File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.128: download - view: text, annotated - select for diffs
Tue Dec 20 19:31:46 2016 UTC (2 years, 11 months ago) by gdt
Branches: MAIN
CVS tags: HEAD
Add more xentools3* data on netbsd-6 i386

(Yes, I realize this doesn't really matter, but I am cleaning up old
data.)

    1: Introduction
    2: ============
    3: 
    4: [![[Xen
    5: screenshot]](http://www.netbsd.org/gallery/in-Action/hubertf-xens.png)](http://www.netbsd.org/gallery/in-Action/hubertf-xen.png)
    6: 
    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.
   16: 
   17: Xen supports two styles of guests.  The original is Para-Virtualized
   18: (PV) which means that the guest OS does not attempt to access hardware
   19: directly, but instead makes hypercalls to the hypervisor.  This is
   20: analogous to a user-space program making system calls.  (The dom0
   21: operating system uses PV calls for some functions, such as updating
   22: memory mapping page tables, but has direct hardware access for disk
   23: and network.)   PV guests must be specifically coded for Xen.
   24: 
   25: The more recent style is HVM, which means that the guest does not have
   26: code for Xen and need not be aware that it is running under Xen.
   27: Attempts to access hardware registers are trapped and emulated.  This
   28: style is less efficient but can run unmodified guests.
   29: 
   30: 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.
   40: 
   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.
   43: 
   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.)
   47: 
   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.
   53: 
   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.
   58: 
   59: NetBSD 6 and earlier supported Xen 2; support was removed from NetBSD
   60: 7.  Xen 2 has been removed from pkgsrc.
   61: 
   62: Prerequisites
   63: -------------
   64: 
   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.
   70: 
   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](http://www.xenproject.org/).
   75: 
   76: Versions of Xen and NetBSD
   77: ==========================
   78: 
   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.
   83: 
   84: Xen
   85: ---
   86: 
   87: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
   88: xenkernel and xentools.  We will refer only to the kernel versions,
   89: but note that both packages must be installed together and must have
   90: matching versions.
   91: 
   92: xenkernel3 provides Xen 3.1.  It is no longer maintained by Xen, and the last applied security patch was in
   93: 2011. Thus, it should not be used.  It supports PCI passthrough,
   94: which is why people use it anyway. Xen 3.1 supports i386, both PAE and
   95: non-PAE.
   96: 
   97: xenkernel33 provides Xen 3.3.  It is no longer maintained by Xen, and
   98: the last applied security patch was in 2012.  Thus, it should not be
   99: used.  Xen 3.3 supports i386, but only in PAE mode.  There are no good
  100: reasons to run this version.
  101: 
  102: xenkernel41 provides Xen 4.1.  It is no longer maintained by Xen, but
  103: as of 2016-12 received backported security patches.  Xen 4.1 supports
  104: i386, but only in PAE mode.  There are no good reasons to run this
  105: version.
  106: 
  107: xenkernel42 provides Xen 4.2.  It is no longer maintained by Xen, but
  108: as of 2016-12 received backported security patches.  Xen 4.2 supports
  109: i386, but only in PAE mode.  The only reason to run this is if you
  110: need to use xm instead of xl, or if you need to run an i386 dom0
  111: (because your hardware is i386 only).
  112: 
  113: xenkernel45 provides Xen 4.5.  As of 2016-12, security patches were
  114: released by Xen and applied to pkgsrc.  Xen 4.5 requires using a dom0
  115: running NetBSD's amd64 port (Intel or AMD hardware is fine), but domUs
  116: can be amd64 or i386 PAE.  TODO: It is either a conservative choice or
  117: somewhat old.
  118: 
  119: xenkernel46 provides Xen 4.6.  It is new to pkgsrc as of 2016-05.  As
  120: of 2016-12, security patches were released by Xen and applied to
  121: pkgsrc.  Xen 4.6 similarly requires a NetBSD/amd64 dom0, but domUs can
  122: be amd64 or i386 PAE.  TODO: It is either a somewhat aggressive choice
  123: or the standard choice.
  124: 
  125: Xen 4.7 (released 2016-06) and 4.8 (released 2016-12) are not yet in
  126: pkgsrc.
  127: 
  128: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
  129: 
  130: Note that NetBSD support is called XEN3.  It works with Xen 3 and Xen
  131: 4 because the hypercall interface has been stable.
  132: 
  133: Xen command program
  134: -------------------
  135: 
  136: Early Xen used a program called xm to manipulate the system from the
  137: dom0.  Starting in 4.1, a replacement program with similar behavior
  138: called xl is provided, but it does not work well in 4.1.  In 4.2, both
  139: xm and xl work fine.  4.4 is the last version that has xm.
  140: 
  141: You must make a global choice to use xm or xl, because it affects not
  142: only which command you use, but the command used by rc.d scripts
  143: (specifically xendomains) and which daemons should be run.  The
  144: xentools packages provide xm for 3.1, 3.3 and 4.1 and xl for 4.2 and up.
  145: 
  146: In 4.2, you can choose to use xm by simply changing the ctl_command
  147: variable.
  148: 
  149: With xl, virtual devices are configured in parallel, which can cause
  150: problems if they are written assuming serial operation (e.g., updating
  151: firewall rules without explicit locking).
  152: 
  153: There is now locking for xl startup, but as of 201612 it has not been
  154: adequately tested.
  155: 
  156: NetBSD
  157: ------
  158: 
  159: The netbsd-6, netbsd-7, and -current branches are all reasonable
  160: choices, with more or less the same considerations for non-Xen use.
  161: Therefore, netbsd-7 is recommended as the stable version of the most
  162: recent release for production use.  In addition, netbsd-7 and -current
  163: have a important scheduler fix (in November of 2015) affecting
  164: contention between dom0 and domUs; see
  165: https://releng.netbsd.org/cgi-bin/req-7.cgi?show=1040 for a
  166: description.  For those wanting to learn Xen or without production
  167: stability concerns, netbsd-7 is still likely most appropriate, but
  168: -current is also a reasonable choice.  (Xen runs ok on netbsd-5, but
  169: the xentools packages are likely difficult to build, and netbsd-5 is
  170: not supported.)
  171: 
  172: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
  173: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
  174: need SMP for dom0 functions; the lack of support is really a problem
  175: when using a dom0 as a normal computer.)
  176: 
  177: Architecture
  178: ------------
  179: 
  180: Xen itself can run on i386 (Xen < 3.1) or amd64 machines (all Xen
  181: versions).  (Practically, almost any computer where one would want to
  182: run Xen today supports amd64.)
  183: 
  184: Xen, the dom0 kernel, and each domU kernel can be either i386 or
  185: amd64.  When building a xenkernel package, one obtains i386 on an i386
  186: host, and amd64 on an amd64 host.  If the Xen kernel is i386, then the
  187: dom0 kernel and all domU kernels must be i386.  With an amd64 Xen
  188: kernel, an amd64 dom0 kernel is known to work, and an i386PAE dom0
  189: kernel should in theory work.  An amd64 Xen/dom0 is known to support
  190: both i386PAE and amd64 domUs.
  191: 
  192: i386 dom0 and domU kernels must be PAE (except for Xen 3.1); these are
  193: built by default.  (Note that emacs (at least) fails if run on i386
  194: with PAE when built without, and vice versa, presumably due to bugs in
  195: the undump code.)
  196: 
  197: Because of the above, the standard approach is to use amd64 for the
  198: dom0.
  199: 
  200: Xen 4.2 is the last version to support i386 as a host.  TODO: Clarify
  201: if this is about the CPU, the Xen kernel, or the dom0 kernel having to
  202: be amd64.
  203: 
  204: 
  205: Stability
  206: ---------
  207: 
  208: Mostly, NetBSD as a dom0 or domU is quite stable.
  209: However, there are some open PRs indicating problems.
  210: 
  211:  - [PR 48125](http://gnats.netbsd.org/48125)
  212:  - [PR 47720](http://gnats.netbsd.org/47720)
  213: 
  214: Note also that there are issues with sparse vnd(4) instances, but
  215: these are not about Xen -- they just are noticed with sparse vnd(4)
  216: instances in support of virtual disks in a dom0.
  217: 
  218: Recommendation
  219: --------------
  220: 
  221: Therefore, this HOWTO recommends running xenkernel45 or xenkernel46,
  222: xl, the NetBSD 7 stable branch, and to use an amd64 kernel as the
  223: dom0.  Either the i386PAE or amd64 version of NetBSD may be used as
  224: domUs.
  225: 
  226: Status
  227: ------
  228: 
  229: Ideally, all versions of Xen in pkgsrc would build on all supported
  230: versions of NetBSD/amd64, to the point where this section would be
  231: silly.  However, that has not always been the case.  Besides aging
  232: code and aging compilers, qemu (included in xentools for HVM support)
  233: is difficult to build.  Note that there is intentionally no data for
  234: 4.5+ up for i386, and often omits xentools info if the corresponding
  235: kernel fails.
  236: 
  237: The following table gives status, with the date last checked
  238: (generally on the most recent quarterly branch).  The first code is
  239: "builds" if it builds ok, and "FAIL" for a failure to build.  The
  240: second code/date only appears for xenkernel* and is "works" if it runs
  241: ok as a dom0, and "FAIL" if it won't boot or run a domU.
  242: 
  243: 	xenkernel3 netbsd-6 i386 FAIL 201612
  244: 	xenkernel33 netbsd-6 i386 FAIL 201612
  245: 	xenkernel41 netbsd-6 i386 builds 201612
  246: 	xenkernel42 netbsd-6 i386 builds 201612
  247: 	xentools3 netbsd-6 i386 FAIL 201612
  248: 	xentools33 netbsd-6 i386 FAIL 201612
  249: 	xentools41 netbsd-6 i386 builds 201612
  250: 	xentools42 netbsd-6 i386 FAIL 201612
  251: 
  252: 	xenkernel3 netbsd-7 i386 FAIL 201412
  253: 	xenkernel33 netbsd-7 i386 FAIL 201412
  254: 	xenkernel41 netbsd-7 i386 builds 201412
  255: 	xenkernel42 netbsd-7 i386 builds 201412
  256: 	xentools41 netbsd-7 i386 builds 201412
  257: 	xentools42 netbsd-7 i386 ??FAIL 201412
  258: 
  259: 	xenkernel3 netbsd-6 amd64 FAIL 201612
  260: 	xenkernel33 netbsd-6 amd64 FAIL 201612
  261: 	xenkernel41 netbsd-6 amd64 builds 201612 works 201612
  262: 	xenkernel42 netbsd-6 amd64 builds 201612 works 201612
  263: 	xenkernel45 netbsd-6 amd64 builds 201612
  264: 	xenkernel46 netbsd-6 amd64 builds 201612
  265: 	xentools41 netbsd-6 amd64 builds 201612
  266: 	xentools42 netbsd-6 amd64 builds 201612
  267: 	xentools45 netbsd-6 amd64 builds 201612
  268: 	xentools46 netbsd-6 amd64 FAIL 201612
  269: 
  270: 	xenkernel3 netbsd-7 amd64 builds 201612
  271: 	xenkernel33 netbsd-7 amd64 builds 201612
  272: 	xenkernel41 netbsd-7 amd64 builds 201612
  273: 	xenkernel42 netbsd-7 amd64 builds 201612
  274: 	xenkernel45 netbsd-7 amd64 builds 201612
  275: 	xenkernel46 netbsd-7 amd64 builds 201612
  276: 	xentools3 netbsd-7 amd64 builds 201612
  277: 	xentools3-hvm netbsd-7 amd64 builds 201612
  278: 	xentools33 netbsd-7 amd64 FAIL 201612
  279: 	xentools41 netbsd-7 amd64 builds 201612
  280: 	xentools42 netbsd-7 amd64 builds 201612
  281: 	xentools45 netbsd-7 amd64 builds 201612
  282: 	xentools46 netbsd-7 amd64 builds 201612
  283: 
  284: NetBSD as a dom0
  285: ================
  286: 
  287: NetBSD can be used as a dom0 and works very well.  The following
  288: sections address installation, updating NetBSD, and updating Xen.
  289: Note that it doesn't make sense to talk about installing a dom0 OS
  290: without also installing Xen itself.  We first address installing
  291: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
  292: NetBSD install to a dom0 install by just changing the kernel and boot
  293: configuration.
  294: 
  295: For experimenting with Xen, a machine with as little as 1G of RAM and
  296: 100G of disk can work.  For running many domUs in productions, far
  297: more will be needed.
  298: 
  299: Styles of dom0 operation
  300: ------------------------
  301: 
  302: There are two basic ways to use Xen.  The traditional method is for
  303: the dom0 to do absolutely nothing other than providing support to some
  304: number of domUs.  Such a system was probably installed for the sole
  305: purpose of hosting domUs, and sits in a server room on a UPS.
  306: 
  307: The other way is to put Xen under a normal-usage computer, so that the
  308: dom0 is what the computer would have been without Xen, perhaps a
  309: desktop or laptop.  Then, one can run domUs at will.  Purists will
  310: deride this as less secure than the previous approach, and for a
  311: computer whose purpose is to run domUs, they are right.  But Xen and a
  312: dom0 (without domUs) is not meaningfully less secure than the same
  313: things running without Xen.  One can boot Xen or boot regular NetBSD
  314: alternately with little problems, simply refraining from starting the
  315: Xen daemons when not running Xen.
  316: 
  317: Note that NetBSD as dom0 does not support multiple CPUs.  This will
  318: limit the performance of the Xen/dom0 workstation approach.  In theory
  319: the only issue is that the "backend drivers" are not yet MPSAFE:
  320:   http://mail-index.netbsd.org/netbsd-users/2014/08/29/msg015195.html
  321: 
  322: Installation of NetBSD
  323: ----------------------
  324: 
  325: First,
  326: [install NetBSD/amd64](/guide/inst/)
  327: just as you would if you were not using Xen.
  328: However, the partitioning approach is very important.
  329: 
  330: If you want to use RAIDframe for the dom0, there are no special issues
  331: for Xen.  Typically one provides RAID storage for the dom0, and the
  332: domU systems are unaware of RAID.  The 2nd-stage loader bootxx_* skips
  333: over a RAID1 header to find /boot from a file system within a RAID
  334: partition; this is no different when booting Xen.
  335: 
  336: There are 4 styles of providing backing storage for the virtual disks
  337: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN.
  338: 
  339: With raw partitions, one has a disklabel (or gpt) partition sized for
  340: each virtual disk to be used by the domU.  (If you are able to predict
  341: how domU usage will evolve, please add an explanation to the HOWTO.
  342: Seriously, needs tend to change over time.)
  343: 
  344: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
  345: for domU disks.  This is almost as efficient as raw disk partitions
  346: and more flexible.  Hence raw disk partitions should typically not
  347: be used.
  348: 
  349: One can use files in the dom0 file system, typically created by dd'ing
  350: /dev/zero to create a specific size.  This is somewhat less efficient,
  351: but very convenient, as one can cp the files for backup, or move them
  352: between dom0 hosts.
  353: 
  354: Finally, in theory one can place the files backing the domU disks in a
  355: SAN.  (This is an invitation for someone who has done this to add a
  356: HOWTO page.)
  357: 
  358: Installation of Xen
  359: -------------------
  360: 
  361: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
  362: pkgsrc (or another matching pair).  See [the pkgsrc
  363: documentation](http://www.NetBSD.org/docs/pkgsrc/) for help with
  364: pkgsrc.  Ensure that your packages are recent; the HOWTO does not
  365: contemplate old builds.
  366: 
  367: 
  368: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm.  More
  369: recent versions have HVM support integrated in the main xentools
  370: package.  It is entirely reasonable to run only PV guests.
  371: 
  372: Next you need to install the selected Xen kernel itself, which is
  373: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz".  Copy it to /.
  374: For debugging, one may copy xen-debug.gz; this is conceptually similar
  375: to DIAGNOSTIC and DEBUG in NetBSD.  xen-debug.gz is basically only
  376: useful with a serial console.  Then, place a NetBSD XEN3_DOM0 kernel
  377: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
  378: of a NetBSD build.  If using i386, use
  379: releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz.  (If using Xen
  380: 3.1 and i386, you may use XEN3_DOM0 with the non-PAE Xen.  But you
  381: should not use Xen 3.1.)  Both xen and the NetBSD kernel may be (and
  382: typically are) left compressed.
  383: 
  384: In a dom0 kernel, kernfs is mandatory for xend to communicate with the
  385: kernel, so ensure that /kern is in fstab.  TODO: Say this is default,
  386: or file a PR and give a reference.
  387: 
  388: Because you already installed NetBSD, you have a working boot setup
  389: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
  390: beginning of your root file system, /boot present, and likely
  391: /boot.cfg.  (If not, fix before continuing!)
  392: 
  393: Add a line to to /boot.cfg to boot Xen.  See boot.cfg(5) for an
  394: example.  The basic line is
  395: 
  396:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
  397: 
  398: which specifies that the dom0 should have 256M, leaving the rest to be
  399: allocated for domUs.  To use a serial console, use
  400: 
  401:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=256M console=com1 com1=9600,8n1
  402: 
  403: which will use the first serial port for Xen (which counts starting
  404: from 1), forcing speed/parity, and also for NetBSD (which counts
  405: starting at 0).  In an attempt to add performance, one can also add
  406: 
  407:         dom0_max_vcpus=1 dom0_vcpus_pin
  408: 
  409: to force only one vcpu to be provided (since NetBSD dom0 can't use
  410: more) and to pin that vcpu to a physical CPU.  TODO: benchmark this.
  411: 
  412: Xen has [many boot
  413: options](http://xenbits.xenproject.org/docs/4.5-testing/misc/xen-command-line.html),
  414: and other than dom0 memory and max_vcpus, they are generally not
  415: necessary.
  416: 
  417: As with non-Xen systems, you should have a line to boot /netbsd (a
  418: kernel that works without Xen).  Consider a line to boot /netbsd.ok (a
  419: fallback version of the non-Xen kernel, updated manually when you are
  420: sure /netbsd is ok).  Consider also a line to boot fallback versions
  421: of Xen and the dom0 kernel, but note that non-Xen NetBSD can be used
  422: to resolve Xen booting issues.
  423: 
  424: Probably you want a default=N line to choose Xen in the absence of
  425: intervention.
  426: 
  427: Now, reboot so that you are running a DOM0 kernel under Xen, rather
  428: than GENERIC without Xen.
  429: 
  430: Using grub (historic)
  431: ---------------------
  432: 
  433: Before NetBSD's native bootloader could support Xen, the use of
  434: grub was recommended.  If necessary, see the
  435: [old grub information](/ports/xen/howto-grub/).
  436: 
  437: The [HowTo on Installing into
  438: RAID-1](http://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
  439: explains how to set up booting a dom0 with Xen using grub with
  440: NetBSD's RAIDframe.  (This is obsolete with the use of NetBSD's native
  441: boot.)
  442: 
  443: Configuring Xen
  444: ---------------
  445: 
  446: Xen logs will be in /var/log/xen.
  447: 
  448: Now, you have a system that will boot Xen and the dom0 kernel, but not
  449: do anything else special.  Make sure that you have rebooted into Xen.
  450: There will be no domUs, and none can be started because you still have
  451: to configure the dom0 daemons.
  452: 
  453: The daemons which should be run vary with Xen version and with whether
  454: one is using xm or xl.  The Xen 3.1, 3.3 and 4.1 packages use xm.  Xen
  455: 4.2 and up packages use xl.  To use xm with 4.2, edit xendomains to
  456: use xm instead.
  457: 
  458: For "xm", you should enable xend and xenbackendd:
  459: 
  460:         xend=YES
  461:         xenbackendd=YES
  462: 
  463: For "xl", you should enable xend and xencommons (xenstored).  Trying
  464: to use xl without xencommons=YES will result in a hang; it is
  465: necessary to hit ^C on the console to let the machine finish booting.
  466: TODO: Check if xend really should be used, and if not why it is
  467: installed.
  468: 
  469:         xend=YES
  470:         xencommons=YES
  471: 
  472: TODO: Recommend for/against xen-watchdog.
  473: 
  474: After you have configured the daemons and either started them (in the
  475: order given) or rebooted, use xm or xl to inspect Xen's boot messages,
  476: available resources, and running domains.  An example with xl follows:
  477: 
  478:         # xl dmesg
  479: 	[xen's boot info]
  480:         # xl info
  481: 	[available memory, etc.]
  482:         # xl list
  483:         Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  484:         Domain-0           0       64    0  r----     58.1
  485: 
  486: ### Issues with xencommons
  487: 
  488: xencommons starts xenstored, which stores data on behalf of dom0 and
  489: domUs.  It does not currently work to stop and start xenstored.
  490: Certainly all domUs should be shutdown first, following the sort order
  491: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
  492: and is not notified when xenstored exits, leading to not recreating
  493: the state when the new xenstored starts.  Until there's a mechanism to
  494: make this work, one should not expect to be able to restart xenstored
  495: (and thus xencommons).  There is currently no reason to expect that
  496: this will get fixed any time soon.
  497: 
  498: ### No-longer needed advice about devices
  499: 
  500: The installation of NetBSD should already have created devices for xen
  501: (xencons, xenevt, xsd_kva), but if they are not present, create them:
  502: 
  503:         cd /dev && sh MAKEDEV xen
  504: 
  505: anita (for testing NetBSD)
  506: --------------------------
  507: 
  508: With the setup so far (assuming 4.2/xl), one should be able to run
  509: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
  510: root, because anita must create a domU):
  511: 
  512:         anita --vmm=xl test file:///usr/obj/i386/
  513: 
  514: Alternatively, one can use --vmm=xm to use xm-based domU creation
  515: instead (and must, on Xen <= 4.1).   TODO: confirm that anita xl really works.
  516:     
  517: Xen-specific NetBSD issues
  518: --------------------------
  519: 
  520: There are (at least) two additional things different about NetBSD as a
  521: dom0 kernel compared to hardware.
  522: 
  523: One is that the module ABI is different because some of the #defines
  524: change, so one must build modules for Xen.  As of netbsd-7, the build
  525: system does this automatically.  TODO: check this.  (Before building
  526: Xen modules was added, it was awkward to use modules to the point
  527: where it was considered that it did not work.)
  528: 
  529: The other difference is that XEN3_DOM0 does not have exactly the same
  530: options as GENERIC.  While it is debatable whether or not this is a
  531: bug, users should be aware of this and can simply add missing config
  532: items if desired.
  533: 
  534: Updating NetBSD in a dom0
  535: -------------------------
  536: 
  537: This is just like updating NetBSD on bare hardware, assuming the new
  538: version supports the version of Xen you are running.  Generally, one
  539: replaces the kernel and reboots, and then overlays userland binaries
  540: and adjusts /etc.
  541: 
  542: Note that one must update both the non-Xen kernel typically used for
  543: rescue purposes and the DOM0 kernel used with Xen.
  544: 
  545: Converting from grub to /boot
  546: -----------------------------
  547: 
  548: These instructions were [TODO: will be] used to convert a system from
  549: grub to /boot.  The system was originally installed in February of
  550: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
  551: over time.  Before these commands, it was running NetBSD 6 i386, Xen
  552: 4.1 and grub, much like the message linked earlier in the grub
  553: section.
  554: 
  555:         # Install MBR bootblocks on both disks. 
  556:         fdisk -i /dev/rwd0d
  557:         fdisk -i /dev/rwd1d
  558:         # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
  559:         installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
  560:         installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
  561:         # Install secondary boot loader
  562:         cp -p /usr/mdec/boot /
  563:         # Create boot.cfg following earlier guidance:
  564:         menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
  565:         menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=256M
  566:         menu=GENERIC:boot
  567:         menu=GENERIC single-user:boot -s
  568:         menu=GENERIC.ok:boot netbsd.ok
  569:         menu=GENERIC.ok single-user:boot netbsd.ok -s
  570:         menu=Drop to boot prompt:prompt
  571:         default=1
  572:         timeout=30
  573: 
  574: TODO: actually do this and fix it if necessary.
  575: 
  576: Upgrading Xen versions
  577: ---------------------
  578: 
  579: Minor version upgrades are trivial.  Just rebuild/replace the
  580: xenkernel version and copy the new xen.gz to / (where /boot.cfg
  581: references it), and reboot.
  582: 
  583: Major version upgrades are conceptually not difficult, but can run
  584: into all the issues found when installing Xen.  Assuming migration
  585: from 4.1 to 4.2, remove the xenkernel41 and xentools41 packages and
  586: install the xenkernel42 and xentools42 packages.  Copy the 4.2 xen.gz
  587: to /.
  588: 
  589: Ensure that the contents of /etc/rc.d/xen* are correct.  Specifically,
  590: they must match the package you just installed and not be left over
  591: from some previous installation.
  592: 
  593: Enable the correct set of daemons; see the configuring section above.
  594: (Upgrading from 3.x to 4.x without doing this will result in a hang.)
  595: 
  596: Ensure that the domU config files are valid for the new version.
  597: Specifically, for 4.x remove autorestart=True, and ensure that disks
  598: are specified with numbers as the second argument, as the examples
  599: above show, and not NetBSD device names.
  600: 
  601: Hardware known to work
  602: ----------------------
  603: 
  604: Arguably, this section is misplaced, and there should be a page of
  605: hardware that runs NetBSD/amd64 well, with the mostly-well-founded
  606: assumption that NetBSD/xen runs fine on any modern hardware that
  607: NetBSD/amd64 runs well on.  Until then, we give motherboard/CPU/RAM
  608: triples to aid those choosing a motherboard.  Note that Xen systems
  609: usually do not run X, so a listing here does not imply that X works at
  610: all.
  611: 
  612:         Supermicro X9SRL-F, Xeon E5-1650 v2, 96 GiB ECC
  613:         Supermicro ??, Atom C2758 (8 core), 32 GiB ECC
  614:         ASUS M5A78L-M/USB3 AM3+ microATX, AMD Piledriver X8 4000MHz, 16 GiB ECC
  615: 
  616: Older hardware:
  617: 
  618:         Intel D915GEV, Pentium4 CPU 3.40GHz, 4GB 533MHz Synchronous DDR2
  619: 
  620: Running Xen under qemu
  621: ----------------------
  622: 
  623: The astute reader will note that this section is somewhat twisted.
  624: However, it can be useful to run Xen under qemu either because the
  625: version of NetBSD as a dom0 does not run on the hardware in use, or to
  626: generate automated test cases involving Xen.
  627: 
  628: In 2015-01, the following combination was reported to mostly work:
  629: 
  630:         host OS: NetBSD/amd64 6.1.4
  631:         qemu: 2.2.0 from pkgsrc
  632:         Xen kernel: xenkernel42-4.2.5nb1 from pkgsrc
  633:         dom0 kernel: NetBSD/amd64 6.1.5
  634:         Xen tools: xentools42-4.2.5 from pkgsrc
  635: 
  636: See [PR 47720](http://gnats.netbsd.org/47720) for a problem with dom0
  637: shutdown.
  638: 
  639: Unprivileged domains (domU)
  640: ===========================
  641: 
  642: This section describes general concepts about domUs.  It does not
  643: address specific domU operating systems or how to install them.  The
  644: config files for domUs are typically in /usr/pkg/etc/xen, and are
  645: typically named so that the file name, domU name and the domU's host
  646: name match.
  647: 
  648: The domU is provided with CPU and memory by Xen, configured by the
  649: dom0.  The domU is provided with disk and network by the dom0,
  650: mediated by Xen, and configured in the dom0.
  651: 
  652: Entropy in domUs can be an issue; physical disks and network are on
  653: the dom0.  NetBSD's /dev/random system works, but is often challenged.
  654: 
  655: Config files
  656: ------------
  657: 
  658: There is no good order to present config files and the concepts
  659: surrounding what is being configured.  We first show an example config
  660: file, and then in the various sections give details.
  661: 
  662: See (at least in xentools41) /usr/pkg/share/examples/xen/xmexample*,
  663: for a large number of well-commented examples, mostly for running
  664: GNU/Linux.
  665: 
  666: The following is an example minimal domain configuration file
  667: "/usr/pkg/etc/xen/foo".  It is (with only a name change) an actual
  668: known working config file on Xen 4.1 (NetBSD 5 amd64 dom0 and NetBSD 5
  669: i386 domU).  The domU serves as a network file server.
  670: 
  671:         # -*- mode: python; -*-
  672: 
  673:         kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
  674:         memory = 1024
  675:         vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
  676:         disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
  677:                  'file:/n0/xen/foo-wd1,0x1,w' ]
  678: 
  679: The domain will have the same name as the file.  The kernel has the
  680: host/domU name in it, so that on the dom0 one can update the various
  681: domUs independently.  The vif line causes an interface to be provided,
  682: with a specific mac address (do not reuse MAC addresses!), in bridge
  683: mode.  Two disks are provided, and they are both writable; the bits
  684: are stored in files and Xen attaches them to a vnd(4) device in the
  685: dom0 on domain creation.  The system treats xbd0 as the boot device
  686: without needing explicit configuration.
  687: 
  688: By default xm looks for domain config files in /usr/pkg/etc/xen.  Note
  689: that "xm create" takes the name of a config file, while other commands
  690: take the name of a domain.  To create the domain, connect to the
  691: console, create the domain while attaching the console, shutdown the
  692: domain, and see if it has finished stopping, do (or xl with Xen >=
  693: 4.2):
  694: 
  695:         xm create foo
  696:         xm console foo
  697:         xm create -c foo
  698:         xm shutdown foo
  699:         xm list
  700: 
  701: Typing ^] will exit the console session.  Shutting down a domain is
  702: equivalent to pushing the power button; a NetBSD domU will receive a
  703: power-press event and do a clean shutdown.  Shutting down the dom0
  704: will trigger controlled shutdowns of all configured domUs.
  705: 
  706: domU kernels
  707: ------------
  708: 
  709: On a physical computer, the BIOS reads sector 0, and a chain of boot
  710: loaders finds and loads a kernel.  Normally this comes from the root
  711: file system.  With Xen domUs, the process is totally different.  The
  712: normal path is for the domU kernel to be a file in the dom0's
  713: file system.  At the request of the dom0, Xen loads that kernel into a
  714: new domU instance and starts execution.  While domU kernels can be
  715: anyplace, reasonable places to store domU kernels on the dom0 are in /
  716: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
  717: config files), or in /u0/xen (where the vdisks are).
  718: 
  719: Note that loading the domU kernel from the dom0 implies that boot
  720: blocks, /boot, /boot.cfg, and so on are all ignored in the domU.
  721: See the VPS section near the end for discussion of alternate ways to
  722: obtain domU kernels.
  723: 
  724: CPU and memory
  725: --------------
  726: 
  727: A domain is provided with some number of vcpus, less than the number
  728: of CPUs seen by the hypervisor.  (For a dom0, this is controlled by
  729: the boot argument "dom0_max_vcpus=1".)  For a domU, it is controlled
  730: from the config file by the "vcpus = N" directive.
  731: 
  732: A domain is provided with memory; this is controlled in the config
  733: file by "memory = N" (in megabytes).  In the straightforward case, the
  734: sum of the the memory allocated to the dom0 and all domUs must be less
  735: than the available memory.
  736: 
  737: Xen also provides a "balloon" driver, which can be used to let domains
  738: use more memory temporarily.  TODO: Explain better, and explain how
  739: well it works with NetBSD.
  740: 
  741: Virtual disks
  742: -------------
  743: 
  744: With the file/vnd style, typically one creates a directory,
  745: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
  746: domUs.  Then, for each domU disk, one writes zeros to a file that then
  747: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
  748: for the first virtual disk for the domU called foo.  Writing zeros to
  749: the file serves two purposes.  One is that preallocating the contents
  750: improves performance.  The other is that vnd on sparse files has
  751: failed to work.  TODO: give working/notworking NetBSD versions for
  752: sparse vnd and gnats reference.  Note that the use of file/vnd for Xen
  753: is not really different than creating a file-backed virtual disk for
  754: some other purpose, except that xentools handles the vnconfig
  755: commands.  To create an empty 4G virtual disk, simply do
  756: 
  757:         dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
  758: 
  759: Do not use qemu-img-xen, because this will create sparse file.  There
  760: have been recent (2015) reports of sparse vnd(4) devices causing
  761: lockups, but there is apparently no PR.
  762: 
  763: With the lvm style, one creates logical devices.  They are then used
  764: similarly to vnds.  TODO: Add an example with lvm.
  765: 
  766: In domU config files, the disks are defined as a sequence of 3-tuples.
  767: The first element is "method:/path/to/disk".  Common methods are
  768: "file:" for file-backed vnd. and "phy:" for something that is already
  769: a (TODO: character or block) device.
  770: 
  771: The second element is an artifact of how virtual disks are passed to
  772: Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
  773: are given a device name to associate with the disk, and values like
  774: "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
  775: as xbd0, the second as xbd1, and so on.  However, xm/xl demand a
  776: second argument.  The name given is converted to a major/minor by
  777: calling stat(2) on the name in /dev and this is passed to the domU.
  778: In the general case, the dom0 and domU can be different operating
  779: systems, and it is an unwarranted assumption that they have consistent
  780: numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
  781: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
  782: for the second works fine and avoids this issue.  For a GNU/Linux
  783: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
  784: /dev/hda1.
  785: 
  786: The third element is "w" for writable disks, and "r" for read-only
  787: disks.
  788: 
  789: Note that NetBSD by default creates only vnd[0123].  If you need more
  790: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
  791: dom0.
  792: 
  793: Note that NetBSD by default creates only xbd[0123].  If you need more
  794: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
  795: 
  796: Virtual Networking
  797: ------------------
  798: 
  799: Xen provides virtual Ethernets, each of which connects the dom0 and a
  800: domU.  For each virtual network, there is an interface "xvifN.M" in
  801: the dom0, and in domU index N, a matching interface xennetM (NetBSD
  802: name).  The interfaces behave as if there is an Ethernet with two
  803: adapters connected.  From this primitive, one can construct various
  804: configurations.  We focus on two common and useful cases for which
  805: there are existing scripts: bridging and NAT.
  806: 
  807: With bridging (in the example above), the domU perceives itself to be
  808: on the same network as the dom0.  For server virtualization, this is
  809: usually best.  Bridging is accomplished by creating a bridge(4) device
  810: and adding the dom0's physical interface and the various xvifN.0
  811: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
  812: config file.  The bridge must be set up already in the dom0; an
  813: example /etc/ifconfig.bridge0 is:
  814: 
  815:         create
  816:         up
  817:         !brconfig bridge0 add wm0
  818: 
  819: With NAT, the domU perceives itself to be behind a NAT running on the
  820: dom0.  This is often appropriate when running Xen on a workstation.
  821: TODO: NAT appears to be configured by "vif = [ '' ]".
  822: 
  823: The MAC address specified is the one used for the interface in the new
  824: domain.  The interface in dom0 will use this address XOR'd with
  825: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
  826: 
  827: Sizing domains
  828: --------------
  829: 
  830: Modern x86 hardware has vast amounts of resources.  However, many
  831: virtual servers can function just fine on far less.  A system with
  832: 256M of RAM and a 4G disk can be a reasonable choice.  Note that it is
  833: far easier to adjust virtual resources than physical ones.  For
  834: memory, it's just a config file edit and a reboot.  For disk, one can
  835: create a new file and vnconfig it (or lvm), and then dump/restore,
  836: just like updating physical disks, but without having to be there and
  837: without those pesky connectors.
  838: 
  839: Starting domains automatically
  840: ------------------------------
  841: 
  842: To start domains foo at bar at boot and shut them down cleanly on dom0
  843: shutdown, in rc.conf add:
  844: 
  845:         xendomains="foo bar"
  846: 
  847: Note that earlier versions of the xentools41 xendomains rc.d script
  848: used xl, when one should use xm with 4.1.
  849: 
  850: Creating specific unprivileged domains (domU)
  851: =============================================
  852: 
  853: Creating domUs is almost entirely independent of operating system.  We
  854: have already presented the basics of config files.  Note that you must
  855: have already completed the dom0 setup so that "xl list" (or "xm list")
  856: works.
  857: 
  858: Creating an unprivileged NetBSD domain (domU)
  859: ---------------------------------------------
  860: 
  861: See the earlier config file, and adjust memory.  Decide on how much
  862: storage you will provide, and prepare it (file or lvm).
  863: 
  864: While the kernel will be obtained from the dom0 file system, the same
  865: file should be present in the domU as /netbsd so that tools like
  866: savecore(8) can work.   (This is helpful but not necessary.)
  867: 
  868: The kernel must be specifically for Xen and for use as a domU.  The
  869: i386 and amd64 provide the following kernels:
  870: 
  871:         i386 XEN3_DOMU
  872:         i386 XEN3PAE_DOMU
  873:         amd64 XEN3_DOMU
  874: 
  875: Unless using Xen 3.1 (and you shouldn't) with i386-mode Xen, you must
  876: use the PAE version of the i386 kernel.
  877: 
  878: This will boot NetBSD, but this is not that useful if the disk is
  879: empty.  One approach is to unpack sets onto the disk outside of xen
  880: (by mounting it, just as you would prepare a physical disk for a
  881: system you can't run the installer on).
  882: 
  883: A second approach is to run an INSTALL kernel, which has a miniroot
  884: and can load sets from the network.  To do this, copy the INSTALL
  885: kernel to / and change the kernel line in the config file to:
  886: 
  887:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  888: 
  889: Then, start the domain as "xl create -c configname".
  890: 
  891: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
  892: line should be used in the config file.
  893: 
  894:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  895: 
  896: After booting the domain, the option to install via CDROM may be
  897: selected.  The CDROM device should be changed to `xbd1d`.
  898: 
  899: Once done installing, "halt -p" the new domain (don't reboot or halt,
  900: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
  901: config file), switch the config file back to the XEN3_DOMU kernel,
  902: and start the new domain again. Now it should be able to use "root on
  903: xbd0a" and you should have a, functional NetBSD domU.
  904: 
  905: TODO: check if this is still accurate.
  906: When the new domain is booting you'll see some warnings about *wscons*
  907: and the pseudo-terminals. These can be fixed by editing the files
  908: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  909: `/etc/ttys`, except *console*, like this:
  910: 
  911:     console "/usr/libexec/getty Pc"         vt100   on secure
  912:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  913:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  914:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  915:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  916: 
  917: Finally, all screens must be commented out from `/etc/wscons.conf`.
  918: 
  919: It is also desirable to add
  920: 
  921:         powerd=YES
  922: 
  923: in rc.conf. This way, the domain will be properly shut down if
  924: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
  925: 
  926: It is not strictly necessary to have a kernel (as /netbsd) in the domU
  927: file system.  However, various programs (e.g. netstat) will use that
  928: kernel to look up symbols to read from kernel virtual memory.  If
  929: /netbsd is not the running kernel, those lookups will fail.  (This is
  930: not really a Xen-specific issue, but because the domU kernel is
  931: obtained from the dom0, it is far more likely to be out of sync or
  932: missing with Xen.)
  933: 
  934: Creating an unprivileged Linux domain (domU)
  935: --------------------------------------------
  936: 
  937: Creating unprivileged Linux domains isn't much different from
  938: unprivileged NetBSD domains, but there are some details to know.
  939: 
  940: First, the second parameter passed to the disk declaration (the '0x1' in
  941: the example below)
  942: 
  943:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  944: 
  945: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  946: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
  947: So, hda1 which has major 3 and minor 1 on a Linux system will have
  948: device number 0x301.  Alternatively, devices names can be used (hda,
  949: hdb, ...)  as xentools has a table to map these names to devices
  950: numbers.  To export a partition to a Linux guest we can use:
  951: 
  952:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
  953:         root = "/dev/hda1 ro"
  954: 
  955: and it will appear as /dev/hda on the Linux system, and be used as root
  956: partition.
  957: 
  958: To install the Linux system on the partition to be exported to the
  959: guest domain, the following method can be used: install
  960: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
  961: that will be the root partition of your Linux domain, and mount it.
  962: Then copy the files from a working Linux system, make adjustments in
  963: `/etc` (fstab, network config).  It should also be possible to extract
  964: binary packages such as .rpm or .deb directly to the mounted partition
  965: using the appropriate tool, possibly running under NetBSD's Linux
  966: emulation.  Once the file system has been populated, umount it.  If
  967: desirable, the file system can be converted to ext3 using tune2fs -j.
  968: It should now be possible to boot the Linux guest domain, using one of
  969: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
  970: 
  971: To get the Linux console right, you need to add:
  972: 
  973:     extra = "xencons=tty1"
  974: 
  975: to your configuration since not all Linux distributions auto-attach a
  976: tty to the xen console.
  977: 
  978: Creating an unprivileged Solaris domain (domU)
  979: ----------------------------------------------
  980: 
  981: See possibly outdated
  982: [Solaris domU instructions](/ports/xen/howto-solaris/).
  983: 
  984: 
  985: PCI passthrough: Using PCI devices in guest domains
  986: ---------------------------------------------------
  987: 
  988: The dom0 can give other domains access to selected PCI
  989: devices. This can allow, for example, a non-privileged domain to have
  990: access to a physical network interface or disk controller.  However,
  991: keep in mind that giving a domain access to a PCI device most likely
  992: will give the domain read/write access to the whole physical memory,
  993: as PCs don't have an IOMMU to restrict memory access to DMA-capable
  994: device.  Also, it's not possible to export ISA devices to non-dom0
  995: domains, which means that the primary VGA adapter can't be exported.
  996: A guest domain trying to access the VGA registers will panic.
  997: 
  998: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
  999: not been ported to later versions at this time.
 1000: 
 1001: For a PCI device to be exported to a domU, is has to be attached to
 1002: the "pciback" driver in dom0.  Devices passed to the dom0 via the
 1003: pciback.hide boot parameter will attach to "pciback" instead of the
 1004: usual driver.  The list of devices is specified as "(bus:dev.func)",
 1005: where bus and dev are 2-digit hexadecimal numbers, and func a
 1006: single-digit number:
 1007: 
 1008:         pciback.hide=(00:0a.0)(00:06.0)
 1009: 
 1010: pciback devices should show up in the dom0's boot messages, and the
 1011: devices should be listed in the `/kern/xen/pci` directory.
 1012: 
 1013: PCI devices to be exported to a domU are listed in the "pci" array of
 1014: the domU's config file, with the format "0000:bus:dev.func".
 1015: 
 1016:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
 1017: 
 1018: In the domU an "xpci" device will show up, to which one or more pci
 1019: buses will attach.  Then the PCI drivers will attach to PCI buses as
 1020: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
 1021: or any PCI drivers built in by default; you have to build your own
 1022: kernel to use PCI devices in a domU.  Here's a kernel config example;
 1023: note that only the "xpci" lines are unusual.
 1024: 
 1025:         include         "arch/i386/conf/XEN3_DOMU"
 1026: 
 1027:         # Add support for PCI buses to the XEN3_DOMU kernel
 1028:         xpci* at xenbus ?
 1029:         pci* at xpci ?
 1030: 
 1031:         # PCI USB controllers
 1032:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
 1033: 
 1034:         # USB bus support
 1035:         usb*    at uhci?
 1036: 
 1037:         # USB Hubs
 1038:         uhub*   at usb?
 1039:         uhub*   at uhub? port ? configuration ? interface ?
 1040: 
 1041:         # USB Mass Storage
 1042:         umass*  at uhub? port ? configuration ? interface ?
 1043:         wd*     at umass?
 1044:         # SCSI controllers
 1045:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
 1046: 
 1047:         # SCSI bus support (for both ahc and umass)
 1048:         scsibus* at scsi?
 1049: 
 1050:         # SCSI devices
 1051:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
 1052:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
 1053: 
 1054: 
 1055: NetBSD as a domU in a VPS
 1056: =========================
 1057: 
 1058: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
 1059: hardware.  This section explains how to deal with Xen in a domU as a
 1060: virtual private server where you do not control or have access to the
 1061: dom0.  This is not intended to be an exhaustive list of VPS providers;
 1062: only a few are mentioned that specifically support NetBSD.
 1063: 
 1064: VPS operators provide varying degrees of access and mechanisms for
 1065: configuration.  The big issue is usually how one controls which kernel
 1066: is booted, because the kernel is nominally in the dom0 file system (to
 1067: which VPS users do not normally have access).  A second issue is how
 1068: to install NetBSD.
 1069: A VPS user may want to compile a kernel for security updates, to run
 1070: npf, run IPsec, or any other reason why someone would want to change
 1071: their kernel.
 1072: 
 1073: One approach is to have an administrative interface to upload a kernel,
 1074: or to select from a prepopulated list.  Other approaches are pygrub
 1075: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
 1076: kernel from the domU file system.  This is closer to a regular physical
 1077: computer, where someone who controls a machine can replace the kernel.
 1078: 
 1079: A second issue is multiple CPUs.  With NetBSD 6, domUs support
 1080: multiple vcpus, and it is typical for VPS providers to enable multiple
 1081: CPUs for NetBSD domUs.
 1082: 
 1083: pygrub
 1084: -------
 1085: 
 1086: pygrub runs in the dom0 and looks into the domU file system.  This
 1087: implies that the domU must have a kernel in a file system in a format
 1088: known to pygrub.  As of 2014, pygrub seems to be of mostly historical
 1089: interest.
 1090: 
 1091: pvgrub
 1092: ------
 1093: 
 1094: pvgrub is a version of grub that uses PV operations instead of BIOS
 1095: calls.  It is booted from the dom0 as the domU kernel, and then reads
 1096: /grub/menu.lst and loads a kernel from the domU file system.
 1097: 
 1098: [Panix](http://www.panix.com/) lets users use pvgrub.  Panix reports
 1099: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
 1100: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
 1101: page](http://www.panix.com/v-colo/grub.html), which describes only
 1102: Linux but should be updated to cover NetBSD :-).
 1103: 
 1104: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
 1105: their own kernel.  See then [prgmr.com NetBSD
 1106: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
 1107: (which is in need of updating).
 1108: 
 1109: It appears that [grub's FFS
 1110: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
 1111: does not support all aspects of modern FFS, but there are also reports
 1112: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
 1113: partition for the kernel with the intent that grub can understand it,
 1114: which leads to /netbsd not being the actual kernel.  One must remember
 1115: to update the special boot partition.
 1116: 
 1117: Amazon
 1118: ------
 1119: 
 1120: See the [Amazon EC2 page](../amazon_ec2/).
 1121: 
 1122: Using npf
 1123: ---------
 1124: 
 1125: In standard kernels, npf is a module, and thus cannot be loaded in a
 1126: DOMU kernel.
 1127: 
 1128: TODO: Explain how to compile npf into a custom kernel, answering (but
 1129: note that the problem was caused by not booting the right kernel)
 1130: [this email to
 1131: netbsd-users](http://mail-index.netbsd.org/netbsd-users/2014/12/26/msg015576.html).
 1132: 
 1133: TODO items for improving NetBSD/xen
 1134: ===================================
 1135: 
 1136: * Make the NetBSD dom0 kernel work with SMP.
 1137: * Test the Xen 4.5 packages adequately to be able to recommend them as
 1138:   the standard approach.
 1139: * Get PCI passthrough working on Xen 4.5
 1140: * Get pvgrub into pkgsrc, either via xentools or separately.
 1141: * grub
 1142:   * Check/add support to pkgsrc grub2 for UFS2 and arbitrary
 1143:     fragsize/blocksize (UFS2 support may be present; the point is to
 1144:     make it so that with any UFS1/UFS2 file system setup that works
 1145:     with NetBSD grub will also work).
 1146:     See [pkg/40258](http://gnats.netbsd.org/40258).
 1147:   * Push patches upstream.
 1148:   * Get UFS2 patches into pvgrub.
 1149: * Add support for PV ops to a version of /boot, and make it usable as
 1150:   a kernel in Xen, similar to pvgrub.
 1151: * Solve somehow the issue with modules for GENERIC not being loadable
 1152:   in a Xen dom0 or domU kernel.
 1153: 
 1154: Random pointers
 1155: ===============
 1156: 
 1157: This section contains links from elsewhere not yet integrated into the
 1158: HOWTO, and other guides.
 1159: 
 1160: * http://www.lumbercartel.ca/library/xen/
 1161: * http://pbraun.nethence.com/doc/sysutils/xen_netbsd_dom0.html
 1162: * https://gmplib.org/~tege/xen.html

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