File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.62: download - view: text, annotated - select for diffs
Mon Dec 29 00:13:15 2014 UTC (4 years, 11 months ago) by gdt
Branches: MAIN
CVS tags: HEAD
update what builds and what doesn't

    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 amd64 machine will work with Xen and PV guests.  In
   31: theory i386 computers without amd64 support can be used for Xen <=
   32: 4.2, but we have no recent reports of this working (this is a hint).
   33: For HVM guests, the VT or VMX cpu feature (Intel) or SVM/HVM/VT
   34: (amd64) is needed; "cpuctl identify 0" will show this.  TODO: Clean up
   35: and check the above features.
   36: 
   37: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
   38: The dom0 can start one or more domUs.  (Booting is explained in detail
   39: in the dom0 section.)
   40: 
   41: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
   42: and that Xen kernels and tools are available in pkgsrc.  This HOWTO
   43: attempts to address both the case of running a NetBSD dom0 on hardware
   44: and running domUs under it (NetBSD and other), and also running NetBSD
   45: as a domU in a VPS.
   46: 
   47: Some versions of Xen support "PCI passthrough", which means that
   48: specific PCI devices can be made available to a specific domU instead
   49: of the dom0.  This can be useful to let a domU run X11, or access some
   50: network interface or other peripheral.
   51: 
   52: NetBSD used to support Xen2; this has been removed.
   53: 
   54: Prerequisites
   55: -------------
   56: 
   57: Installing NetBSD/Xen is not extremely difficult, but it is more
   58: complex than a normal installation of NetBSD.
   59: In general, this HOWTO is occasionally overly restrictive about how
   60: things must be done, guiding the reader to stay on the established
   61: path when there are no known good reasons to stray.
   62: 
   63: This HOWTO presumes a basic familiarity with the Xen system
   64: architecture.  This HOWTO presumes familiarity with installing NetBSD
   65: on i386/amd64 hardware and installing software from pkgsrc.
   66: See also the [Xen website](http://www.xenproject.org/).
   67: 
   68: Versions of Xen and NetBSD
   69: ==========================
   70: 
   71: Most of the installation concepts and instructions are independent
   72: of Xen version and NetBSD version.  This section gives advice on
   73: which version to choose.  Versions not in pkgsrc and older unsupported
   74: versions of NetBSD are intentionally ignored.
   75: 
   76: Xen
   77: ---
   78: 
   79: In NetBSD, xen is provided in pkgsrc, via matching pairs of packages
   80: xenkernel and xentools.  We will refer only to the kernel versions,
   81: but note that both packages must be installed together and must have
   82: matching versions.
   83: 
   84: xenkernel3 and xenkernel33 provide Xen 3.1 and 3.3.  These no longer
   85: receive security patches and should not be used.  Xen 3.1 supports PCI
   86: passthrough.  Xen 3.1 supports non-PAE on i386.
   87: 
   88: xenkernel41 provides Xen 4.1.  This is no longer maintained by Xen,
   89: but as of 2014-12 receives backported security patches.  It is a
   90: reasonable although trailing-edge choice.
   91: 
   92: xenkernel42 provides Xen 4.2.  This is maintained by Xen, but old as
   93: of 2014-12.
   94: 
   95: Ideally newer versions of Xen will be added to pkgsrc.
   96: 
   97: Note that NetBSD support is called XEN3.  It works with 3.1 through
   98: 4.2 because the hypercall interface has been stable.
   99: 
  100: Xen command program
  101: -------------------
  102: 
  103: Early Xen used a program called "xm" to manipulate the system from the
  104: dom0.  Starting in 4.1, a replacement program with similar behavior
  105: called "xl" is provided.  In 4.2 and later, "xl" is preferred.  4.4 is
  106: the last version that has "xm".
  107: 
  108: NetBSD
  109: ------
  110: 
  111: The netbsd-5, netbsd-6, netbsd-7, and -current branches are all
  112: reasonable choices, with more or less the same considerations for
  113: non-Xen use.  Therefore, netbsd-6 is recommended as the stable version
  114: of the most recent release for production use.  For those wanting to
  115: learn Xen or without production stability concerns, netbsd-7 is likely
  116: most appropriate.
  117: 
  118: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
  119: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
  120: need SMP; the lack of support is really a problem when using a dom0 as
  121: a normal computer.)
  122: 
  123: Architecture
  124: ------------
  125: 
  126: Xen itself can run on i386 or amd64 machines.  (Practically, almost
  127: any computer where one would want to run Xen supports amd64.)  If
  128: using an i386 NetBSD kernel for the dom0, PAE is required (PAE
  129: versions are built by default).  While i386 dom0 works fine, amd64 is
  130: recommended as more normal.
  131: 
  132: Xen 4.2 is the last version to support i386 as a host.  TODO: Clarify
  133: if this is about the CPU having to be amd64, or about the dom0 kernel
  134: having to be amd64.
  135: 
  136: One can then run i386 domUs and amd64 domUs, in any combination.  If
  137: running an i386 NetBSD kernel as a domU, the PAE version is required.
  138: (Note that emacs (at least) fails if run on i386 with PAE when built
  139: without, and vice versa, presumably due to bugs in the undump code.)
  140: 
  141: Recommendation
  142: --------------
  143: 
  144: Therefore, this HOWTO recommends running xenkernel42 (and xentools42),
  145: xl, the NetBSD 6 stable branch, and to use an amd64 kernel as the
  146: dom0.  Either the i386 or amd64 of NetBSD may be used as domUs.
  147: 
  148: Build problems
  149: --------------
  150: 
  151: Ideally, all versions of Xen in pkgsrc would build on all versions of
  152: NetBSD on both i386 and amd64.  However, that isn't the case.  Besides
  153: aging code and aging compilers, qemu (included in xentools for HVM
  154: support) is difficult to build.  The following are known to work or FAIL:
  155: 
  156:         xenkernel3 netbsd-5 amd64
  157:         xenkernel33 netbsd-5 amd64
  158:         xenkernel41 netbsd-5 amd64
  159:         xentools41 netbsd-5 amd64
  160:         xenkernel42 netbsd-5 amd64
  161: 
  162:         xenkernel3 netbsd-6 i386 FAIL
  163:         xentools3 netbsd-6 i386
  164: 	xentools3-hvm netbsd-6 i386 FAIL (dependencies fail)
  165: 	xenkernel33 netbsd-6 i386
  166: 	xentools33 netbsd-6 i386
  167:         xenkernel41 netbsd-6 i386
  168:         xentools41 netbsd-6 i386
  169: 	xenkernel42 netbsd-6 i386
  170:         xentools42 netbsd-6 i386 FAIL
  171: 
  172: NetBSD as a dom0
  173: ================
  174: 
  175: NetBSD can be used as a dom0 and works very well.  The following
  176: sections address installation, updating NetBSD, and updating Xen.
  177: Note that it doesn't make sense to talk about installing a dom0 OS
  178: without also installing Xen itself.  We first address installing
  179: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
  180: NetBSD install to a dom0 install by just changing the kernel and boot
  181: configuration.
  182: 
  183: For experimenting with Xen, a machine with as little as 1G of RAM and
  184: 100G of disk can work.  For running many domUs in productions, far
  185: more will be needed.
  186: 
  187: Styles of dom0 operation
  188: ------------------------
  189: 
  190: There are two basic ways to use Xen.  The traditional method is for
  191: the dom0 to do absolutely nothing other than providing support to some
  192: number of domUs.  Such a system was probably installed for the sole
  193: purpose of hosting domUs, and sits in a server room on a UPS.
  194: 
  195: The other way is to put Xen under a normal-usage computer, so that the
  196: dom0 is what the computer would have been without Xen, perhaps a
  197: desktop or laptop.  Then, one can run domUs at will.  Purists will
  198: deride this as less secure than the previous approach, and for a
  199: computer whose purpose is to run domUs, they are right.  But Xen and a
  200: dom0 (without domUs) is not meaingfully less secure than the same
  201: things running without Xen.  One can boot Xen or boot regular NetBSD
  202: alternately with little problems, simply refraining from starting the
  203: Xen daemons when not running Xen.
  204: 
  205: Note that NetBSD as dom0 does not support multiple CPUs.  This will
  206: limit the performance of the Xen/dom0 workstation approach.  In theory
  207: the only issue is that the "backend drivers" are not yet MPSAFE:
  208:   http://mail-index.netbsd.org/netbsd-users/2014/08/29/msg015195.html
  209: 
  210: Installation of NetBSD
  211: ----------------------
  212: 
  213: First,
  214: [install NetBSD/amd64](/guide/inst/)
  215: just as you would if you were not using Xen.
  216: However, the partitioning approach is very important.
  217: 
  218: If you want to use RAIDframe for the dom0, there are no special issues
  219: for Xen.  Typically one provides RAID storage for the dom0, and the
  220: domU systems are unaware of RAID.  The 2nd-stage loader bootxx_* skips
  221: over a RAID1 header to find /boot from a filesystem within a RAID
  222: partition; this is no different when booting Xen.
  223: 
  224: There are 4 styles of providing backing storage for the virtual disks
  225: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN,
  226: 
  227: With raw partitions, one has a disklabel (or gpt) partition sized for
  228: each virtual disk to be used by the domU.  (If you are able to predict
  229: how domU usage will evolve, please add an explanation to the HOWTO.
  230: Seriously, needs tend to change over time.)
  231: 
  232: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
  233: for domU disks.  This is almost as efficient as raw disk partitions
  234: and more flexible.  Hence raw disk partitions should typically not
  235: be used.
  236: 
  237: One can use files in the dom0 filesystem, typically created by dd'ing
  238: /dev/zero to create a specific size.  This is somewhat less efficient,
  239: but very convenient, as one can cp the files for backup, or move them
  240: between dom0 hosts.
  241: 
  242: Finally, in theory one can place the files backing the domU disks in a
  243: SAN.  (This is an invitation for someone who has done this to add a
  244: HOWTO page.)
  245: 
  246: Installation of Xen
  247: -------------------
  248: 
  249: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
  250: pkgsrc (or another matching pair).
  251: See [the pkgsrc
  252: documentation](http://www.NetBSD.org/docs/pkgsrc/) for help with pkgsrc.
  253: 
  254: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm.  More
  255: recent versions have HVM support integrated in the main xentools
  256: package.  It is entirely reasonable to run only PV guests.
  257: 
  258: Next you need to install the selected Xen kernel itself, which is
  259: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz".  Copy it to /.
  260: For debugging, one may copy xen-debug.gz; this is conceptually similar
  261: to DIAGNOSTIC and DEBUG in NetBSD.  xen-debug.gz is basically only
  262: useful with a serial console.  Then, place a NetBSD XEN3_DOM0 kernel
  263: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
  264: of a NetBSD build.  Both xen and NetBSD may be left compressed.  (If
  265: using i386, use releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz.)
  266: 
  267: With Xen as the kernel, you must provide a dom0 NetBSD kernel to be
  268: used as a module; place this in /.  Suitable kernels are provided in
  269: releasedir/binary/kernel:
  270: 
  271:         i386 XEN3_DOM0
  272:         i386 XEN3PAE_DOM0
  273: 	amd64 XEN3_DOM0
  274: 
  275: The first one is only for use with Xen 3.1 and i386-mode Xen (and you
  276: should not do this).  Current Xen always uses PAE on i386, but you
  277: should generally use amd64 for the dom0.  In a dom0 kernel, kernfs is
  278: mandatory for xend to comunicate with the kernel, so ensure that /kern
  279: is in fstab.  TODO: Say this is default, or file a PR and give a
  280: reference.
  281: 
  282: Because you already installed NetBSD, you have a working boot setup
  283: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
  284: beginning of your root filesystem, /boot present, and likely
  285: /boot.cfg.  (If not, fix before continuing!)
  286: 
  287: See boot.cfg(5) for an example.  The basic line is
  288: 
  289:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
  290: 
  291: which specifies that the dom0 should have 256M, leaving the rest to be
  292: allocated for domUs.  In an attempt to add performance, one can also
  293: add
  294: 
  295:         dom0_max_vcpus=1 dom0_vcpus_pin
  296: 
  297: to force only one vcpu to be provided (since NetBSD dom0 can't use
  298: more) and to pin that vcpu to a physical cpu.  TODO: benchmark this.
  299: 
  300: As with non-Xen systems, you should have a line to boot /netbsd (a
  301: kernel that works without Xen) and fallback versions of the non-Xen
  302: kernel, Xen, and the dom0 kernel.
  303: 
  304: Using grub (historic)
  305: ---------------------
  306: 
  307: Before NetBSD's native bootloader could support Xen, the use of
  308: grub was recommended.  If necessary, see the
  309: [old grub information](/ports/xen/howto-grub/).
  310: 
  311: The [HowTo on Installing into
  312: RAID-1](http://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
  313: explains how to set up booting a dom0 with Xen using grub with
  314: NetBSD's RAIDframe.  (This is obsolete with the use of NetBSD's native
  315: boot.)
  316: 
  317: Configuring Xen
  318: ---------------
  319: 
  320: Xen logs will be in /var/log/xen.
  321: 
  322: Now, you have a system that will boot Xen and the dom0 kernel, and
  323: just run the dom0 kernel.  There will be no domUs, and none can be
  324: started because you still have to configure the dom0 tools.  The
  325: daemons which should be run vary with Xen version and with whether one
  326: is using xm or xl.  Note that xend is for supporting "xm", and should
  327: only be used if you plan on using "xm".  Do NOT enable xend if you
  328: plan on using "xl" as it will cause problems.
  329: 
  330: The installation of NetBSD should already have created devices for xen
  331: (xencons, xenevt), but if they are not present, create them:
  332: 
  333:         cd /dev && sh MAKEDEV xen
  334: 
  335: TODO: Give 3.1 advice (or remove it from pkgsrc).
  336: 
  337: For 3.3 (and thus xm), add to rc.conf (but note that you should have
  338: installed 4.1 or 4.2):
  339: 
  340:         xend=YES
  341:         xenbackendd=YES
  342: 
  343: For 4.1 (and thus xm; xl is believed not to work well), add to rc.conf:
  344: 
  345:         xencommons=YES
  346:         xend=YES
  347: 
  348: (If you are using xentools41 from before 2014-12-26, change
  349: rc.d/xendomains to use xm rather than xl.)
  350: 
  351: For 4.2 with xm, add to rc.conf
  352: 
  353:         xencommons=YES
  354:         xend=YES
  355: 
  356: For 4.2 with xl (preferred), add to rc.conf:
  357: 
  358:         xencommons=YES
  359:         TODO: explain if there is a xend replacement
  360: 
  361: TODO: Recommend for/against xen-watchdog.
  362: 
  363: After you have configured the daemons and either started them (in the
  364: order given) or rebooted, run the following (or use xl) to inspect
  365: Xen's boot messages, available resources, and running domains:
  366: 
  367:         # xm dmesg
  368: 	[xen's boot info]
  369:         # xm info
  370: 	[available memory, etc.]
  371:         # xm list
  372:         Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  373:         Domain-0           0       64    0  r----     58.1
  374: 
  375: anita (for testing NetBSD)
  376: --------------------------
  377: 
  378: With the setup so far, one should be able to run anita (see
  379: pkgsrc/sysutils/py-anita) to test NetBSD releases, by doing (as root,
  380: because anita must create a domU):
  381: 
  382:         anita --vmm=xm test file:///usr/obj/i386/
  383: 
  384: Alternatively, one can use --vmm=xl to use xl-based domU creation instead.
  385: TODO: check this.
  386: 
  387: Xen-specific NetBSD issues
  388: --------------------------
  389: 
  390: There are (at least) two additional things different about NetBSD as a
  391: dom0 kernel compared to hardware.
  392: 
  393: One is that modules are not usable in DOM0 kernels, so one must
  394: compile in what's needed.  It's not really that modules cannot work,
  395: but that modules must be built for XEN3_DOM0 because some of the
  396: defines change and the normal module builds don't do this.  Basically,
  397: enabling Xen changes the kernel ABI, and the module build system
  398: doesn't cope with this.
  399: 
  400: The other difference is that XEN3_DOM0 does not have exactly the same
  401: options as GENERIC.  While it is debatable whether or not this is a
  402: bug, users should be aware of this and can simply add missing config
  403: items if desired.
  404: 
  405: Updating NetBSD in a dom0
  406: -------------------------
  407: 
  408: This is just like updating NetBSD on bare hardware, assuming the new
  409: version supports the version of Xen you are running.  Generally, one
  410: replaces the kernel and reboots, and then overlays userland binaries
  411: and adjusts /etc.
  412: 
  413: Note that one must update both the non-Xen kernel typically used for
  414: rescue purposes and the DOM0 kernel used with Xen.
  415: 
  416: Converting from grub to /boot
  417: -----------------------------
  418: 
  419: These instructions were [TODO: will be] used to convert a system from
  420: grub to /boot.  The system was originally installed in February of
  421: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
  422: over time.  Before these commands, it was running NetBSD 6 i386, Xen
  423: 4.1 and grub, much like the message linked earlier in the grub
  424: section.
  425: 
  426:         # Install mbr bootblocks on both disks. 
  427:         fdisk -i /dev/rwd0d
  428:         fdisk -i /dev/rwd1d
  429:         # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
  430:         installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
  431:         installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
  432:         # Install secondary boot loader
  433:         cp -p /usr/mdec/boot /
  434:         # Create boog.cfg following earlier guidance:
  435:         menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
  436:         menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=256M
  437:         menu=GENERIC:boot
  438:         menu=GENERIC single-user:boot -s
  439:         menu=GENERIC.ok:boot netbsd.ok
  440:         menu=GENERIC.ok single-user:boot netbsd.ok -s
  441:         menu=Drop to boot prompt:prompt
  442:         default=1
  443:         timeout=30
  444: 
  445: TODO: actually do this and fix it if necessary.
  446: 
  447: Updating Xen versions
  448: ---------------------
  449: 
  450: Updating Xen is conceptually not difficult, but can run into all the
  451: issues found when installing Xen.  Assuming migration from 4.1 to 4.2,
  452: remove the xenkernel41 and xentools41 packages and install the
  453: xenkernel42 and xentools42 packages.  Copy the 4.2 xen.gz to /.
  454: 
  455: Ensure that the contents of /etc/rc.d/xen* are correct.  Enable the
  456: correct set of daemons.  Ensure that the domU config files are valid
  457: for the new version.
  458: 
  459: 
  460: Unprivileged domains (domU)
  461: ===========================
  462: 
  463: This section describes general concepts about domUs.  It does not
  464: address specific domU operating systems or how to install them.  The
  465: config files for domUs are typically in /usr/pkg/etc/xen, and are
  466: typically named so that the file name, domU name and the domU's host
  467: name match.
  468: 
  469: The domU is provided with cpu and memory by Xen, configured by the
  470: dom0.  The domU is provided with disk and network by the dom0,
  471: mediated by Xen, and configured in the dom0.
  472: 
  473: Entropy in domUs can be an issue; physical disks and network are on
  474: the dom0.  NetBSD's /dev/random system works, but is often challenged.
  475: 
  476: Config files
  477: ------------
  478: 
  479: There is no good order to present config files and the concepts
  480: surrounding what is being configured.  We first show an example config
  481: file, and then in the various sections give details.
  482: 
  483: See (at least in xentools41) /usr/pkg/share/examples/xen/xmexample*,
  484: for a large number of well-commented examples, mostly for running
  485: GNU/Linux.
  486: 
  487: The following is an example minimal domain configuration file
  488: "/usr/pkg/etc/xen/foo".  It is (with only a name change) an actual
  489: known working config file on Xen 4.1 (NetBSD 5 amd64 dom0 and NetBSD 5
  490: i386 domU).  The domU serves as a network file server.
  491: 
  492:         # -*- mode: python; -*-
  493: 
  494:         kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
  495:         memory = 1024
  496:         vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
  497:         disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
  498:                  'file:/n0/xen/foo-wd1,0x1,w' ]
  499: 
  500: The domain will have the same name as the file.  The kernel has the
  501: host/domU name in it, so that on the dom0 one can update the various
  502: domUs independently.  The vif line causes an interface to be provided,
  503: with a specific mac address (do not reuse MAC addresses!), in bridge
  504: mode.  Two disks are provided, and they are both writable; the bits
  505: are stored in files and Xen attaches them to a vnd(4) device in the
  506: dom0 on domain creation.  The system treates xbd0 as the boot device
  507: without needing explicit configuration.
  508: 
  509: By default xm looks for domain config files in /usr/pkg/etc/xen.  Note
  510: that "xm create" takes the name of a config file, while other commands
  511: take the name of a domain.  To create the domain, connect to the
  512: console, create the domain while attaching the console, shutdown the
  513: domain, and see if it has finished stopping, do (or xl with Xen >=
  514: 4.2):
  515: 
  516:         xm create foo
  517:         xm console foo
  518:         xm create -c foo
  519:         xm shutdown foo
  520: 	xm list
  521: 
  522: Typing ^] will exit the console session.  Shutting down a domain is
  523: equivalent to pushing the power button; a NetBSD domU will receive a
  524: power-press event and do a clean shutdown.  Shutting down the dom0
  525: will trigger controlled shutdowns of all configured domUs.
  526: 
  527: domU kernels
  528: ------------
  529: 
  530: On a physical computer, the BIOS reads sector 0, and a chain of boot
  531: loaders finds and loads a kernel.  Normally this comes from the root
  532: filesystem.  With Xen domUs, the process is totally different.  The
  533: normal path is for the domU kernel to be a file in the dom0's
  534: filesystem.  At the request of the dom0, Xen loads that kernel into a
  535: new domU instance and starts execution.  While domU kernels can be
  536: anyplace, reasonable places to store domU kernels on the dom0 are in /
  537: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
  538: config files), or in /u0/xen (where the vdisks are).
  539: 
  540: Note that loading the domU kernel from the dom0 implies that boot
  541: blocks, /boot, /boot.cfg, and so on are all ignored in the domU.
  542: See the VPS section near the end for discussion of alternate ways to
  543: obtain domU kernels.
  544: 
  545: CPU and memory
  546: --------------
  547: 
  548: A domain is provided with some number of vcpus, less than the number
  549: of cpus seen by the hypervisor.  (For a dom0, this is controlled by
  550: the boot argument "dom0_max_vcpus=1".)  For a domU, it is controlled
  551: from the config file by the "vcpus = N" directive.
  552: 
  553: A domain is provided with memory; this is controlled in the config
  554: file by "memory = N" (in megabytes).  In the straightforward case, the
  555: sum of the the memory allocated to the dom0 and all domUs must be less
  556: than the available memory.
  557: 
  558: Xen also provides a "balloon" driver, which can be used to let domains
  559: use more memory temporarily.  TODO: Explain better, and explain how
  560: well it works with NetBSD.
  561: 
  562: Virtual disks
  563: -------------
  564: 
  565: With the file/vnd style, typically one creates a directory,
  566: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
  567: domUs.  Then, for each domU disk, one writes zeros to a file that then
  568: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
  569: for the first virtual disk for the domU called foo.  Writing zeros to
  570: the file serves two purposes.  One is that preallocating the contents
  571: improves performance.  The other is that vnd on sparse files has
  572: failed to work.  TODO: give working/notworking NetBSD versions for
  573: sparse vnd.  Note that the use of file/vnd for Xen is not really
  574: different than creating a file-backed virtual disk for some other
  575: purpose, except that xentools handles the vnconfig commands.  To
  576: create an empty 4G virtual disk, simply do
  577: 
  578:         dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
  579: 
  580: With the lvm style, one creates logical devices.  They are then used
  581: similarly to vnds.  TODO: Add an example with lvm.
  582: 
  583: In domU config files, the disks are defined as a sequence of 3-tuples.
  584: The first element is "method:/path/to/disk".  Common methods are
  585: "file:" for file-backed vnd. and "phy:" for something that is already
  586: a (TODO: character or block) device.
  587: 
  588: The second element is an artifact of how virtual disks are passed to
  589: Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
  590: are given a device name to associate with the disk, and values like
  591: "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
  592: as xbd0, the second as xbd1, and so on.  However, xm/xl demand a
  593: second argument.  The name given is converted to a major/minor by
  594: calling stat(2) on the name in /dev and this is passed to the domU.
  595: In the general case, the dom0 and domU can be different operating
  596: systems, and it is an unwarranted assumption that they have consistent
  597: numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
  598: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
  599: for the second works fine and avoids this issue.  For a GNU/Linux
  600: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
  601: /dev/hda1.
  602: 
  603: The third element is "w" for writable disks, and "r" for read-only
  604: disks.
  605: 
  606: Virtual Networking
  607: ------------------
  608: 
  609: Xen provides virtual ethernets, each of which connects the dom0 and a
  610: domU.  For each virtual network, there is an interface "xvifN.M" in
  611: the dom0, and in domU index N, a matching interface xennetM (NetBSD
  612: name).  The interfaces behave as if there is an Ethernet with two
  613: adaptors connected.  From this primitive, one can construct various
  614: configurations.  We focus on two common and useful cases for which
  615: there are existing scripts: bridging and NAT.
  616: 
  617: With bridging (in the example above), the domU perceives itself to be
  618: on the same network as the dom0.  For server virtualization, this is
  619: usually best.  Bridging is accomplished by creating a bridge(4) device
  620: and adding the dom0's physical interface and the various xvifN.0
  621: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
  622: config file.  The bridge must be set up already in the dom0; an
  623: example /etc/ifconfig.bridge0 is:
  624: 
  625:         create
  626:         up
  627:         !brconfig bridge0 add wm0
  628: 
  629: With NAT, the domU perceives itself to be behind a NAT running on the
  630: dom0.  This is often appropriate when running Xen on a workstation.
  631: TODO: NAT appears to be configured by "vif = [ '' ]".
  632: 
  633: The MAC address specified is the one used for the interface in the new
  634: domain.  The interface in dom0 will use this address XOR'd with
  635: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
  636: 
  637: Sizing domains
  638: --------------
  639: 
  640: Modern x86 hardware has vast amounts of resources.  However, many
  641: virtual servers can function just fine on far less.  A system with
  642: 256M of RAM and a 4G disk can be a reasonable choice.  Note that it is
  643: far easier to adjust virtual resources than physical ones.  For
  644: memory, it's just a config file edit and a reboot.  For disk, one can
  645: create a new file and vnconfig it (or lvm), and then dump/restore,
  646: just like updating physical disks, but without having to be there and
  647: without those pesky connectors.
  648: 
  649: Starting domains automatically
  650: ------------------------------
  651: 
  652: To start domains foo at bar at boot and shut them down cleanly on dom0
  653: shutdown, in rc.conf add:
  654: 
  655:         xendomains="foo bar"
  656: 
  657: TODO: Explain why 4.1 rc.d/xendomains has xl, when one should use xm
  658: on 4.1.  Or fix the xentools41 package to have xm
  659: 
  660: Creating specific unprivileged domains (domU)
  661: =============================================
  662: 
  663: Creating domUs is almost entirely independent of operating system.  We
  664: have already presented the basics of config files.  Note that you must
  665: have already completed the dom0 setup so that "xl list" (or "xm list")
  666: works.
  667: 
  668: Creating an unprivileged NetBSD domain (domU)
  669: ---------------------------------------------
  670: 
  671: See the earlier config file, and adjust memory.  Decide on how much
  672: storage you will provide, and prepare it (file or lvm).
  673: 
  674: While the kernel will be obtained from the dom0 filesystem, the same
  675: file should be present in the domU as /netbsd so that tools like
  676: savecore(8) can work.   (This is helpful but not necessary.)
  677: 
  678: The kernel must be specifically for Xen and for use as a domU.  The
  679: i386 and amd64 provide the following kernels:
  680: 
  681:         i386 XEN3_DOMU
  682:         i386 XEN3PAE_DOMU
  683: 	amd64 XEN3_DOMU
  684: 
  685: Unless using Xen 3.1 (and you shouldn't) with i386-mode Xen, you must
  686: use the PAE version of the i386 kernel.
  687: 
  688: This will boot NetBSD, but this is not that useful if the disk is
  689: empty.  One approach is to unpack sets onto the disk outside of xen
  690: (by mounting it, just as you would prepare a physical disk for a
  691: system you can't run the installer on).
  692: 
  693: A second approach is to run an INSTALL kernel, which has a miniroot
  694: and can load sets from the network.  To do this, copy the INSTALL
  695: kernel to / and change the kernel line in the config file to:
  696: 
  697:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  698: 
  699: Then, start the domain as "xl create -c configname".
  700: 
  701: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
  702: line should be used in the config file.
  703: 
  704:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  705: 
  706: After booting the domain, the option to install via CDROM may be
  707: selected.  The CDROM device should be changed to `xbd1d`.
  708: 
  709: Once done installing, "halt -p" the new domain (don't reboot or halt,
  710: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
  711: config file), switch the config file back to the XEN3_DOMU kernel,
  712: and start the new domain again. Now it should be able to use "root on
  713: xbd0a" and you should have a, functional NetBSD domU.
  714: 
  715: TODO: check if this is still accurate.
  716: When the new domain is booting you'll see some warnings about *wscons*
  717: and the pseudo-terminals. These can be fixed by editing the files
  718: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  719: `/etc/ttys`, except *console*, like this:
  720: 
  721:     console "/usr/libexec/getty Pc"         vt100   on secure
  722:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  723:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  724:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  725:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  726: 
  727: Finally, all screens must be commented out from `/etc/wscons.conf`.
  728: 
  729: It is also desirable to add
  730: 
  731:         powerd=YES
  732: 
  733: in rc.conf. This way, the domain will be properly shut down if
  734: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
  735: 
  736: Your domain should be now ready to work, enjoy.
  737: 
  738: Creating an unprivileged Linux domain (domU)
  739: --------------------------------------------
  740: 
  741: Creating unprivileged Linux domains isn't much different from
  742: unprivileged NetBSD domains, but there are some details to know.
  743: 
  744: First, the second parameter passed to the disk declaration (the '0x1' in
  745: the example below)
  746: 
  747:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  748: 
  749: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  750: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
  751: So, hda1 which has major 3 and minor 1 on a Linux system will have
  752: device number 0x301.  Alternatively, devices names can be used (hda,
  753: hdb, ...)  as xentools has a table to map these names to devices
  754: numbers.  To export a partition to a Linux guest we can use:
  755: 
  756:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
  757:         root = "/dev/hda1 ro"
  758: 
  759: and it will appear as /dev/hda on the Linux system, and be used as root
  760: partition.
  761: 
  762: To install the Linux system on the partition to be exported to the
  763: guest domain, the following method can be used: install
  764: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
  765: that will be the root partition of your Linux domain, and mount it.
  766: Then copy the files from a working Linux system, make adjustments in
  767: `/etc` (fstab, network config).  It should also be possible to extract
  768: binary packages such as .rpm or .deb directly to the mounted partition
  769: using the appropriate tool, possibly running under NetBSD's Linux
  770: emulation.  Once the filesystem has been populated, umount it.  If
  771: desirable, the filesystem can be converted to ext3 using tune2fs -j.
  772: It should now be possible to boot the Linux guest domain, using one of
  773: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
  774: 
  775: To get the linux console right, you need to add:
  776: 
  777:     extra = "xencons=tty1"
  778: 
  779: to your configuration since not all linux distributions auto-attach a
  780: tty to the xen console.
  781: 
  782: Creating an unprivileged Solaris domain (domU)
  783: ----------------------------------------------
  784: 
  785: See possibly outdated
  786: [Solaris domU instructions](/ports/xen/howto-solaris/).
  787: 
  788: 
  789: PCI passthrough: Using PCI devices in guest domains
  790: ---------------------------------------------------
  791: 
  792: The dom0 can give other domains access to selected PCI
  793: devices. This can allow, for example, a non-privileged domain to have
  794: access to a physical network interface or disk controller.  However,
  795: keep in mind that giving a domain access to a PCI device most likely
  796: will give the domain read/write access to the whole physical memory,
  797: as PCs don't have an IOMMU to restrict memory access to DMA-capable
  798: device.  Also, it's not possible to export ISA devices to non-dom0
  799: domains, which means that the primary VGA adapter can't be exported.
  800: A guest domain trying to access the VGA registers will panic.
  801: 
  802: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
  803: not been ported to later versions at this time.
  804: 
  805: For a PCI device to be exported to a domU, is has to be attached to
  806: the "pciback" driver in dom0.  Devices passed to the dom0 via the
  807: pciback.hide boot parameter will attach to "pciback" instead of the
  808: usual driver.  The list of devices is specified as "(bus:dev.func)",
  809: where bus and dev are 2-digit hexadecimal numbers, and func a
  810: single-digit number:
  811: 
  812:         pciback.hide=(00:0a.0)(00:06.0)
  813: 
  814: pciback devices should show up in the dom0's boot messages, and the
  815: devices should be listed in the `/kern/xen/pci` directory.
  816: 
  817: PCI devices to be exported to a domU are listed in the "pci" array of
  818: the domU's config file, with the format "0000:bus:dev.func".
  819: 
  820:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
  821: 
  822: In the domU an "xpci" device will show up, to which one or more pci
  823: busses will attach.  Then the PCI drivers will attach to PCI busses as
  824: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
  825: or any PCI drivers built in by default; you have to build your own
  826: kernel to use PCI devices in a domU.  Here's a kernel config example;
  827: note that only the "xpci" lines are unusual.
  828: 
  829:         include         "arch/i386/conf/XEN3_DOMU"
  830: 
  831:         # Add support for PCI busses to the XEN3_DOMU kernel
  832:         xpci* at xenbus ?
  833:         pci* at xpci ?
  834: 
  835:         # PCI USB controllers
  836:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
  837: 
  838:         # USB bus support
  839:         usb*    at uhci?
  840: 
  841:         # USB Hubs
  842:         uhub*   at usb?
  843:         uhub*   at uhub? port ? configuration ? interface ?
  844: 
  845:         # USB Mass Storage
  846:         umass*  at uhub? port ? configuration ? interface ?
  847:         wd*     at umass?
  848:         # SCSI controllers
  849:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
  850: 
  851:         # SCSI bus support (for both ahc and umass)
  852:         scsibus* at scsi?
  853: 
  854:         # SCSI devices
  855:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
  856:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
  857: 
  858: 
  859: NetBSD as a domU in a VPS
  860: =========================
  861: 
  862: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
  863: hardware.  This section explains how to deal with Xen in a domU as a
  864: virtual private server where you do not control or have access to the
  865: dom0.
  866: 
  867: VPS operators provide varying degrees of access and mechanisms for
  868: configuration.  The big issue is usually how one controls which kernel
  869: is booted, because the kernel is nominally in the dom0 filesystem (to
  870: which VPS users do not normally have acesss).
  871: 
  872: A VPS user may want to compile a kernel for security updates, to run
  873: npf, run IPsec, or any other reason why someone would want to change
  874: their kernel.
  875: 
  876: One approach is to have an adminstrative interface to upload a kernel,
  877: or to select from a prepopulated list.  Other approaches are py-grub
  878: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
  879: kernel from the domU filesystem.  This is closer to a regular physical
  880: computer, where someone who controls a machine can replace the kernel.
  881: 
  882: py-grub
  883: -------
  884: 
  885: py-grub runs in the dom0 and looks into the domU filesystem.  This
  886: implies that the domU must have a kernel in a filesystem in a format
  887: known to py-grub.  As of 2014, py-grub seems to be of mostly historical interest.
  888: 
  889: pvgrub
  890: ------
  891: 
  892: pvgrub is a version of grub that uses PV operations instead of BIOS
  893: calls.  It is booted from the dom0 as the domU kernel, and then reads
  894: /grub/menu.lst and loads a kernel from the domU filesystem.
  895: 
  896: [prgmr.com](http://prgmr.com/) uses this approach to let users choose
  897: their own operating system and kernel.  See then [prgmr.com NetBSD
  898: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU).
  899: 
  900: Because [grub's FFS code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
  901: appears not to support all aspects of modern FFS,
  902: typically one has an ext2 or FAT partition for the kernel, so that
  903: grub can understand it, which leads to /netbsd not being the actual
  904: kernel.  One must remember to update the special boot partiion.
  905: 
  906: Amazon
  907: ------
  908: 
  909: TODO: add link to NetBSD amazon howto.
  910: 
  911: Using npf
  912: ---------
  913: 
  914: In standard kernels, npf is a module, and thus cannot be loadeed in a
  915: DOMU kernel.
  916: 
  917: TODO: explain how to compile npf into a custom kernel, answering (but
  918: note that the problem was caused by not booting the right kernel):
  919: http://mail-index.netbsd.org/netbsd-users/2014/12/26/msg015576.html

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