd64 builds 201612
  294: 	xenkernel42 netbsd-7 amd64 builds 201612
  295: 	xenkernel45 netbsd-7 amd64 builds 201612
  296: 	xenkernel46 netbsd-7 amd64 builds 201612
  297: 	xentools3 netbsd-7 amd64 builds 201612
  298: 	xentools3-hvm netbsd-7 amd64 builds 201612
  299: 	xentools33 netbsd-7 amd64 FAIL 201612
  300: 	xentools41 netbsd-7 amd64 builds 201612
  301: 	xentools42 netbsd-7 amd64 builds 201612
  302: 	xentools45 netbsd-7 amd64 builds 201612
  303: 	xentools46 netbsd-7 amd64 builds 201612
  305: NetBSD as a dom0
  306: ================
  308: NetBSD can be used as a dom0 and works very well.  The following
  309: sections address installation, updating NetBSD, and updating Xen.
  310: Note that it doesn't make sense to talk about installing a dom0 OS
  311: without also installing Xen itself.  We first address installing
  312: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
  313: NetBSD install to a dom0 install by just changing the kernel and boot
  314: configuration.
  316: For experimenting with Xen, a machine with as little as 1G of RAM and
  317: 100G of disk can work.  For running many domUs in productions, far
  318: more will be needed; e.g. 4-8G and 1T of disk is reasonable for a
  319: half-dozen domUs of 512M and 32G each.  Basically, the RAM and disk
  320: have to be bigger than the sum of the RAM/disk needs of the dom0 and
  321: all the domUs.
  323: Styles of dom0 operation
  324: ------------------------
  326: There are two basic ways to use Xen.  The traditional method is for
  327: the dom0 to do absolutely nothing other than providing support to some
  328: number of domUs.  Such a system was probably installed for the sole
  329: purpose of hosting domUs, and sits in a server room on a UPS.
  331: The other way is to put Xen under a normal-usage computer, so that the
  332: dom0 is what the computer would have been without Xen, perhaps a
  333: desktop or laptop.  Then, one can run domUs at will.  Purists will
  334: deride this as less secure than the previous approach, and for a
  335: computer whose purpose is to run domUs, they are right.  But Xen and a
  336: dom0 (without domUs) is not meaningfully less secure than the same
  337: things running without Xen.  One can boot Xen or boot regular NetBSD
  338: alternately with little problems, simply refraining from starting the
  339: Xen daemons when not running Xen.
  341: Note that NetBSD as dom0 does not support multiple CPUs.  This will
  342: limit the performance of the Xen/dom0 workstation approach.  In theory
  343: the only issue is that the "backend drivers" are not yet MPSAFE:
  346: Installation of NetBSD
  347: ----------------------
  349: First,
  350: [install NetBSD/amd64](/guide/inst/)
  351: just as you would if you were not using Xen.
  352: However, the partitioning approach is very important.
  354: If you want to use RAIDframe for the dom0, there are no special issues
  355: for Xen.  Typically one provides RAID storage for the dom0, and the
  356: domU systems are unaware of RAID.  The 2nd-stage loader bootxx_* skips
  357: over a RAID1 header to find /boot from a file system within a RAID
  358: partition; this is no different when booting Xen.
  360: There are 4 styles of providing backing storage for the virtual disks
  361: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN.
  363: With raw partitions, one has a disklabel (or gpt) partition sized for
  364: each virtual disk to be used by the domU.  (If you are able to predict
  365: how domU usage will evolve, please add an explanation to the HOWTO.
  366: Seriously, needs tend to change over time.)
  368: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
  369: for domU disks.  This is almost as efficient as raw disk partitions
  370: and more flexible.  Hence raw disk partitions should typically not
  371: be used.
  373: One can use files in the dom0 file system, typically created by dd'ing
  374: /dev/zero to create a specific size.  This is somewhat less efficient,
  375: but very convenient, as one can cp the files for backup, or move them
  376: between dom0 hosts.
  378: Finally, in theory one can place the files backing the domU disks in a
  379: SAN.  (This is an invitation for someone who has done this to add a
  380: HOWTO page.)
  382: Installation of Xen
  383: -------------------
  385: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
  386: pkgsrc (or another matching pair).  See [the pkgsrc
  387: documentation]( for help with
  388: pkgsrc.  Ensure that your packages are recent; the HOWTO does not
  389: contemplate old builds.
  392: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm.  More
  393: recent versions have HVM support integrated in the main xentools
  394: package.  It is entirely reasonable to run only PV guests.
  396: Next you need to install the selected Xen kernel itself, which is
  397: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz".  Copy it to /.
  398: For debugging, one may copy xen-debug.gz; this is conceptually similar
  399: to DIAGNOSTIC and DEBUG in NetBSD.  xen-debug.gz is basically only
  400: useful with a serial console.  Then, place a NetBSD XEN3_DOM0 kernel
  401: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
  402: of a NetBSD build.  If using i386, use
  403: releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz.  (If using Xen
  404: 3.1 and i386, you may use XEN3_DOM0 with the non-PAE Xen.  But you
  405: should not use Xen 3.1.)  Both xen and the NetBSD kernel may be (and
  406: typically are) left compressed.
  408: In a dom0, kernfs is mandatory for xend to communicate with the
  409: kernel, so ensure that /kern is in fstab.  (A standard NetBSD install
  410: should already mount /kern.)
  412: Because you already installed NetBSD, you have a working boot setup
  413: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
  414: beginning of your root file system, have /boot, and likely also
  415: /boot.cfg.  (If not, fix before continuing!)
  417: Add a line to to /boot.cfg to boot Xen.  See boot.cfg(5) for an
  418: example.  The basic line is
  420:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
  422: which specifies that the dom0 should have 256M, leaving the rest to be
  423: allocated for domUs.  To use a serial console, use
  425:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=256M console=com1 com1=9600,8n1
  427: which will use the first serial port for Xen (which counts starting
  428: from 1), forcing speed/parity, and also for NetBSD (which counts
  429: starting at 0).  In an attempt to add performance, one can also add
  431:         dom0_max_vcpus=1 dom0_vcpus_pin
  433: to force only one vcpu to be provided (since NetBSD dom0 can't use
  434: more) and to pin that vcpu to a physical CPU.  TODO: benchmark this.
  436: Xen has [many boot
  437: options](,
  438: and other than dom0 memory and max_vcpus, they are generally not
  439: necessary.
  441: As with non-Xen systems, you should have a line to boot /netbsd (a
  442: kernel that works without Xen).  Consider a line to boot /netbsd.ok (a
  443: fallback version of the non-Xen kernel, updated manually when you are
  444: sure /netbsd is ok).  Consider also a line to boot fallback versions
  445: of Xen and the dom0 kernel, but note that non-Xen NetBSD can be used
  446: to resolve Xen booting issues.
  448: Probably you want a default=N line to choose Xen in the absence of
  449: intervention.
  451: Now, reboot so that you are running a DOM0 kernel under Xen, rather
  452: than GENERIC without Xen.
  454: Using grub (historic)
  455: ---------------------
  457: Before NetBSD's native bootloader could support Xen, the use of
  458: grub was recommended.  If necessary, see the
  459: [old grub information](/ports/xen/howto-grub).
  461: The [HowTo on Installing into
  462: RAID-1](
  463: explains how to set up booting a dom0 with Xen using grub with
  464: NetBSD's RAIDframe.  (This is obsolete with the use of NetBSD's native
  465: boot.  Now, just create a system with RAID-1, and alter /boot.cfg as
  466: described above.)
  468: Configuring Xen
  469: ---------------
  471: Xen logs will be in /var/log/xen.
  473: Now, you have a system that will boot Xen and the dom0 kernel, but not
  474: do anything else special.  Make sure that you have rebooted into Xen.
  475: There will be no domUs, and none can be started because you still have
  476: to configure the dom0 daemons.
  478: The daemons which should be run vary with Xen version and with whether
  479: one is using xm or xl.  The Xen 3.1, 3.3 and 4.1 packages use xm.  Xen
  480: 4.2 and up packages use xl.  To use xm with 4.2, edit xendomains to
  481: use xm instead.
  483: For 3.1 and 3.3, you should enable xend and xenbackendd:
  485:         xend=YES
  486:         xenbackendd=YES
  488: For 4.1 and up, you should enable xencommons.  Not enabling xencommons
  489: will result in a hang; it is necessary to hit ^C on the console to let
  490: the machine finish booting.  If you are using xm (default in 4.1, or
  491: if you changed xendomains in 4.2), you should also enable xend:
  493:         xend=YES # only if using xm, and only installed <= 4.2
  494:         xencommons=YES
  496: TODO: Recommend for/against xen-watchdog.
  498: After you have configured the daemons and either started them (in the
  499: order given) or rebooted, use xm or xl to inspect Xen's boot messages,
  500: available resources, and running domains.  An example with xl follows:
  502:         # xl dmesg
  503: 	[xen's boot info]
  504:         # xl info
  505: 	[available memory, etc.]
  506:         # xl list
  507:         Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  508:         Domain-0           0       64    0  r----     58.1
  510: ### Issues with xencommons
  512: xencommons starts xenstored, which stores data on behalf of dom0 and
  513: domUs.  It does not currently work to stop and start xenstored.
  514: Certainly all domUs should be shutdown first, following the sort order
  515: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
  516: and is not notified when xenstored exits, leading to not recreating
  517: the state when the new xenstored starts.  Until there's a mechanism to
  518: make this work, one should not expect to be able to restart xenstored
  519: (and thus xencommons).  There is currently no reason to expect that
  520: this will get fixed any time soon.
  522: ### No-longer needed advice about devices
  524: The installation of NetBSD should already have created devices for xen
  525: (xencons, xenevt, xsd_kva), but if they are not present, create them:
  527:         cd /dev && sh MAKEDEV xen
  529: anita (for testing NetBSD)
  530: --------------------------
  532: With the setup so far (assuming 4.2/xl), one should be able to run
  533: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
  534: root, because anita must create a domU):
  536:         anita --vmm=xl test file:///usr/obj/i386/
  538: Alternatively, one can use --vmm=xm to use xm-based domU creation
  539: instead (and must, on Xen <= 4.1).   TODO: confirm that anita xl really works.
  541: Xen-specific NetBSD issues
  542: --------------------------
  544: There are (at least) two additional things different about NetBSD as a
  545: dom0 kernel compared to hardware.
  547: One is that the module ABI is different because some of the #defines
  548: change, so one must build modules for Xen.  As of netbsd-7, the build
  549: system does this automatically.  TODO: check this.  (Before building
  550: Xen modules was added, it was awkward to use modules to the point
  551: where it was considered that it did not work.)
  553: The other difference is that XEN3_DOM0 does not have exactly the same
  554: options as GENERIC.  While it is debatable whether or not this is a
  555: bug, users should be aware of this and can simply add missing config
  556: items if desired.
  558: Updating NetBSD in a dom0
  559: -------------------------
  561: This is just like updating NetBSD on bare hardware, assuming the new
  562: version supports the version of Xen you are running.  Generally, one
  563: replaces the kernel and reboots, and then overlays userland binaries
  564: and adjusts /etc.
  566: Note that one must update both the non-Xen kernel typically used for
  567: rescue purposes and the DOM0 kernel used with Xen.
  569: Converting from grub to /boot
  570: -----------------------------
  572: These instructions were [TODO: will be] used to convert a system from
  573: grub to /boot.  The system was originally installed in February of
  574: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
  575: over time.  Before these commands, it was running NetBSD 6 i386, Xen
  576: 4.1 and grub, much like the message linked earlier in the grub
  577: section.
  579:         # Install MBR bootblocks on both disks. 
  580:         fdisk -i /dev/rwd0d
  581:         fdisk -i /dev/rwd1d
  582:         # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
  583:         installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
  584:         installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
  585:         # Install secondary boot loader
  586:         cp -p /usr/mdec/boot /
  587:         # Create boot.cfg following earlier guidance:
  588:         menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
  589:         menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=256M
  590:         menu=GENERIC:boot
  591:         menu=GENERIC single-user:boot -s
  592:         menu=GENERIC.ok:boot netbsd.ok
  593:         menu=GENERIC.ok single-user:boot netbsd.ok -s
  594:         menu=Drop to boot prompt:prompt
  595:         default=1
  596:         timeout=30
  598: TODO: actually do this and fix it if necessary.
  600: Upgrading Xen versions
  601: ---------------------
  603: Minor version upgrades are trivial.  Just rebuild/replace the
  604: xenkernel version and copy the new xen.gz to / (where /boot.cfg
  605: references it), and reboot.
  607: Major version upgrades are conceptually not difficult, but can run
  608: into all the issues found when installing Xen.  Assuming migration
  609: from 4.1 to 4.2, remove the xenkernel41 and xentools41 packages and
  610: install the xenkernel42 and xentools42 packages.  Copy the 4.2 xen.gz
  611: to /.
  613: Ensure that the contents of /etc/rc.d/xen* are correct.  Specifically,
  614: they must match the package you just installed and not be left over
  615: from some previous installation.
  617: Enable the correct set of daemons; see the configuring section above.
  618: (Upgrading from 3.x to 4.x without doing this will result in a hang.)
  620: Ensure that the domU config files are valid for the new version.
  621: Specifically, for 4.x remove autorestart=True, and ensure that disks
  622: are specified with numbers as the second argument, as the examples
  623: above show, and not NetBSD device names.
  625: Hardware known to work
  626: ----------------------
  628: Arguably, this section is misplaced, and there should be a page of
  629: hardware that runs NetBSD/amd64 well, with the mostly-well-founded
  630: assumption that NetBSD/xen runs fine on any modern hardware that
  631: NetBSD/amd64 runs well on.  Until then, we give motherboard/CPU (and
  632: sometimes RAM) pairs/triples to aid those choosing a motherboard.
  633: Note that Xen systems usually do not run X, so a listing here does not
  634: imply that X works at all.
  636:         Supermicro X9SRL-F, Xeon E5-1650 v2, 96 GiB ECC
  637:         Supermicro ??, Atom C2758 (8 core), 32 GiB ECC
  638:         ASUS M5A78L-M/USB3 AM3+ microATX, AMD Piledriver X8 4000MHz, 16 GiB ECC
  640: Older hardware:
  642:         Intel D915GEV, Pentium4 CPU 3.40GHz, 4GB 533MHz Synchronous DDR2
  643:         INTEL DG33FB, "Intel(R) Core(TM)2 Duo CPU     E6850  @ 3.00GHz"
  644:         INTEL DG33FB, "Intel(R) Core(TM)2 Duo CPU     E8400  @ 3.00GHz"
  646: Running Xen under qemu
  647: ----------------------
  649: The astute reader will note that this section is somewhat twisted.
  650: However, it can be useful to run Xen under qemu either because the
  651: version of NetBSD as a dom0 does not run on the hardware in use, or to
  652: generate automated test cases involving Xen.
  654: In 2015-01, the following combination was reported to mostly work:
  656:         host OS: NetBSD/amd64 6.1.4
  657:         qemu: 2.2.0 from pkgsrc
  658:         Xen kernel: xenkernel42-4.2.5nb1 from pkgsrc
  659:         dom0 kernel: NetBSD/amd64 6.1.5
  660:         Xen tools: xentools42-4.2.5 from pkgsrc
  662: See [PR 47720]( for a problem with dom0
  663: shutdown.
  665: Unprivileged domains (domU)
  666: ===========================
  668: This section describes general concepts about domUs.  It does not
  669: address specific domU operating systems or how to install them.  The
  670: config files for domUs are typically in /usr/pkg/etc/xen, and are
  671: typically named so that the file name, domU name and the domU's host
  672: name match.
  674: The domU is provided with CPU and memory by Xen, configured by the
  675: dom0.  The domU is provided with disk and network by the dom0,
  676: mediated by Xen, and configured in the dom0.
  678: Entropy in domUs can be an issue; physical disks and network are on
  679: the dom0.  NetBSD's /dev/random system works, but is often challenged.
  681: Config files
  682: ------------
  684: There is no good order to present config files and the concepts
  685: surrounding what is being configured.  We first show an example config
  686: file, and then in the various sections give details.
  688: See (at least in xentools41) /usr/pkg/share/examples/xen/xmexample*,
  689: for a large number of well-commented examples, mostly for running
  690: GNU/Linux.
  692: The following is an example minimal domain configuration file
  693: "/usr/pkg/etc/xen/foo".  It is (with only a name change) an actual
  694: known working config file on Xen 4.1 (NetBSD 5 amd64 dom0 and NetBSD 5
  695: i386 domU).  The domU serves as a network file server.
  697:         # -*- mode: python; -*-
  699:         kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
  700:         memory = 1024
  701:         vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
  702:         disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
  703:                  'file:/n0/xen/foo-wd1,0x1,w' ]
  705: The domain will have the same name as the file.  The kernel has the
  706: host/domU name in it, so that on the dom0 one can update the various
  707: domUs independently.  The vif line causes an interface to be provided,
  708: with a specific mac address (do not reuse MAC addresses!), in bridge
  709: mode.  Two disks are provided, and they are both writable; the bits
  710: are stored in files and Xen attaches them to a vnd(4) device in the
  711: dom0 on domain creation.  The system treats xbd0 as the boot device
  712: without needing explicit configuration.
  714: By default xm looks for domain config files in /usr/pkg/etc/xen.  Note
  715: that "xm create" takes the name of a config file, while other commands
  716: take the name of a domain.  To create the domain, connect to the
  717: console, create the domain while attaching the console, shutdown the
  718: domain, and see if it has finished stopping, do (or xl with Xen >=
  719: 4.2):
  721:         xm create foo
  722:         xm console foo
  723:         xm create -c foo
  724:         xm shutdown foo
  725:         xm list
  727: Typing ^] will exit the console session.  Shutting down a domain is
  728: equivalent to pushing the power button; a NetBSD domU will receive a
  729: power-press event and do a clean shutdown.  Shutting down the dom0
  730: will trigger controlled shutdowns of all configured domUs.
  732: domU kernels
  733: ------------
  735: On a physical computer, the BIOS reads sector 0, and a chain of boot
  736: loaders finds and loads a kernel.  Normally this comes from the root
  737: file system.  With Xen domUs, the process is totally different.  The
  738: normal path is for the domU kernel to be a file in the dom0's
  739: file system.  At the request of the dom0, Xen loads that kernel into a
  740: new domU instance and starts execution.  While domU kernels can be
  741: anyplace, reasonable places to store domU kernels on the dom0 are in /
  742: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
  743: config files), or in /u0/xen (where the vdisks are).
  745: Note that loading the domU kernel from the dom0 implies that boot
  746: blocks, /boot, /boot.cfg, and so on are all ignored in the domU.
  747: See the VPS section near the end for discussion of alternate ways to
  748: obtain domU kernels.
  750: CPU and memory
  751: --------------
  753: A domain is provided with some number of vcpus, less than the number
  754: of CPUs seen by the hypervisor.  (For a dom0, this is controlled by
  755: the boot argument "dom0_max_vcpus=1".)  For a domU, it is controlled
  756: from the config file by the "vcpus = N" directive.
  758: A domain is provided with memory; this is controlled in the config
  759: file by "memory = N" (in megabytes).  In the straightforward case, the
  760: sum of the the memory allocated to the dom0 and all domUs must be less
  761: than the available memory.
  763: Xen also provides a "balloon" driver, which can be used to let domains
  764: use more memory temporarily.  TODO: Explain better, and explain how
  765: well it works with NetBSD.
  767: Virtual disks
  768: -------------
  770: With the file/vnd style, typically one creates a directory,
  771: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
  772: domUs.  Then, for each domU disk, one writes zeros to a file that then
  773: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
  774: for the first virtual disk for the domU called foo.  Writing zeros to
  775: the file serves two purposes.  One is that preallocating the contents
  776: improves performance.  The other is that vnd on sparse files has
  777: failed to work.  TODO: give working/notworking NetBSD versions for
  778: sparse vnd and gnats reference.  Note that the use of file/vnd for Xen
  779: is not really different than creating a file-backed virtual disk for
  780: some other purpose, except that xentools handles the vnconfig
  781: commands.  To create an empty 4G virtual disk, simply do
  783:         dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
  785: Do not use qemu-img-xen, because this will create sparse file.  There
  786: have been recent (2015) reports of sparse vnd(4) devices causing
  787: lockups, but there is apparently no PR.
  789: With the lvm style, one creates logical devices.  They are then used
  790: similarly to vnds.  TODO: Add an example with lvm.
  792: In domU config files, the disks are defined as a sequence of 3-tuples.
  793: The first element is "method:/path/to/disk".  Common methods are
  794: "file:" for file-backed vnd. and "phy:" for something that is already
  795: a (TODO: character or block) device.
  797: The second element is an artifact of how virtual disks are passed to
  798: Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
  799: are given a device name to associate with the disk, and values like
  800: "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
  801: as xbd0, the second as xbd1, and so on.  However, xm/xl demand a
  802: second argument.  The name given is converted to a major/minor by
  803: calling stat(2) on the name in /dev and this is passed to the domU.
  804: In the general case, the dom0 and domU can be different operating
  805: systems, and it is an unwarranted assumption that they have consistent
  806: numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
  807: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
  808: for the second works fine and avoids this issue.  For a GNU/Linux
  809: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
  810: /dev/hda1.
  812: The third element is "w" for writable disks, and "r" for read-only
  813: disks.
  815: Note that NetBSD by default creates only vnd[0123].  If you need more
  816: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
  817: dom0.
  819: Note that NetBSD by default creates only xbd[0123].  If you need more
  820: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
  822: Virtual Networking
  823: ------------------
  825: Xen provides virtual Ethernets, each of which connects the dom0 and a
  826: domU.  For each virtual network, there is an interface "xvifN.M" in
  827: the dom0, and in domU index N, a matching interface xennetM (NetBSD
  828: name).  The interfaces behave as if there is an Ethernet with two
  829: adapters connected.  From this primitive, one can construct various
  830: configurations.  We focus on two common and useful cases for which
  831: there are existing scripts: bridging and NAT.
  833: With bridging (in the example above), the domU perceives itself to be
  834: on the same network as the dom0.  For server virtualization, this is
  835: usually best.  Bridging is accomplished by creating a bridge(4) device
  836: and adding the dom0's physical interface and the various xvifN.0
  837: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
  838: config file.  The bridge must be set up already in the dom0; an
  839: example /etc/ifconfig.bridge0 is:
  841:         create
  842:         up
  843:         !brconfig bridge0 add wm0
  845: With NAT, the domU perceives itself to be behind a NAT running on the
  846: dom0.  This is often appropriate when running Xen on a workstation.
  847: TODO: NAT appears to be configured by "vif = [ '' ]".
  849: The MAC address specified is the one used for the interface in the new
  850: domain.  The interface in dom0 will use this address XOR'd with
  851: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
  853: Sizing domains
  854: --------------
  856: Modern x86 hardware has vast amounts of resources.  However, many
  857: virtual servers can function just fine on far less.  A system with
  858: 256M of RAM and a 4G disk can be a reasonable choice.  Note that it is
  859: far easier to adjust virtual resources than physical ones.  For
  860: memory, it's just a config file edit and a reboot.  For disk, one can
  861: create a new file and vnconfig it (or lvm), and then dump/restore,
  862: just like updating physical disks, but without having to be there and
  863: without those pesky connectors.
  865: Starting domains automatically
  866: ------------------------------
  868: To start domains foo at bar at boot and shut them down cleanly on dom0
  869: shutdown, in rc.conf add:
  871:         xendomains="foo bar"
  873: Note that earlier versions of the xentools41 xendomains rc.d script
  874: used xl, when one should use xm with 4.1.
  876: Creating specific unprivileged domains (domU)
  877: =============================================
  879: Creating domUs is almost entirely independent of operating system.  We
  880: have already presented the basics of config files.  Note that you must
  881: have already completed the dom0 setup so that "xl list" (or "xm list")
  882: works.
  884: Creating an unprivileged NetBSD domain (domU)
  885: ---------------------------------------------
  887: See the earlier config file, and adjust memory.  Decide on how much
  888: storage you will provide, and prepare it (file or lvm).
  890: While the kernel will be obtained from the dom0 file system, the same
  891: file should be present in the domU as /netbsd so that tools like
  892: savecore(8) can work.   (This is helpful but not necessary.)
  894: The kernel must be specifically for Xen and for use as a domU.  The
  895: i386 and amd64 provide the following kernels:
  897:         i386 XEN3_DOMU
  898:         i386 XEN3PAE_DOMU
  899:         amd64 XEN3_DOMU
  901: Unless using Xen 3.1 (and you shouldn't) with i386-mode Xen, you must
  902: use the PAE version of the i386 kernel.
  904: This will boot NetBSD, but this is not that useful if the disk is
  905: empty.  One approach is to unpack sets onto the disk outside of xen
  906: (by mounting it, just as you would prepare a physical disk for a
  907: system you can't run the installer on).
  909: A second approach is to run an INSTALL kernel, which has a miniroot
  910: and can load sets from the network.  To do this, copy the INSTALL
  911: kernel to / and change the kernel line in the config file to:
  913:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  915: Then, start the domain as "xl create -c configname".
  917: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
  918: line should be used in the config file.
  920:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  922: After booting the domain, the option to install via CDROM may be
  923: selected.  The CDROM device should be changed to `xbd1d`.
  925: Once done installing, "halt -p" the new domain (don't reboot or halt,
  926: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
  927: config file), switch the config file back to the XEN3_DOMU kernel,
  928: and start the new domain again. Now it should be able to use "root on
  929: xbd0a" and you should have a, functional NetBSD domU.
  931: TODO: check if this is still accurate.
  932: When the new domain is booting you'll see some warnings about *wscons*
  933: and the pseudo-terminals. These can be fixed by editing the files
  934: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  935: `/etc/ttys`, except *console*, like this:
  937:     console "/usr/libexec/getty Pc"         vt100   on secure
  938:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  939:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  940:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  941:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  943: Finally, all screens must be commented out from `/etc/wscons.conf`.
  945: It is also desirable to add
  947:         powerd=YES
  949: in rc.conf. This way, the domain will be properly shut down if
  950: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
  952: It is not strictly necessary to have a kernel (as /netbsd) in the domU
  953: file system.  However, various programs (e.g. netstat) will use that
  954: kernel to look up symbols to read from kernel virtual memory.  If
  955: /netbsd is not the running kernel, those lookups will fail.  (This is
  956: not really a Xen-specific issue, but because the domU kernel is
  957: obtained from the dom0, it is far more likely to be out of sync or
  958: missing with Xen.)
  960: Creating an unprivileged Linux domain (domU)
  961: --------------------------------------------
  963: Creating unprivileged Linux domains isn't much different from
  964: unprivileged NetBSD domains, but there are some details to know.
  966: First, the second parameter passed to the disk declaration (the '0x1' in
  967: the example below)
  969:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  971: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  972: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
  973: So, hda1 which has major 3 and minor 1 on a Linux system will have
  974: device number 0x301.  Alternatively, devices names can be used (hda,
  975: hdb, ...)  as xentools has a table to map these names to devices
  976: numbers.  To export a partition to a Linux guest we can use:
  978:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
  979:         root = "/dev/hda1 ro"
  981: and it will appear as /dev/hda on the Linux system, and be used as root
  982: partition.
  984: To install the Linux system on the partition to be exported to the
  985: guest domain, the following method can be used: install
  986: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
  987: that will be the root partition of your Linux domain, and mount it.
  988: Then copy the files from a working Linux system, make adjustments in
  989: `/etc` (fstab, network config).  It should also be possible to extract
  990: binary packages such as .rpm or .deb directly to the mounted partition
  991: using the appropriate tool, possibly running under NetBSD's Linux
  992: emulation.  Once the file system has been populated, umount it.  If
  993: desirable, the file system can be converted to ext3 using tune2fs -j.
  994: It should now be possible to boot the Linux guest domain, using one of
  995: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
  997: To get the Linux console right, you need to add:
  999:     extra = "xencons=tty1"
 1001: to your configuration since not all Linux distributions auto-attach a
 1002: tty to the xen console.
 1004: Creating an unprivileged Solaris domain (domU)
 1005: ----------------------------------------------
 1007: See possibly outdated
 1008: [Solaris domU instructions](/ports/xen/howto-solaris/).
 1011: PCI passthrough: Using PCI devices in guest domains
 1012: ---------------------------------------------------
 1014: The dom0 can give other domains access to selected PCI
 1015: devices. This can allow, for example, a non-privileged domain to have
 1016: access to a physical network interface or disk controller.  However,
 1017: keep in mind that giving a domain access to a PCI device most likely
 1018: will give the domain read/write access to the whole physical memory,
 1019: as PCs don't have an IOMMU to restrict memory access to DMA-capable
 1020: device.  Also, it's not possible to export ISA devices to non-dom0
 1021: domains, which means that the primary VGA adapter can't be exported.
 1022: A guest domain trying to access the VGA registers will panic.
 1024: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
 1025: not been ported to later versions at this time.
 1027: For a PCI device to be exported to a domU, is has to be attached to
 1028: the "pciback" driver in dom0.  Devices passed to the dom0 via the
 1029: pciback.hide boot parameter will attach to "pciback" instead of the
 1030: usual driver.  The list of devices is specified as "(bus:dev.func)",
 1031: where bus and dev are 2-digit hexadecimal numbers, and func a
 1032: single-digit number:
 1034:         pciback.hide=(00:0a.0)(00:06.0)
 1036: pciback devices should show up in the dom0's boot messages, and the
 1037: devices should be listed in the `/kern/xen/pci` directory.
 1039: PCI devices to be exported to a domU are listed in the "pci" array of
 1040: the domU's config file, with the format "0000:bus:dev.func".
 1042:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
 1044: In the domU an "xpci" device will show up, to which one or more pci
 1045: buses will attach.  Then the PCI drivers will attach to PCI buses as
 1046: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
 1047: or any PCI drivers built in by default; you have to build your own
 1048: kernel to use PCI devices in a domU.  Here's a kernel config example;
 1049: note that only the "xpci" lines are unusual.
 1051:         include         "arch/i386/conf/XEN3_DOMU"
 1053:         # Add support for PCI buses to the XEN3_DOMU kernel
 1054:         xpci* at xenbus ?
 1055:         pci* at xpci ?
 1057:         # PCI USB controllers
 1058:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
 1060:         # USB bus support
 1061:         usb*    at uhci?
 1063:         # USB Hubs
 1064:         uhub*   at usb?
 1065:         uhub*   at uhub? port ? configuration ? interface ?
 1067:         # USB Mass Storage
 1068:         umass*  at uhub? port ? configuration ? interface ?
 1069:         wd*     at umass?
 1070:         # SCSI controllers
 1071:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
 1073:         # SCSI bus support (for both ahc and umass)
 1074:         scsibus* at scsi?
 1076:         # SCSI devices
 1077:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
 1078:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
 1081: NetBSD as a domU in a VPS
 1082: =========================
 1084: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
 1085: hardware.  This section explains how to deal with Xen in a domU as a
 1086: virtual private server where you do not control or have access to the
 1087: dom0.  This is not intended to be an exhaustive list of VPS providers;
 1088: only a few are mentioned that specifically support NetBSD.
 1090: VPS operators provide varying degrees of access and mechanisms for
 1091: configuration.  The big issue is usually how one controls which kernel
 1092: is booted, because the kernel is nominally in the dom0 file system (to
 1093: which VPS users do not normally have access).  A second issue is how
 1094: to install NetBSD.
 1095: A VPS user may want to compile a kernel for security updates, to run
 1096: npf, run IPsec, or any other reason why someone would want to change
 1097: their kernel.
 1099: One approach is to have an administrative interface to upload a kernel,
 1100: or to select from a prepopulated list.  Other approaches are pygrub
 1101: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
 1102: kernel from the domU file system.  This is closer to a regular physical
 1103: computer, where someone who controls a machine can replace the kernel.
 1105: A second issue is multiple CPUs.  With NetBSD 6, domUs support
 1106: multiple vcpus, and it is typical for VPS providers to enable multiple
 1107: CPUs for NetBSD domUs.
 1109: pygrub
 1110: -------
 1112: pygrub runs in the dom0 and looks into the domU file system.  This
 1113: implies that the domU must have a kernel in a file system in a format
 1114: known to pygrub.  As of 2014, pygrub seems to be of mostly historical
 1115: interest.
 1117: pvgrub
 1118: ------
 1120: pvgrub is a version of grub that uses PV operations instead of BIOS
 1121: calls.  It is booted from the dom0 as the domU kernel, and then reads
 1122: /grub/menu.lst and loads a kernel from the domU file system.
 1124: [Panix]( lets users use pvgrub.  Panix reports
 1125: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
 1126: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
 1127: page](, which describes only
 1128: Linux but should be updated to cover NetBSD :-).
 1130: []( also lets users with pvgrub to boot
 1131: their own kernel.  See then [ NetBSD
 1132: HOWTO](
 1133: (which is in need of updating).
 1135: It appears that [grub's FFS
 1136: code](
 1137: does not support all aspects of modern FFS, but there are also reports
 1138: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
 1139: partition for the kernel with the intent that grub can understand it,
 1140: which leads to /netbsd not being the actual kernel.  One must remember
 1141: to update the special boot partition.
 1143: Amazon
 1144: ------
 1146: See the [Amazon EC2 page](../amazon_ec2/).
 1148: Using npf
 1149: ---------
 1151: In standard kernels, npf is a module, and thus cannot be loaded in a
 1152: DOMU kernel.
 1154: TODO: Explain how to compile npf into a custom kernel, answering (but
 1155: note that the problem was caused by not booting the right kernel)
 1156: [this email to
 1157: netbsd-users](
 1159: TODO items for improving NetBSD/xen
 1160: ===================================
 1162: * Make the NetBSD dom0 kernel work with SMP.
 1163: * Test the Xen 4.5 packages adequately to be able to recommend them as
 1164:   the standard approach.
 1165: * Get PCI passthrough working on Xen 4.5
 1166: * Get pvgrub into pkgsrc, either via xentools or separately.
 1167: * grub
 1168:   * Check/add support to pkgsrc grub2 for UFS2 and arbitrary
 1169:     fragsize/blocksize (UFS2 support may be present; the point is to
 1170:     make it so that with any UFS1/UFS2 file system setup that works
 1171:     with NetBSD grub will also work).
 1172:     See [pkg/40258](
 1173:   * Push patches upstream.
 1174:   * Get UFS2 patches into pvgrub.
 1175: * Add support for PV ops to a version of /boot, and make it usable as
 1176:   a kernel in Xen, similar to pvgrub.
 1177: * Solve somehow the issue with modules for GENERIC not being loadable
 1178:   in a Xen dom0 or domU kernel.
 1180: Random pointers
 1181: ===============
 1183: This section contains links from elsewhere not yet integrated into the
 1184: HOWTO, and other guides.
 1186: *
 1187: *
 1188: *

CVSweb for NetBSD wikisrc <> software: FreeBSD-CVSweb