File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.91: download - view: text, annotated - select for diffs
Wed Mar 4 01:32:16 2015 UTC (4 years, 2 months ago) by gdt
Branches: MAIN
CVS tags: HEAD
Turn PR links into real links.

    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: xenkernel45 provides Xen 4.5.  This is new to pkgsrc as of 2015-01 and
   96: not yet recommended for other than experimental/testing use.
   97: 
   98: Ideally newer versions of Xen will be added to pkgsrc.
   99: 
  100: Note that NetBSD support is called XEN3.  It works with Xen 3 and Xen
  101: 4 because the hypercall interface has been stable.
  102: 
  103: Xen command program
  104: -------------------
  105: 
  106: Early Xen used a program called xm to manipulate the system from the
  107: dom0.  Starting in 4.1, a replacement program with similar behavior
  108: called xl is provided, but it does not work well in 4.1.  In 4.2, both
  109: xm and xl work fine.  4.4 is the last version that has xm.  You must
  110: choose one or the other, because it affects which daemons you run.
  111: 
  112: NetBSD
  113: ------
  114: 
  115: The netbsd-5, netbsd-6, netbsd-7, and -current branches are all
  116: reasonable choices, with more or less the same considerations for
  117: non-Xen use.  Therefore, netbsd-6 is recommended as the stable version
  118: of the most recent release for production use.  For those wanting to
  119: learn Xen or without production stability concerns, netbsd-7 is likely
  120: most appropriate.
  121: 
  122: As of NetBSD 6, a NetBSD domU will support multiple vcpus.  There is
  123: no SMP support for NetBSD as dom0.  (The dom0 itself doesn't really
  124: need SMP; the lack of support is really a problem when using a dom0 as
  125: a normal computer.)
  126: 
  127: Architecture
  128: ------------
  129: 
  130: Xen itself can run on i386 or amd64 machines.  (Practically, almost
  131: any computer where one would want to run Xen supports amd64.)  If
  132: using an i386 NetBSD kernel for the dom0, PAE is required (PAE
  133: versions are built by default).  While i386 dom0 works fine, amd64 is
  134: recommended as more normal.
  135: 
  136: Xen 4.2 is the last version to support i386 as a host.  TODO: Clarify
  137: if this is about the CPU having to be amd64, or about the dom0 kernel
  138: having to be amd64.
  139: 
  140: One can then run i386 domUs and amd64 domUs, in any combination.  If
  141: running an i386 NetBSD kernel as a domU, the PAE version is required.
  142: (Note that emacs (at least) fails if run on i386 with PAE when built
  143: without, and vice versa, presumably due to bugs in the undump code.)
  144: 
  145: Stability
  146: ---------
  147: 
  148: Mostly, NetBSD as a dom0 or domU is quite stable.
  149: However, there are some open PRs indicating problems.
  150: 
  151:  - [PR 48125](http://gnats.netbsd.org/48125)
  152:  - [PR 47720](http://gnats.netbsd.org/47720)
  153: 
  154: Note also that there are issues with sparse vnd(4) instances, but
  155: these are not about Xen.
  156: 
  157: Recommendation
  158: --------------
  159: 
  160: Therefore, this HOWTO recommends running xenkernel42 (and xentools42),
  161: xl, the NetBSD 6 stable branch, and to use an amd64 kernel as the
  162: dom0.  Either the i386 or amd64 of NetBSD may be used as domUs.
  163: 
  164: Build problems
  165: --------------
  166: 
  167: Ideally, all versions of Xen in pkgsrc would build on all versions of
  168: NetBSD on both i386 and amd64.  However, that isn't the case.  Besides
  169: aging code and aging compilers, qemu (included in xentools for HVM
  170: support) is difficult to build.  The following are known to work or FAIL:
  171: 
  172:         xenkernel3 netbsd-5 amd64
  173:         xentools3 netbsd-5 amd64
  174:         xentools3=hvm netbsd-5 amd64 ????
  175:         xenkernel33 netbsd-5 amd64
  176:         xentools33 netbsd-5 amd64
  177:         xenkernel41 netbsd-5 amd64
  178:         xentools41 netbsd-5 amd64
  179:         xenkernel42 netbsd-5 amd64
  180:         xentools42 netbsd-5 amd64
  181: 
  182:         xenkernel3 netbsd-6 i386 FAIL
  183:         xentools3 netbsd-6 i386
  184:         xentools3-hvm netbsd-6 i386 FAIL (dependencies fail)
  185:         xenkernel33 netbsd-6 i386
  186:         xentools33 netbsd-6 i386
  187:         xenkernel41 netbsd-6 i386
  188:         xentools41 netbsd-6 i386
  189:         xenkernel42 netbsd-6 i386
  190:         xentools42 netbsd-6 i386 *MIXED
  191: 
  192: 	(all 3 and 33 seem to FAIL)
  193:         xenkernel41 netbsd-7 i386
  194:         xentools41 netbsd-7 i386
  195:         xenkernel42 netbsd-7 i386
  196:         xentools42 netbsd-7 i386 ??FAIL
  197: 
  198: (*On netbsd-6 i386, there is a xentools42 in the 2014Q3 official builds,
  199: but it does not build for gdt.)
  200: 
  201: NetBSD as a dom0
  202: ================
  203: 
  204: NetBSD can be used as a dom0 and works very well.  The following
  205: sections address installation, updating NetBSD, and updating Xen.
  206: Note that it doesn't make sense to talk about installing a dom0 OS
  207: without also installing Xen itself.  We first address installing
  208: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
  209: NetBSD install to a dom0 install by just changing the kernel and boot
  210: configuration.
  211: 
  212: For experimenting with Xen, a machine with as little as 1G of RAM and
  213: 100G of disk can work.  For running many domUs in productions, far
  214: more will be needed.
  215: 
  216: Styles of dom0 operation
  217: ------------------------
  218: 
  219: There are two basic ways to use Xen.  The traditional method is for
  220: the dom0 to do absolutely nothing other than providing support to some
  221: number of domUs.  Such a system was probably installed for the sole
  222: purpose of hosting domUs, and sits in a server room on a UPS.
  223: 
  224: The other way is to put Xen under a normal-usage computer, so that the
  225: dom0 is what the computer would have been without Xen, perhaps a
  226: desktop or laptop.  Then, one can run domUs at will.  Purists will
  227: deride this as less secure than the previous approach, and for a
  228: computer whose purpose is to run domUs, they are right.  But Xen and a
  229: dom0 (without domUs) is not meaingfully less secure than the same
  230: things running without Xen.  One can boot Xen or boot regular NetBSD
  231: alternately with little problems, simply refraining from starting the
  232: Xen daemons when not running Xen.
  233: 
  234: Note that NetBSD as dom0 does not support multiple CPUs.  This will
  235: limit the performance of the Xen/dom0 workstation approach.  In theory
  236: the only issue is that the "backend drivers" are not yet MPSAFE:
  237:   http://mail-index.netbsd.org/netbsd-users/2014/08/29/msg015195.html
  238: 
  239: Installation of NetBSD
  240: ----------------------
  241: 
  242: First,
  243: [install NetBSD/amd64](/guide/inst/)
  244: just as you would if you were not using Xen.
  245: However, the partitioning approach is very important.
  246: 
  247: If you want to use RAIDframe for the dom0, there are no special issues
  248: for Xen.  Typically one provides RAID storage for the dom0, and the
  249: domU systems are unaware of RAID.  The 2nd-stage loader bootxx_* skips
  250: over a RAID1 header to find /boot from a filesystem within a RAID
  251: partition; this is no different when booting Xen.
  252: 
  253: There are 4 styles of providing backing storage for the virtual disks
  254: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN,
  255: 
  256: With raw partitions, one has a disklabel (or gpt) partition sized for
  257: each virtual disk to be used by the domU.  (If you are able to predict
  258: how domU usage will evolve, please add an explanation to the HOWTO.
  259: Seriously, needs tend to change over time.)
  260: 
  261: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
  262: for domU disks.  This is almost as efficient as raw disk partitions
  263: and more flexible.  Hence raw disk partitions should typically not
  264: be used.
  265: 
  266: One can use files in the dom0 filesystem, typically created by dd'ing
  267: /dev/zero to create a specific size.  This is somewhat less efficient,
  268: but very convenient, as one can cp the files for backup, or move them
  269: between dom0 hosts.
  270: 
  271: Finally, in theory one can place the files backing the domU disks in a
  272: SAN.  (This is an invitation for someone who has done this to add a
  273: HOWTO page.)
  274: 
  275: Installation of Xen
  276: -------------------
  277: 
  278: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
  279: pkgsrc (or another matching pair).
  280: See [the pkgsrc
  281: documentation](http://www.NetBSD.org/docs/pkgsrc/) for help with pkgsrc.
  282: 
  283: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm.  More
  284: recent versions have HVM support integrated in the main xentools
  285: package.  It is entirely reasonable to run only PV guests.
  286: 
  287: Next you need to install the selected Xen kernel itself, which is
  288: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz".  Copy it to /.
  289: For debugging, one may copy xen-debug.gz; this is conceptually similar
  290: to DIAGNOSTIC and DEBUG in NetBSD.  xen-debug.gz is basically only
  291: useful with a serial console.  Then, place a NetBSD XEN3_DOM0 kernel
  292: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
  293: of a NetBSD build.  If using i386, use
  294: releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz.  (If using Xen
  295: 3.1 and i386, you may use XEN3_DOM0 with the non-PAE Xen.  But you
  296: should not use Xen 3.1.)  Both xen and the NetBSD kernel may be (and
  297: typically are) left compressed.
  298: 
  299: In a dom0 kernel, kernfs is mandatory for xend to comunicate with the
  300: kernel, so ensure that /kern is in fstab.  TODO: Say this is default,
  301: or file a PR and give a reference.
  302: 
  303: Because you already installed NetBSD, you have a working boot setup
  304: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
  305: beginning of your root filesystem, /boot present, and likely
  306: /boot.cfg.  (If not, fix before continuing!)
  307: 
  308: Add a line to to /boot.cfg to boot Xen.  See boot.cfg(5) for an
  309: example.  The basic line is
  310: 
  311:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
  312: 
  313: which specifies that the dom0 should have 256M, leaving the rest to be
  314: allocated for domUs.  To use a serial console, use
  315: 
  316:         menu=Xen:load /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=256M console=com1 com1=9600,8n1
  317: 
  318: which will use the first serial port for Xen (which counts starting
  319: from 1), forcing speed/parity, and also for NetBSD (which counts
  320: starting at 0).  In an attempt to add performance, one can also add
  321: 
  322:         dom0_max_vcpus=1 dom0_vcpus_pin
  323: 
  324: to force only one vcpu to be provided (since NetBSD dom0 can't use
  325: more) and to pin that vcpu to a physical cpu.  TODO: benchmark this.
  326: 
  327: As with non-Xen systems, you should have a line to boot /netbsd (a
  328: kernel that works without Xen) and fallback versions of the non-Xen
  329: kernel, Xen, and the dom0 kernel.
  330: 
  331: Now, reboot so that you are running a DOM0 kernel under Xen, rather
  332: than GENERIC without Xen.
  333: 
  334: Using grub (historic)
  335: ---------------------
  336: 
  337: Before NetBSD's native bootloader could support Xen, the use of
  338: grub was recommended.  If necessary, see the
  339: [old grub information](/ports/xen/howto-grub/).
  340: 
  341: The [HowTo on Installing into
  342: RAID-1](http://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
  343: explains how to set up booting a dom0 with Xen using grub with
  344: NetBSD's RAIDframe.  (This is obsolete with the use of NetBSD's native
  345: boot.)
  346: 
  347: Configuring Xen
  348: ---------------
  349: 
  350: Xen logs will be in /var/log/xen.
  351: 
  352: Now, you have a system that will boot Xen and the dom0 kernel, but not
  353: do anything else special.  Make sure that you have rebooted into Xen.
  354: There will be no domUs, and none can be started because you still have
  355: to configure the dom0 tools.  The daemons which should be run vary
  356: with Xen version and with whether one is using xm or xl.  Note that
  357: xend is for supporting "xm", and should only be used if you plan on
  358: using "xm".  Do NOT enable xend if you plan on using "xl" as it will
  359: cause problems.  Running xl without xencommons=YES (and starting it)
  360: will result in a hang (so don't do that; follow the HOWTO!).
  361: 
  362: The installation of NetBSD should already have created devices for xen
  363: (xencons, xenevt), but if they are not present, create them:
  364: 
  365:         cd /dev && sh MAKEDEV xen
  366: 
  367: TODO: Give 3.1 advice (or remove it from pkgsrc).
  368: 
  369: For 3.3 (and thus xm), add to rc.conf (but note that you should have
  370: installed 4.1 or 4.2):
  371: 
  372:         xend=YES
  373:         xenbackendd=YES
  374: 
  375: For 4.1 (and thus xm; xl is believed not to work well), add to rc.conf:
  376: 
  377:         xencommons=YES
  378:         xend=YES
  379: 
  380: (If you are using xentools41 from before 2014-12-26, change
  381: rc.d/xendomains to use xm rather than xl.)
  382: 
  383: For 4.2 with xm, add to rc.conf
  384: 
  385:         xencommons=YES
  386:         xend=YES
  387: 
  388: For 4.2 with xl, add to rc.conf:
  389: 
  390:         xencommons=YES
  391:         TODO: explain if there is a xend replacement
  392: 
  393: For 4.5 (and thus with xl), add to rc.conf:
  394: 
  395:         xencommons=YES
  396:         TODO: explain if there is a xend replacement
  397: 
  398: TODO: Recommend for/against xen-watchdog.
  399: 
  400: After you have configured the daemons and either started them (in the
  401: order given) or rebooted, use xm or xl to inspect Xen's boot messages,
  402: available resources, and running domains.  An example with xm follows:
  403: 
  404:         # xm dmesg
  405: 	[xen's boot info]
  406:         # xm info
  407: 	[available memory, etc.]
  408:         # xm list
  409:         Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  410:         Domain-0           0       64    0  r----     58.1
  411: 
  412: With xl, the commands are the same, and the output may be slightly
  413: different.  TODO: add example output for xl before the xm example,
  414: after confirming on 4.2 and resolving the TODO about rc.conf.
  415: 
  416: ### Issues with xencommons
  417: 
  418: xencommons starts xenstored, which stores data on behalf of dom0 and
  419: domUs.  It does not currently work to stop and start xenstored.
  420: Certainly all domUs should be shutdown first, following the sort order
  421: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
  422: and is not notified when xenstored exits, leading to not recreating
  423: the state when the new xenstored starts.  Until there's a mechanism to
  424: make this work, one should not expect to be able to restart xenstored
  425: (and thus xencommons).  There is currently no reason to expect that
  426: this will get fixed any time soon.
  427: 
  428: anita (for testing NetBSD)
  429: --------------------------
  430: 
  431: With the setup so far (assuming 4.2/xl), one should be able to run
  432: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
  433: root, because anita must create a domU):
  434: 
  435:         anita --vmm=xl test file:///usr/obj/i386/
  436: 
  437: Alternatively, one can use --vmm=xm to use xm-based domU creation
  438: instead (and must, on Xen <= 4.1).   TODO: confirm that anita xl really works.
  439:     
  440: Xen-specific NetBSD issues
  441: --------------------------
  442: 
  443: There are (at least) two additional things different about NetBSD as a
  444: dom0 kernel compared to hardware.
  445: 
  446: One is that modules are not usable in DOM0 kernels, so one must
  447: compile in what's needed.  It's not really that modules cannot work,
  448: but that modules must be built for XEN3_DOM0 because some of the
  449: defines change and the normal module builds don't do this.  Basically,
  450: enabling Xen changes the kernel ABI, and the module build system
  451: doesn't cope with this.
  452: 
  453: The other difference is that XEN3_DOM0 does not have exactly the same
  454: options as GENERIC.  While it is debatable whether or not this is a
  455: bug, users should be aware of this and can simply add missing config
  456: items if desired.
  457: 
  458: Updating NetBSD in a dom0
  459: -------------------------
  460: 
  461: This is just like updating NetBSD on bare hardware, assuming the new
  462: version supports the version of Xen you are running.  Generally, one
  463: replaces the kernel and reboots, and then overlays userland binaries
  464: and adjusts /etc.
  465: 
  466: Note that one must update both the non-Xen kernel typically used for
  467: rescue purposes and the DOM0 kernel used with Xen.
  468: 
  469: Converting from grub to /boot
  470: -----------------------------
  471: 
  472: These instructions were [TODO: will be] used to convert a system from
  473: grub to /boot.  The system was originally installed in February of
  474: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
  475: over time.  Before these commands, it was running NetBSD 6 i386, Xen
  476: 4.1 and grub, much like the message linked earlier in the grub
  477: section.
  478: 
  479:         # Install mbr bootblocks on both disks. 
  480:         fdisk -i /dev/rwd0d
  481:         fdisk -i /dev/rwd1d
  482:         # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
  483:         installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
  484:         installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
  485:         # Install secondary boot loader
  486:         cp -p /usr/mdec/boot /
  487:         # Create boog.cfg following earlier guidance:
  488:         menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
  489:         menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=256M
  490:         menu=GENERIC:boot
  491:         menu=GENERIC single-user:boot -s
  492:         menu=GENERIC.ok:boot netbsd.ok
  493:         menu=GENERIC.ok single-user:boot netbsd.ok -s
  494:         menu=Drop to boot prompt:prompt
  495:         default=1
  496:         timeout=30
  497: 
  498: TODO: actually do this and fix it if necessary.
  499: 
  500: Updating Xen versions
  501: ---------------------
  502: 
  503: Updating Xen is conceptually not difficult, but can run into all the
  504: issues found when installing Xen.  Assuming migration from 4.1 to 4.2,
  505: remove the xenkernel41 and xentools41 packages and install the
  506: xenkernel42 and xentools42 packages.  Copy the 4.2 xen.gz to /.
  507: 
  508: Ensure that the contents of /etc/rc.d/xen* are correct.  Enable the
  509: correct set of daemons.  Ensure that the domU config files are valid
  510: for the new version.
  511: 
  512: 
  513: Running Xen under qemu
  514: ----------------------
  515: 
  516: The astute reader will note that this section is somewhat twisted.
  517: However, it can be useful to run Xen under qemu either because the
  518: version of NetBSD as a dom0 does not run on the hardware in use, or to
  519: generate automated test cases involving Xen.
  520: 
  521: In 2015-01, the following combination was reported to mostly work:
  522: 
  523:         host OS: NetBSD/amd64 6.1.4
  524:         qemu: 2.2.0 from pkgsrc
  525:         Xen kernel: xenkernel42-4.2.5nb1 from pkgsrc
  526:         dom0 kernel: NetBSD/amd64 6.1.5
  527:         Xen tools: xentools42-4.2.5 from pkgsrc
  528: 
  529: See [PR 47720](http://gnats.netbsd.org/47720) for a problem with dom0
  530: shutdown.
  531: 
  532: Unprivileged domains (domU)
  533: ===========================
  534: 
  535: This section describes general concepts about domUs.  It does not
  536: address specific domU operating systems or how to install them.  The
  537: config files for domUs are typically in /usr/pkg/etc/xen, and are
  538: typically named so that the file name, domU name and the domU's host
  539: name match.
  540: 
  541: The domU is provided with cpu and memory by Xen, configured by the
  542: dom0.  The domU is provided with disk and network by the dom0,
  543: mediated by Xen, and configured in the dom0.
  544: 
  545: Entropy in domUs can be an issue; physical disks and network are on
  546: the dom0.  NetBSD's /dev/random system works, but is often challenged.
  547: 
  548: Config files
  549: ------------
  550: 
  551: There is no good order to present config files and the concepts
  552: surrounding what is being configured.  We first show an example config
  553: file, and then in the various sections give details.
  554: 
  555: See (at least in xentools41) /usr/pkg/share/examples/xen/xmexample*,
  556: for a large number of well-commented examples, mostly for running
  557: GNU/Linux.
  558: 
  559: The following is an example minimal domain configuration file
  560: "/usr/pkg/etc/xen/foo".  It is (with only a name change) an actual
  561: known working config file on Xen 4.1 (NetBSD 5 amd64 dom0 and NetBSD 5
  562: i386 domU).  The domU serves as a network file server.
  563: 
  564:         # -*- mode: python; -*-
  565: 
  566:         kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
  567:         memory = 1024
  568:         vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
  569:         disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
  570:                  'file:/n0/xen/foo-wd1,0x1,w' ]
  571: 
  572: The domain will have the same name as the file.  The kernel has the
  573: host/domU name in it, so that on the dom0 one can update the various
  574: domUs independently.  The vif line causes an interface to be provided,
  575: with a specific mac address (do not reuse MAC addresses!), in bridge
  576: mode.  Two disks are provided, and they are both writable; the bits
  577: are stored in files and Xen attaches them to a vnd(4) device in the
  578: dom0 on domain creation.  The system treates xbd0 as the boot device
  579: without needing explicit configuration.
  580: 
  581: By default xm looks for domain config files in /usr/pkg/etc/xen.  Note
  582: that "xm create" takes the name of a config file, while other commands
  583: take the name of a domain.  To create the domain, connect to the
  584: console, create the domain while attaching the console, shutdown the
  585: domain, and see if it has finished stopping, do (or xl with Xen >=
  586: 4.2):
  587: 
  588:         xm create foo
  589:         xm console foo
  590:         xm create -c foo
  591:         xm shutdown foo
  592:         xm list
  593: 
  594: Typing ^] will exit the console session.  Shutting down a domain is
  595: equivalent to pushing the power button; a NetBSD domU will receive a
  596: power-press event and do a clean shutdown.  Shutting down the dom0
  597: will trigger controlled shutdowns of all configured domUs.
  598: 
  599: domU kernels
  600: ------------
  601: 
  602: On a physical computer, the BIOS reads sector 0, and a chain of boot
  603: loaders finds and loads a kernel.  Normally this comes from the root
  604: filesystem.  With Xen domUs, the process is totally different.  The
  605: normal path is for the domU kernel to be a file in the dom0's
  606: filesystem.  At the request of the dom0, Xen loads that kernel into a
  607: new domU instance and starts execution.  While domU kernels can be
  608: anyplace, reasonable places to store domU kernels on the dom0 are in /
  609: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
  610: config files), or in /u0/xen (where the vdisks are).
  611: 
  612: Note that loading the domU kernel from the dom0 implies that boot
  613: blocks, /boot, /boot.cfg, and so on are all ignored in the domU.
  614: See the VPS section near the end for discussion of alternate ways to
  615: obtain domU kernels.
  616: 
  617: CPU and memory
  618: --------------
  619: 
  620: A domain is provided with some number of vcpus, less than the number
  621: of cpus seen by the hypervisor.  (For a dom0, this is controlled by
  622: the boot argument "dom0_max_vcpus=1".)  For a domU, it is controlled
  623: from the config file by the "vcpus = N" directive.
  624: 
  625: A domain is provided with memory; this is controlled in the config
  626: file by "memory = N" (in megabytes).  In the straightforward case, the
  627: sum of the the memory allocated to the dom0 and all domUs must be less
  628: than the available memory.
  629: 
  630: Xen also provides a "balloon" driver, which can be used to let domains
  631: use more memory temporarily.  TODO: Explain better, and explain how
  632: well it works with NetBSD.
  633: 
  634: Virtual disks
  635: -------------
  636: 
  637: With the file/vnd style, typically one creates a directory,
  638: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
  639: domUs.  Then, for each domU disk, one writes zeros to a file that then
  640: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
  641: for the first virtual disk for the domU called foo.  Writing zeros to
  642: the file serves two purposes.  One is that preallocating the contents
  643: improves performance.  The other is that vnd on sparse files has
  644: failed to work.  TODO: give working/notworking NetBSD versions for
  645: sparse vnd.  Note that the use of file/vnd for Xen is not really
  646: different than creating a file-backed virtual disk for some other
  647: purpose, except that xentools handles the vnconfig commands.  To
  648: create an empty 4G virtual disk, simply do
  649: 
  650:         dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
  651: 
  652: Do not use qemu-img-xen, because this will create sparse file.  There
  653: have been recent (2015) reports of sparse vnd(4) devices causing
  654: lockups, but there is apparently no PR.
  655: 
  656: With the lvm style, one creates logical devices.  They are then used
  657: similarly to vnds.  TODO: Add an example with lvm.
  658: 
  659: In domU config files, the disks are defined as a sequence of 3-tuples.
  660: The first element is "method:/path/to/disk".  Common methods are
  661: "file:" for file-backed vnd. and "phy:" for something that is already
  662: a (TODO: character or block) device.
  663: 
  664: The second element is an artifact of how virtual disks are passed to
  665: Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
  666: are given a device name to associate with the disk, and values like
  667: "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
  668: as xbd0, the second as xbd1, and so on.  However, xm/xl demand a
  669: second argument.  The name given is converted to a major/minor by
  670: calling stat(2) on the name in /dev and this is passed to the domU.
  671: In the general case, the dom0 and domU can be different operating
  672: systems, and it is an unwarranted assumption that they have consistent
  673: numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
  674: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
  675: for the second works fine and avoids this issue.  For a GNU/Linux
  676: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
  677: /dev/hda1.
  678: 
  679: The third element is "w" for writable disks, and "r" for read-only
  680: disks.
  681: 
  682: Virtual Networking
  683: ------------------
  684: 
  685: Xen provides virtual ethernets, each of which connects the dom0 and a
  686: domU.  For each virtual network, there is an interface "xvifN.M" in
  687: the dom0, and in domU index N, a matching interface xennetM (NetBSD
  688: name).  The interfaces behave as if there is an Ethernet with two
  689: adaptors connected.  From this primitive, one can construct various
  690: configurations.  We focus on two common and useful cases for which
  691: there are existing scripts: bridging and NAT.
  692: 
  693: With bridging (in the example above), the domU perceives itself to be
  694: on the same network as the dom0.  For server virtualization, this is
  695: usually best.  Bridging is accomplished by creating a bridge(4) device
  696: and adding the dom0's physical interface and the various xvifN.0
  697: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
  698: config file.  The bridge must be set up already in the dom0; an
  699: example /etc/ifconfig.bridge0 is:
  700: 
  701:         create
  702:         up
  703:         !brconfig bridge0 add wm0
  704: 
  705: With NAT, the domU perceives itself to be behind a NAT running on the
  706: dom0.  This is often appropriate when running Xen on a workstation.
  707: TODO: NAT appears to be configured by "vif = [ '' ]".
  708: 
  709: The MAC address specified is the one used for the interface in the new
  710: domain.  The interface in dom0 will use this address XOR'd with
  711: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
  712: 
  713: Sizing domains
  714: --------------
  715: 
  716: Modern x86 hardware has vast amounts of resources.  However, many
  717: virtual servers can function just fine on far less.  A system with
  718: 256M of RAM and a 4G disk can be a reasonable choice.  Note that it is
  719: far easier to adjust virtual resources than physical ones.  For
  720: memory, it's just a config file edit and a reboot.  For disk, one can
  721: create a new file and vnconfig it (or lvm), and then dump/restore,
  722: just like updating physical disks, but without having to be there and
  723: without those pesky connectors.
  724: 
  725: Starting domains automatically
  726: ------------------------------
  727: 
  728: To start domains foo at bar at boot and shut them down cleanly on dom0
  729: shutdown, in rc.conf add:
  730: 
  731:         xendomains="foo bar"
  732: 
  733: Note that earlier versions of the xentools41 xendomains rc.d scripth
  734: usd xl, when one should use xm with 4.1.
  735: 
  736: Creating specific unprivileged domains (domU)
  737: =============================================
  738: 
  739: Creating domUs is almost entirely independent of operating system.  We
  740: have already presented the basics of config files.  Note that you must
  741: have already completed the dom0 setup so that "xl list" (or "xm list")
  742: works.
  743: 
  744: Creating an unprivileged NetBSD domain (domU)
  745: ---------------------------------------------
  746: 
  747: See the earlier config file, and adjust memory.  Decide on how much
  748: storage you will provide, and prepare it (file or lvm).
  749: 
  750: While the kernel will be obtained from the dom0 filesystem, the same
  751: file should be present in the domU as /netbsd so that tools like
  752: savecore(8) can work.   (This is helpful but not necessary.)
  753: 
  754: The kernel must be specifically for Xen and for use as a domU.  The
  755: i386 and amd64 provide the following kernels:
  756: 
  757:         i386 XEN3_DOMU
  758:         i386 XEN3PAE_DOMU
  759: 	amd64 XEN3_DOMU
  760: 
  761: Unless using Xen 3.1 (and you shouldn't) with i386-mode Xen, you must
  762: use the PAE version of the i386 kernel.
  763: 
  764: This will boot NetBSD, but this is not that useful if the disk is
  765: empty.  One approach is to unpack sets onto the disk outside of xen
  766: (by mounting it, just as you would prepare a physical disk for a
  767: system you can't run the installer on).
  768: 
  769: A second approach is to run an INSTALL kernel, which has a miniroot
  770: and can load sets from the network.  To do this, copy the INSTALL
  771: kernel to / and change the kernel line in the config file to:
  772: 
  773:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  774: 
  775: Then, start the domain as "xl create -c configname".
  776: 
  777: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
  778: line should be used in the config file.
  779: 
  780:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  781: 
  782: After booting the domain, the option to install via CDROM may be
  783: selected.  The CDROM device should be changed to `xbd1d`.
  784: 
  785: Once done installing, "halt -p" the new domain (don't reboot or halt,
  786: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
  787: config file), switch the config file back to the XEN3_DOMU kernel,
  788: and start the new domain again. Now it should be able to use "root on
  789: xbd0a" and you should have a, functional NetBSD domU.
  790: 
  791: TODO: check if this is still accurate.
  792: When the new domain is booting you'll see some warnings about *wscons*
  793: and the pseudo-terminals. These can be fixed by editing the files
  794: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  795: `/etc/ttys`, except *console*, like this:
  796: 
  797:     console "/usr/libexec/getty Pc"         vt100   on secure
  798:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  799:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  800:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  801:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  802: 
  803: Finally, all screens must be commented out from `/etc/wscons.conf`.
  804: 
  805: It is also desirable to add
  806: 
  807:         powerd=YES
  808: 
  809: in rc.conf. This way, the domain will be properly shut down if
  810: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
  811: 
  812: Your domain should be now ready to work, enjoy.
  813: 
  814: Creating an unprivileged Linux domain (domU)
  815: --------------------------------------------
  816: 
  817: Creating unprivileged Linux domains isn't much different from
  818: unprivileged NetBSD domains, but there are some details to know.
  819: 
  820: First, the second parameter passed to the disk declaration (the '0x1' in
  821: the example below)
  822: 
  823:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  824: 
  825: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  826: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
  827: So, hda1 which has major 3 and minor 1 on a Linux system will have
  828: device number 0x301.  Alternatively, devices names can be used (hda,
  829: hdb, ...)  as xentools has a table to map these names to devices
  830: numbers.  To export a partition to a Linux guest we can use:
  831: 
  832:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
  833:         root = "/dev/hda1 ro"
  834: 
  835: and it will appear as /dev/hda on the Linux system, and be used as root
  836: partition.
  837: 
  838: To install the Linux system on the partition to be exported to the
  839: guest domain, the following method can be used: install
  840: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
  841: that will be the root partition of your Linux domain, and mount it.
  842: Then copy the files from a working Linux system, make adjustments in
  843: `/etc` (fstab, network config).  It should also be possible to extract
  844: binary packages such as .rpm or .deb directly to the mounted partition
  845: using the appropriate tool, possibly running under NetBSD's Linux
  846: emulation.  Once the filesystem has been populated, umount it.  If
  847: desirable, the filesystem can be converted to ext3 using tune2fs -j.
  848: It should now be possible to boot the Linux guest domain, using one of
  849: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
  850: 
  851: To get the linux console right, you need to add:
  852: 
  853:     extra = "xencons=tty1"
  854: 
  855: to your configuration since not all linux distributions auto-attach a
  856: tty to the xen console.
  857: 
  858: Creating an unprivileged Solaris domain (domU)
  859: ----------------------------------------------
  860: 
  861: See possibly outdated
  862: [Solaris domU instructions](/ports/xen/howto-solaris/).
  863: 
  864: 
  865: PCI passthrough: Using PCI devices in guest domains
  866: ---------------------------------------------------
  867: 
  868: The dom0 can give other domains access to selected PCI
  869: devices. This can allow, for example, a non-privileged domain to have
  870: access to a physical network interface or disk controller.  However,
  871: keep in mind that giving a domain access to a PCI device most likely
  872: will give the domain read/write access to the whole physical memory,
  873: as PCs don't have an IOMMU to restrict memory access to DMA-capable
  874: device.  Also, it's not possible to export ISA devices to non-dom0
  875: domains, which means that the primary VGA adapter can't be exported.
  876: A guest domain trying to access the VGA registers will panic.
  877: 
  878: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
  879: not been ported to later versions at this time.
  880: 
  881: For a PCI device to be exported to a domU, is has to be attached to
  882: the "pciback" driver in dom0.  Devices passed to the dom0 via the
  883: pciback.hide boot parameter will attach to "pciback" instead of the
  884: usual driver.  The list of devices is specified as "(bus:dev.func)",
  885: where bus and dev are 2-digit hexadecimal numbers, and func a
  886: single-digit number:
  887: 
  888:         pciback.hide=(00:0a.0)(00:06.0)
  889: 
  890: pciback devices should show up in the dom0's boot messages, and the
  891: devices should be listed in the `/kern/xen/pci` directory.
  892: 
  893: PCI devices to be exported to a domU are listed in the "pci" array of
  894: the domU's config file, with the format "0000:bus:dev.func".
  895: 
  896:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
  897: 
  898: In the domU an "xpci" device will show up, to which one or more pci
  899: busses will attach.  Then the PCI drivers will attach to PCI busses as
  900: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
  901: or any PCI drivers built in by default; you have to build your own
  902: kernel to use PCI devices in a domU.  Here's a kernel config example;
  903: note that only the "xpci" lines are unusual.
  904: 
  905:         include         "arch/i386/conf/XEN3_DOMU"
  906: 
  907:         # Add support for PCI busses to the XEN3_DOMU kernel
  908:         xpci* at xenbus ?
  909:         pci* at xpci ?
  910: 
  911:         # PCI USB controllers
  912:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
  913: 
  914:         # USB bus support
  915:         usb*    at uhci?
  916: 
  917:         # USB Hubs
  918:         uhub*   at usb?
  919:         uhub*   at uhub? port ? configuration ? interface ?
  920: 
  921:         # USB Mass Storage
  922:         umass*  at uhub? port ? configuration ? interface ?
  923:         wd*     at umass?
  924:         # SCSI controllers
  925:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
  926: 
  927:         # SCSI bus support (for both ahc and umass)
  928:         scsibus* at scsi?
  929: 
  930:         # SCSI devices
  931:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
  932:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
  933: 
  934: 
  935: NetBSD as a domU in a VPS
  936: =========================
  937: 
  938: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
  939: hardware.  This section explains how to deal with Xen in a domU as a
  940: virtual private server where you do not control or have access to the
  941: dom0.  This is not intended to be an exhaustive list of VPS providers;
  942: only a few are mentioned that specifically support NetBSD.
  943: 
  944: VPS operators provide varying degrees of access and mechanisms for
  945: configuration.  The big issue is usually how one controls which kernel
  946: is booted, because the kernel is nominally in the dom0 filesystem (to
  947: which VPS users do not normally have acesss).  A second issue is how
  948: to install NetBSD.
  949: A VPS user may want to compile a kernel for security updates, to run
  950: npf, run IPsec, or any other reason why someone would want to change
  951: their kernel.
  952: 
  953: One approach is to have an adminstrative interface to upload a kernel,
  954: or to select from a prepopulated list.  Other approaches are pygrub
  955: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
  956: kernel from the domU filesystem.  This is closer to a regular physical
  957: computer, where someone who controls a machine can replace the kernel.
  958: 
  959: A second issue is multiple CPUs.  With NetBSD 6, domUs support
  960: multiple vcpus, and it is typical for VPS providers to enable multiple
  961: CPUs for NetBSD domUs.
  962: 
  963: pygrub
  964: -------
  965: 
  966: pygrub runs in the dom0 and looks into the domU filesystem.  This
  967: implies that the domU must have a kernel in a filesystem in a format
  968: known to pygrub.  As of 2014, pygrub seems to be of mostly historical
  969: interest.
  970: 
  971: pvgrub
  972: ------
  973: 
  974: pvgrub is a version of grub that uses PV operations instead of BIOS
  975: calls.  It is booted from the dom0 as the domU kernel, and then reads
  976: /grub/menu.lst and loads a kernel from the domU filesystem.
  977: 
  978: [Panix](http://www.panix.com/) lets users use pvgrub.  Panix reports
  979: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
  980: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
  981: page](http://www.panix.com/v-colo/grub.html), which describes only
  982: Linux but should be updated to cover NetBSD :-).
  983: 
  984: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
  985: their own kernel.  See then [prgmr.com NetBSD
  986: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
  987: (which is in need of updating).
  988: 
  989: It appears that [grub's FFS
  990: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
  991: does not support all aspects of modern FFS, but there are also reports
  992: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
  993: partition for the kernel with the intent that grub can understand it,
  994: which leads to /netbsd not being the actual kernel.  One must remember
  995: to update the special boot partiion.
  996: 
  997: Amazon
  998: ------
  999: 
 1000: TODO: add link to NetBSD amazon howto.
 1001: 
 1002: Using npf
 1003: ---------
 1004: 
 1005: In standard kernels, npf is a module, and thus cannot be loaded in a
 1006: DOMU kernel.
 1007: 
 1008: TODO: explain how to compile npf into a custom kernel, answering (but
 1009: note that the problem was caused by not booting the right kernel):
 1010: http://mail-index.netbsd.org/netbsd-users/2014/12/26/msg015576.html
 1011: 
 1012: TODO items for improving NetBSD/xen
 1013: ===================================
 1014: 
 1015: * Package Xen 4.4.
 1016: * Get PCI passthrough working on Xen 4.2 (or 4.4).
 1017: * Get pvgrub into pkgsrc, either via xentools or separately.
 1018: * grub
 1019:   * Check/add support to pkgsrc grub2 for UFS2 and arbitrary
 1020:     fragsize/blocksize (UFS2 support may be present; the point is to
 1021:     make it so that with any UFS1/UFS2 filesystem setup that works
 1022:     with NetBSD grub will also work).
 1023:     See [pkg/40258](http://gnats.netbsd.org/40258).
 1024:   * Push patches upstream.
 1025:   * Get UFS2 patches into pvgrub.
 1026: * Add support for PV ops to a version of /boot, and make it usable as
 1027:   a kernel in Xen, similar to pvgrub.

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