File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.205: download - view: text, annotated - select for diffs
Mon Mar 15 23:01:56 2021 UTC (2 months ago) by gdt
Branches: MAIN
CVS tags: HEAD
xen howo: rototill some more

redo virtualization types

Separate provider-specific pvgrub info into sections.  Declare that
only providers that actually support NetBSD are listed.

general editing; no really significant changes.

    1: [[!meta title="Xen Status and HowTo"]]
    3: Xen is a Type 1 hypervisor which supports running multiple guest operating
    4: systems on a single physical machine. One uses the Xen kernel to control the
    5: CPU, memory and console, a dom0 operating system which mediates access to
    6: other hardware (e.g., disks, network, USB), and one or more domU operating
    7: systems which operate in an unprivileged virtualized environment. IO requests
    8: from the domU systems are forwarded by the Xen hypervisor to the dom0 to be
    9: fulfilled.
   11: This document provides status on what Xen things work on NetBSD
   12: (upstream documentation might say something works if it works on some
   13: particular Linux system).
   15: This document is also a HOWTO that presumes a basic familiarity with
   16: the Xen system architecture, with installing NetBSD on amd64 hardware,
   17: and with installing software from pkgsrc.  See also the [Xen
   18: website](
   20: If this document says that something works, and you find that it does
   21: not, it is best to ask on port-xen and if you are correct to file a
   22: PR.
   24: [[!toc]]
   26: # Overview
   28: The basic concept of Xen is that the hypervisor (xenkernel) runs on
   29: the hardware, and runs a privileged domain ("dom0") that can access
   30: disks/networking/etc.  One then runs additional unprivileged domains
   31: (each a "domU"), presumably to do something useful.
   33: This HOWTO addresses how to run a NetBSD dom0 (and hence also build
   34: xen itself).  It also addresses how to run domUs in that environment,
   35: and how to deal with having a domU in a Xen environment run by someone
   36: else and/or not running NetBSD.
   38: There are many choices one can make; the HOWTO recommends the standard
   39: approach and limits discussion of alternatives in many cases.
   41: ## Guest Styles
   43: Xen supports different styles of guests.  See
   44: for a
   45: discussion.
   47: This table shows the styles, and if a NetBSD dom0 can run in that
   48: style, if a NetBSD dom0 can sypport that style of guest in a domU, and
   49: if NetBSD as a domU can support that style.
   51: [[!table data="""
   52: Style of guest	|dom0 can be?	|dom0 can support?	|domU can be?
   53: PV		|yes		|yes			|yes
   54: HVM		|N/A		|yes			|yes
   55: PVHVM		|N/A		|yes			|current only
   56: PVH		|not yet	|current only		|current only
   57: """]]
   59: In PV (paravirtualized) mode, the guest OS does not attempt to access
   60: hardware directly, but instead makes hypercalls to the hypervisor; PV
   61: guests must be specifically coded for Xen.  See
   62: [PV](\)).
   64: In HVM (Hardware Virtual Machine) mode, no guest modification is
   65: required.  However, hardware support is required, such as VT-x on
   66: Intel CPUs and SVM on AMD CPUs to assist with the processor emulation.
   67: The dom0 runs qemu to emulate hardware other than the processor.  It
   68: is therefore non-sensical to have an HVM dom0, because there is no
   69: underlying system to provide emulation.
   71: In PVHVM mode, the guest runs as HVM, but additionally uses PV
   72: drivers for efficiency.  Therefore it is non-sensical for to have a
   73: PVHVM dom0.  See [PV on HVM](
   75: There have been two PVH modes: original PVH and PVHv2.  Original PVH
   76: was based on PV mode and is no longer relevant at all.  Therefore
   77: PVHv2 is written as PVH, here and elsewhere.  PVH is basically
   78: lightweight HVM with PV drivers.  A critical feature of it is that
   79: qemu is not needed; the hypervisor can do the emulation that is
   80: required.  Thus, a dom0 can be PVH.  The source code uses PVH and
   81: config files use pvh, but NB that this refers to PVHv2.  See
   82: [PVH(v2)](\)_Domu).
   84: At system boot, the dom0 kernel is loaded as a module with Xen as the
   85: kernel.  The dom0 can start one or more domUs.  (Booting is explained
   86: in detail in the dom0 section.)
   88: ## CPU Architecture
   90: Xen runs on x86_64 hardware (the NetBSD amd64 port).
   92: There is a concept of Xen running on ARM, but there are no reports of this working with NetBSD.
   94: The dom0 system should be amd64.  (Instructions for i386PAE dom0 have been removed from the HOWTO.)
   96: The domU can be i386 PAE or amd64.
   97: i386 PAE at one point was considered as [faster]( than amd64.
   98: However, as of 2021 it is normal to use amd64 as the domU architecture, and use of i386 is dwindling.
  100: ## Xen Versions
  102: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
  103: xenkernel and xentools.  We will refer only to the kernel versions,
  104: but note that both packages must be installed together and must have
  105: matching versions.
  107: Versions available in pkgsrc:
  109: [[!table data="""
  110: Xen Version	|Package Name	|Xen CPU Support	|EOL'ed By Upstream
  111: 4.11		|xenkernel411	|x86_64			|No
  112: 4.13		|xenkernel413	|x86_64			|No
  113: """]]
  115: See also the [Xen Security Advisory page](
  117: Older Xen had a python-based management tool called xm; this has been
  118: replaced by xl.
  120: ## NetBSD versions
  122: Xen has been supported in NetBSD for a long time, at least since 2005.
  123: Initially Xen was PV only.
  125: NetBSD Xen has always supported PV, in both dom0 and domU; for a long
  126: time this was the only way.  NetBSD >=8 as a dom0 supports HVM mode in
  127: domUs.
  129: Support for PVHVM and PVH is available only in NetBSD-current; this is
  130: currently somewhat experimental, although PVHVM appears reasonably
  131: solid.
  133: NetBSD up to and including NetBSD 9 as a dom0 cannot safely run SMP.
  134: Even if one added "options MULTIPROCESSOR" and configured multiple
  135: vcpus, the kernel is likely to crash because of drivers without
  136: adequate locking.
  138: NetBSD-current supports SMP in dom0, and XEN3_DOM0 includes "options
  141: NetBSD (since NetBSD 6), when run as a domU, can run SMP, using
  142: multiple CPUs if provided.  The XEN3_DOMU kernel is built
  143: with "options MULITPROCESSOR".
  145: Note that while Xen 4.13 is current, the kernel support is still
  146: called XEN3, because the hypercall interface has not changed
  147: significantly.
  149: # Creating a NetBSD dom0
  151: In order to install a NetBSD as a dom0, one first installs a normal
  152: NetBSD system, and then pivot the install to a dom0 install by
  153: changing the kernel and boot configuration.
  155: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
  156: 512M it worked reliably.  This does not make sense, but if you see
  157: "not ELF" after Xen boots, try increasing dom0 RAM.
  159: ## Installation of NetBSD
  161: [Install NetBSD/amd64](/guide/inst/) just as you would if you were not
  162: using Xen.  Therefore, use the most recent release, or a build from
  163: the most recent stable branch.  Alternatively, use -current, being
  164: mindful of all the usual caveats of lower stability of current, and
  165: likely a bit more so.  Think about how you will provide storage for
  166: disk images.
  168: ## Installation of Xen
  170: ### Building Xen
  172: Use the most recent version of Xen in pkgsrc, unless the DESCR says
  173: that it is not suitable.  Therefore, choose 4.13.  In the dom0,
  174: install xenkernel413 and xentools413 from pkgsrc.
  176: Once this is done, copy the Xen kernel from where pkgsrc puts it to
  177: where the boot process will be able to find it:
  179: [[!template id=programlisting text="""
  180: # cp -p /usr/pkg/xen413-kernel/xen.gz /
  181: """]]
  183: Then, place a NetBSD XEN3_DOM0 kernel in the `/` directory. Such
  184: kernel can either be taken from a local release run, compiled
  185: manually, or downloaded from the NetBSD FTP, for example at:
  187: [[!template id=programlisting text="""
  189: """]]
  191: ### Configuring booting
  193: Read boot.cfg(8) carefully.  Add lines to /boot.cfg to boot Xen,
  194: adjusting for your root filesystem:
  196: [[!template id=filecontent name="/boot.cfg" text="""
  197: menu=Xen:load /netbsd-XEN3_DOM0.gz root=wd0a console=pc;multiboot /xen.gz dom0_mem=512M
  198: menu=Xen single user:load /netbsd-XEN3_DOM0.gz root=wd0a console=pc -s;multiboot /xen.gz dom0_mem=512M
  199: """]]
  201: This specifies that the dom0 should have 512MB of ram, leaving the rest
  202: to be allocated for domUs.
  204: NB: This says add, not replace, so that you will be able to more
  205: easily boot a NetBSD kernel without Xen.  Once Xen boots ok, you may
  206: want to set it as default.  It is highly likely that you will have
  207: trouble at some point, and keeping an up-to-date GENERIC for use in
  208: fixing problems is the standard prudent approach.
  210: \todo Explain why rndseed is not set with Xen as part of the dom0
  211: subconfiguration.
  213: Note that you are likely to have to set root= because the boot device
  214: from /boot is not passed via Xen to the dom0 kernel.  With one disk,
  215: it will work, but e.g. plugging in USB disk to a machine with root on
  216: wd0a causes boot to fail.
  218: Beware that userconf statements must be attached to the dom0 load, and
  219: may not be at top-level, because then they would try to configure the
  220: hypervisor, if there is a way to pass them via multiboot.  It appears
  221: that adding `userconf=pckbc` to `/boot.cfg` causes Xen to crash very
  222: early with a heap overflow.
  224: ### Console selection
  226: See boot_console(8).  Understand that you should start from a place of
  227: having console setup correct for booting GENERIC before trying to
  228: configure Xen.
  230: Generally for GENERIC, one sets the console in bootxx_ffsv1 or
  231: equivalent, and this is passed on to /boot (where one typically does
  232: not set the console).  This configuration of bootxx_ffsv1 should also
  233: be in place for Xen systems, to allow seeing messages from /boot and
  234: use of a keyboard to select a line from the menu.  And, one should
  235: have a working boot path to GENERIC for rescue situations.
  237: With GENERIC, the boot options are passed on to /netbsd, but there is
  238: currently no mechanism to pass these via multiboot to the hypervisor.
  239: Thus, in addition to configuring the console in the boot blocks, one
  240: must also configure it for Xen.
  242: By default, the hypervisor (Xen itself) will use some sort of vga
  243: device as the console, much like GENERIC uses by default.  The vga
  244: console is relinquished at the conclusion of hypervisor boot, before
  245: the dom0 is started.  Xen when using a vga console does not process
  246: console input.
  248: The hypervisor can be configured to use a serial port console, e.g.
  249: [[!template id=filecontent name="/boot.cfg" text="""
  250: menu=Xen:load /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
  251: """]]
  252: This example uses the first serial port (Xen counts from 1; this is
  253: what NetBSD would call com0), and sets speed and parity.  (The dom0 is
  254: then configured to use the same serial port in this example.)
  256: With the hypervisor configured for a serial console, it can get input,
  257: and there is a notion of passing this input to the dom0.  \todo
  258: Explain why, if Xen has a serial console, the dom0 console is
  259: typically also configured to open that same serial port, instead of
  260: getting the passthrough input via the xen console.
  262: One also configures the console for the dom0.  While one might expect
  263: console=pc to be default, following behavior of GENERIC, a hasty read
  264: of the code suggests there is no default and booting without a
  265: selected console might lead to a panic.  Also, there is merit in
  266: explicit configuration.  Therefore the standard approach is to place
  267: console=pc as part of the load statement for the dom0 kernel, or
  268: alternatively console=com0.
  270: The NetBSD dom0 kernel will attach xencons(4) (the man page does not
  271: exist), but this is not used as a console.  It is used to obtain the
  272: messages from the hypervisor's console; run `xl dmesg` to see them.
  274: ### Tuning
  276: In an attempt to add performance, one can also add `dom0_max_vcpus=1
  277: dom0_vcpus_pin`, to force only one vcpu to be provided (since NetBSD
  278: dom0 can't use more) and to pin that vcpu to a physical CPU. Xen has
  279: [many boot
  280: options](,
  281: and other than dom0 memory and max_vcpus, they are generally not
  282: necessary.
  284: \todo Revisit this advice with current.
  285: \todo Explain if anyone has ever actually measured that this helps.
  287: ### rc.conf
  289: Ensure that the boot scripts installed in
  290: `/usr/pkg/share/examples/rc.d` are in `/etc/rc.d`, either because you
  291: have `PKG_RCD_SCRIPTS=yes`, or manually.  (This is not special to Xen,
  292: but a normal part of pkgsrc usage.)
  294: Set `xencommons=YES` in rc.conf:
  296: [[!template id=filecontent name="/etc/rc.conf" text="""
  297: xencommons=YES
  298: """]]
  300: \todo Recommend for/against xen-watchdog.
  302: ### Testing
  304: Now, reboot so that you are running a DOM0 kernel under Xen, rather
  305: than GENERIC without Xen.
  307: Once the reboot is done, use `xl` to inspect Xen's boot messages,
  308: available resources, and running domains.  For example:
  310: [[!template id=programlisting text="""
  311: # xl dmesg
  312: ... xen's boot info ...
  313: # xl info
  314: ... available memory, etc ...
  315: # xl list
  316: Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  317: Domain-0           0       64    0  r----     58.1
  318: """]]
  320: Xen logs will be in /var/log/xen.
  322: ### Issues with xencommons
  324: `xencommons` starts `xenstored`, which stores data on behalf of dom0 and
  325: domUs.  It does not currently work to stop and start xenstored.
  326: Certainly all domUs should be shutdown first, following the sort order
  327: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
  328: and is not notified when xenstored exits, leading to not recreating
  329: the state when the new xenstored starts.  Until there's a mechanism to
  330: make this work, one should not expect to be able to restart xenstored
  331: (and thus xencommons).  There is currently no reason to expect that
  332: this will get fixed any time soon.
  333: \todo Confirm if this is still true in 2020.
  335: ## Xen-specific NetBSD issues
  337: There are (at least) two additional things different about NetBSD as a
  338: dom0 kernel compared to hardware.
  340: One is that through NetBSD 9 the module ABI is different because some
  341: of the #defines change, so there are separate sets of modules in
  342: /stand.  (Further, zfs in Xen is troubled because of differing
  343: MAXPHYS; see the zfs howto for more.)  In NetBSD-current, there is
  344: only one set of modules.
  346: The other difference is that XEN3_DOM0 does not have exactly the same
  347: options as GENERIC.  While this is roughly agreed to be in large part
  348: a bug, users should be aware of this and can simply add missing config
  349: items if desired.
  351: Finally, there have been occasional reports of trouble with X11
  352: servers in NetBSD as a dom0.  Some hardware support is intentionally
  353: disabled in XEN3_DOM0.
  355: ## Updating Xen in a dom0
  357: Basically, update the xenkernel and xentools packages and copy the new
  358: Xen kernel into place, and reboot.  This procedure should be usable to
  359: update to a new Xen release, but the reader is reminded that having a
  360: non-Xen boot methods was recommended earlier.
  362: ## Updating NetBSD in a dom0
  364: This is just like updating NetBSD on bare hardware, assuming the new
  365: version supports the version of Xen you are running.  Generally, one
  366: replaces the kernel and reboots, and then overlays userland binaries
  367: and adjusts `/etc`.
  369: Note that one should update both the non-Xen kernel typically used for
  370: rescue purposes, as well as the DOM0 kernel used with Xen.
  372: ## anita (for testing NetBSD)
  374: With a NetBSD dom0, even without any domUs, one can run anita (see
  375: pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as root,
  376: because anita must create a domU):
  378: [[!template id=programlisting text="""
  379: anita --vmm=xl test file:///usr/obj/i386/
  380: """]]
  382: # Unprivileged domains (domU)
  384: This section describes general concepts about domUs.  It does not
  385: address specific domU operating systems or how to install them.  The
  386: config files for domUs are typically in `/usr/pkg/etc/xen`, and are
  387: typically named so that the file name, domU name and the domU's host
  388: name match.
  390: The domU is provided with CPU and memory by Xen, configured by the
  391: dom0.  The domU is provided with disk and network by the dom0,
  392: mediated by Xen, and configured in the dom0.
  394: Entropy in domUs can be an issue; physical disks and network are on
  395: the dom0.  NetBSD's /dev/random system works, but is often challenged.
  397: ## Config files
  399: See /usr/pkg/share/examples/xen/xlexample* for a very small number of
  400: examples for running GNU/Linux.
  402: The following is an example minimal domain configuration file.  The
  403: domU serves as a network file server.
  405: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
  406: name = "domU-id"
  407: kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
  408: memory = 1024
  409: vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
  410: disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
  411:          'file:/n0/xen/foo-wd1,0x1,w' ]
  412: """]]
  414: The domain will have name given in the `name` setting.  The kernel has the
  415: host/domU name in it, so that on the dom0 one can update the various
  416: domUs independently.  The `vif` line causes an interface to be provided,
  417: with a specific mac address (do not reuse MAC addresses!), in bridge
  418: mode.  Two disks are provided, and they are both writable; the bits
  419: are stored in files and Xen attaches them to a vnd(4) device in the
  420: dom0 on domain creation.  The system treats xbd0 as the boot device
  421: without needing explicit configuration.
  423: There is not a type line; that implicitly defines a pv domU.
  424: Otherwise, one sets type to the lower-case version of the domU type in
  425: the table above; see later sections.
  427: By convention, domain config files are kept in `/usr/pkg/etc/xen`.  Note
  428: that "xl create" takes the name of a config file, while other commands
  429: take the name of a domain.
  431: Examples of commands:
  433: [[!template id=programlisting text="""
  434: xl create /usr/pkg/etc/xen/foo
  435: xl console domU-id
  436: xl create -c /usr/pkg/etc/xen/foo
  437: xl shutdown domU-id
  438: xl list
  439: """]]
  441: Typing `^]` will exit the console session.  Shutting down a domain is
  442: equivalent to pushing the power button; a NetBSD domU will receive a
  443: power-press event and do a clean shutdown.  Shutting down the dom0
  444: will trigger controlled shutdowns of all configured domUs.
  446: ## CPU and memory
  448: A domain is provided with some number of vcpus; any domain can have up
  449: to the number of CPUs seen by the hypervisor. For a domU, it is
  450: controlled from the config file by the "vcpus = N" directive.  It is
  451: normal to overcommit vcpus; a 4-core machine machine might well provide 4
  452: vcpus to each domU.  One might also configure fewer vcpus for a domU.
  454: A domain is provided with memory; this is controlled in the config
  455: file by "memory = N" (in megabytes).  In the straightforward case, the
  456: sum of the the memory allocated to the dom0 and all domUs must be less
  457: than the available memory.
  459: ## Balloon driver
  461: Xen provides a `balloon` driver, which can be used to let domains use
  462: more memory temporarily.
  464: \todo Explain how to set up a aystem to use the balloon scheme in a
  465: useful manner.
  467: ## Virtual disks
  469: In domU config files, the disks are defined as a sequence of 3-tuples:
  471:  * The first element is "method:/path/to/disk". Common methods are
  472:    "file:" for a file-backed vnd, and "phy:" for something that is already
  473:    a device, such as an LVM logical volume.
  475:  * The second element is an artifact of how virtual disks are passed to
  476:    Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
  477:    are given a device name to associate with the disk, and values like
  478:    "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
  479:    as xbd0, the second as xbd1, and so on.  However, xl demands a
  480:    second argument.  The name given is converted to a major/minor by
  481:    calling stat(2) on the name in /dev and this is passed to the domU.
  482:    In the general case, the dom0 and domU can be different operating
  483:    systems, and it is an unwarranted assumption that they have consistent
  484:    numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
  485:    as both dom0 and domU, using values of 0x0 for the first disk and 0x1
  486:    for the second works fine and avoids this issue.  For a GNU/Linux
  487:    guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
  488:    /dev/hda1.
  490:  * The third element is "w" for writable disks, and "r" for read-only
  491:    disks.
  493: Example:
  494: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
  495: disk = [ 'file:/n0/xen/foo-wd0,0x0,w' ]
  496: """]]
  498: Note that NetBSD by default creates only vnd[0123].  If you need more
  499: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
  500: dom0.
  502: ## Virtual Networking
  504: Xen provides virtual Ethernets, each of which connects the dom0 and a
  505: domU.  For each virtual network, there is an interface "xvifN.M" in
  506: the dom0, and a matching interface xennetM (NetBSD name) in domU index N.
  507: The interfaces behave as if there is an Ethernet with two
  508: adapters connected.  From this primitive, one can construct various
  509: configurations.  We focus on two common and useful cases for which
  510: there are existing scripts: bridging and NAT.
  512: With bridging (in the example above), the domU perceives itself to be
  513: on the same network as the dom0.  For server virtualization, this is
  514: usually best.  Bridging is accomplished by creating a bridge(4) device
  515: and adding the dom0's physical interface and the various xvifN.0
  516: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
  517: config file.  The bridge must be set up already in the dom0; an
  518: example /etc/ifconfig.bridge0 is:
  520: [[!template id=filecontent name="/etc/ifconfig.bridge0" text="""
  521: create
  522: up
  523: !brconfig bridge0 add wm0
  524: """]]
  526: With NAT, the domU perceives itself to be behind a NAT running on the
  527: dom0.  This is often appropriate when running Xen on a workstation.
  528: TODO: NAT appears to be configured by "vif = [ '' ]".
  530: The MAC address specified is the one used for the interface in the new
  531: domain.  The interface in dom0 will use this address XOR'd with
  532: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
  534: ## Starting domains automatically
  536: To start domains `domU-netbsd` and `domU-linux` at boot and shut them
  537: down cleanly on dom0 shutdown, add the following in rc.conf:
  539: [[!template id=filecontent name="/etc/rc.conf" text="""
  540: xendomains="domU-netbsd domU-linux"
  541: """]]
  543: # domU setup for specific systems
  545: Creating domUs is almost entirely independent of operating system.  We
  546: have already presented the basics of config files in the previous system.
  548: Of course, this section presumes that you have a working dom0.
  550: ## Creating a NetBSD PV domU
  552: See the earlier config file, and adjust memory.  Decide on how much
  553: storage you will provide, and prepare it (file or LVM).
  555: While the kernel will be obtained from the dom0 file system, the same
  556: file should be present in the domU as /netbsd so that tools like
  557: savecore(8) can work.   (This is helpful but not necessary.)
  559: The kernel must be specifically built for Xen, to use PV interfacesas
  560: a domU.  NetBSD release builds provide the following kernels:
  562:         i386 XEN3PAE_DOMU
  563:         amd64 XEN3_DOMU
  565: This will boot NetBSD, but this is not that useful if the disk is
  566: empty.  One approach is to unpack sets onto the disk outside of Xen
  567: (by mounting it, just as you would prepare a physical disk for a
  568: system you can't run the installer on).
  570: A second approach is to run an INSTALL kernel, which has a miniroot
  571: and can load sets from the network.  To do this, copy the INSTALL
  572: kernel to / and change the kernel line in the config file to:
  574:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  576: Then, start the domain as "xl create -c configfile".
  578: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
  579: line should be used in the config file.
  581:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  583: After booting the domain, the option to install via CDROM may be
  584: selected.  The CDROM device should be changed to `xbd1d`.
  586: Once done installing, "halt -p" the new domain (don't reboot or halt:
  587: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
  588: config file), switch the config file back to the XEN3_DOMU kernel,
  589: and start the new domain again. Now it should be able to use "root on
  590: xbd0a" and you should have a functional NetBSD domU.
  592: TODO: check if this is still accurate.
  593: When the new domain is booting you'll see some warnings about *wscons*
  594: and the pseudo-terminals. These can be fixed by editing the files
  595: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  596: `/etc/ttys`, except *console*, like this:
  598:     console "/usr/libexec/getty Pc"         vt100   on secure
  599:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  600:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  601:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  602:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  604: Finally, all screens must be commented out from `/etc/wscons.conf`.
  606: One should also run `powerd` in a domU, but this should not need
  607: configuring.  With powerd, the domain will run a controlled shutdown
  608: if `xl shutdown -R` or `xl shutdown -H` is used on the dom0, via
  609: receiving a synthetic `power button pressed` signal.  In 9 and
  610: current, `powerd` is run by default under Xen kernels (or if ACPI is
  611: present), and it can be added to rc.conf if not.
  613: It is not strictly necessary to have a kernel (as /netbsd) in the domU
  614: file system.  However, various programs (e.g. netstat) will use that
  615: kernel to look up symbols to read from kernel virtual memory.  If
  616: /netbsd is not the running kernel, those lookups will fail.  (This is
  617: not really a Xen-specific issue, but because the domU kernel is
  618: obtained from the dom0, it is far more likely to be out of sync or
  619: missing with Xen.)
  621: Note that NetBSD by default creates only xbd[0123].  If you need more
  622: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
  624: ## Creating a Linux PV domU
  626: Creating unprivileged Linux domains isn't much different from
  627: unprivileged NetBSD domains, but there are some details to know.
  629: First, the second parameter passed to the disk declaration (the '0x1' in
  630: the example below)
  632:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  634: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  635: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
  636: So, hda1 which has major 3 and minor 1 on a Linux system will have
  637: device number 0x301.  Alternatively, devices names can be used (hda,
  638: hdb, ...)  as xentools has a table to map these names to devices
  639: numbers.  To export a partition to a Linux guest we can use:
  641:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
  642:         root = "/dev/hda1 ro"
  644: and it will appear as /dev/hda on the Linux system, and be used as root
  645: partition.
  647: To install the Linux system on the partition to be exported to the
  648: guest domain, the following method can be used: install
  649: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
  650: that will be the root partition of your Linux domain, and mount it.
  651: Then copy the files from a working Linux system, make adjustments in
  652: `/etc` (fstab, network config).  It should also be possible to extract
  653: binary packages such as .rpm or .deb directly to the mounted partition
  654: using the appropriate tool, possibly running under NetBSD's Linux
  655: emulation.  Once the file system has been populated, umount it.  If
  656: desirable, the file system can be converted to ext3 using tune2fs -j.
  657: It should now be possible to boot the Linux guest domain, using one of
  658: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
  660: To get the Linux console right, you need to add:
  662:     extra = "xencons=tty1"
  664: to your configuration since not all Linux distributions auto-attach a
  665: tty to the xen console.
  667: ## Creating a NetBSD HVM domU
  669: Use type='hvm', probably.  Use a GENERIC kernel within the disk image.
  671: ## Creating a NetBSD PVH domU
  673: This only works with a current kernel in the domU.
  675: Use type='pvh'.  Probably, use a GENERIC kernel within the disk image,
  676: which in current has PV support.
  678: \todo Verify.
  680: \todo Verify if one can have current PVH domU on a 9 dom0.
  682: ## Creating a Solaris domU
  684: See possibly outdated
  685: [Solaris domU instructions](/ports/xen/howto-solaris/).
  687: ## PCI passthrough: Using PCI devices in guest domains
  689: NB: PCI passthrough only works on some Xen versions and as of 2020 it
  690: is not clear that it works on any version in pkgsrc.  \todo Reports
  691: confirming or denying this notion should be sent to port-xen@.
  693: The dom0 can give other domains access to selected PCI
  694: devices. This can allow, for example, a non-privileged domain to have
  695: access to a physical network interface or disk controller.  However,
  696: keep in mind that giving a domain access to a PCI device most likely
  697: will give the domain read/write access to the whole physical memory,
  698: as PCs don't have an IOMMU to restrict memory access to DMA-capable
  699: device.  Also, it's not possible to export ISA devices to non-dom0
  700: domains, which means that the primary VGA adapter can't be exported.
  701: A guest domain trying to access the VGA registers will panic.
  703: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
  704: not been ported to later versions at this time.
  706: For a PCI device to be exported to a domU, is has to be attached to
  707: the "pciback" driver in dom0.  Devices passed to the dom0 via the
  708: pciback.hide boot parameter will attach to "pciback" instead of the
  709: usual driver.  The list of devices is specified as "(bus:dev.func)",
  710: where bus and dev are 2-digit hexadecimal numbers, and func a
  711: single-digit number:
  713:         pciback.hide=(00:0a.0)(00:06.0)
  715: pciback devices should show up in the dom0's boot messages, and the
  716: devices should be listed in the `/kern/xen/pci` directory.
  718: PCI devices to be exported to a domU are listed in the "pci" array of
  719: the domU's config file, with the format "0000:bus:dev.func".
  721:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
  723: In the domU an "xpci" device will show up, to which one or more pci
  724: buses will attach.  Then the PCI drivers will attach to PCI buses as
  725: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
  726: or any PCI drivers built in by default; you have to build your own
  727: kernel to use PCI devices in a domU.  Here's a kernel config example;
  728: note that only the "xpci" lines are unusual.
  730:         include         "arch/i386/conf/XEN3_DOMU"
  732:         # Add support for PCI buses to the XEN3_DOMU kernel
  733:         xpci* at xenbus ?
  734:         pci* at xpci ?
  736:         # PCI USB controllers
  737:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
  739:         # USB bus support
  740:         usb*    at uhci?
  742:         # USB Hubs
  743:         uhub*   at usb?
  744:         uhub*   at uhub? port ? configuration ? interface ?
  746:         # USB Mass Storage
  747:         umass*  at uhub? port ? configuration ? interface ?
  748:         wd*     at umass?
  749:         # SCSI controllers
  750:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
  752:         # SCSI bus support (for both ahc and umass)
  753:         scsibus* at scsi?
  755:         # SCSI devices
  756:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
  757:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
  760: # Miscellaneous Information
  762: ## Nesting under Linux KVM
  764: It is possible to run Xen and a NetBSD dom0 under Linux KVM.  One
  765: can enable virtio in the dom0 for greater speed.
  767: ## Nesting under qemu
  769: It is possible to run Xen and a NetBSD dom0 under qemu on NetBSD, and
  770: also with nvmm.
  771: \todo Check this.
  773: ## Other nesting
  775: In theory, any full emulation should be able to run Xen and a NetBSD
  776: dom0.  The HOWTO does not currently have information about Xen XVM
  777: mode, Virtualbox, etc.
  779: ## NetBSD 5 as domU
  781: [NetBSD 5 is known to panic.](
  782: (However, NetBSD 5 systems should be updated to a supported version.)
  784: # NetBSD as a domU in a VPS
  786: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
  787: hardware.  This section explains how to deal with Xen in a domU as a
  788: virtual private server where you do not control or have access to the
  789: dom0.  This is not intended to be an exhaustive list of VPS providers;
  790: only a few are mentioned that specifically support NetBSD.
  792: VPS operators provide varying degrees of access and mechanisms for
  793: configuration.  The big issue is usually how one controls which kernel
  794: is booted, because the kernel is nominally in the dom0 file system (to
  795: which VPS users do not normally have access).  A second issue is how
  796: to install NetBSD.
  797: A VPS user may want to compile a kernel for security updates, to run
  798: npf, run IPsec, or any other reason why someone would want to change
  799: their kernel.
  801: One approach is to have an administrative interface to upload a kernel,
  802: or to select from a prepopulated list.  Other approaches are pygrub
  803: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
  804: kernel from the domU file system.  This is closer to a regular physical
  805: computer, where someone who controls a machine can replace the kernel.
  807: A second issue is multiple CPUs.  With NetBSD 6, domUs support
  808: multiple vcpus, and it is typical for VPS providers to enable multiple
  809: CPUs for NetBSD domUs.
  811: ## Complexities due to Xen changes
  813: Xen has many security advisories and people running Xen systems make
  814: different choices.
  816: ### stub domains
  818: Some (Linux) dom0 systems use something called "stub domains" to
  819: isolate qemu from the dom0 system, as a security and reliabilty
  820: mechanism when running HVM domUs.  Somehow, NetBSD's GENERIC kernel
  821: ends up using PIO for disks rather than DMA.  Of course, all of this
  822: is emulated, but emulated PIO is unusably slow.  This problem is not
  823: currently understood.
  825: ### Grant tables
  827: There are multiple versions of using grant tables, and some security
  828: advisories have suggested disabling some versions.  NetBSD through 9
  829: uses version 1 and NetBSD-current uses version 2.  This can lead to
  830: "NetBSD current doesn't run on hosting provider X" situations.
  832: \todo Explain better.
  834: ## Boot methods
  836: ### pvgrub
  838: pvgrub is a version of grub that uses PV operations instead of BIOS
  839: calls.  It is booted from the dom0 as the domU kernel, and then reads
  840: /grub/menu.lst and loads a kernel from the domU file system.
  842: It appears that [grub's FFS
  843: code](
  844: does not support all aspects of modern FFS, but there are also reports
  845: that FFSv2 works fine.
  847: ### pygrub
  849: As of 2014, pygrub seems to be of mostly historical interest.  As of
  850: 2021, the section should perhaps be outright deleted.
  852: pygrub runs in the dom0 and looks into the domU file system.  This
  853: implies that the domU must have a kernel in a file system in a format
  854: known to pygrub.
  856: pygrub doesn't seem to work to load Linux images under NetBSD dom0,
  857: and is inherently less secure than pvgrub due to running inside
  858: dom0. For both these reasons, pygrub should not be used, and is only
  859: still present so that historical DomU images using it still work.
  861: ## Specific Providers
  863: The intent is to list providers only if they document support for
  864: running NetBSD, and to point to their resources briefly.
  866: ###
  868: [Panix]( provides NetBSD as an OS option.  See
  869: for some information.
  870: Users can use pvgrub.  Panix reports that pvgrub works with FFsv2 with
  871: 16K/2K and 32K/4K block/frag sizes (and hence with defaults from
  872: "newfs -O 2").  See [Panix's pvgrub
  873: page]( which describes how to
  874: boot NetBSD.
  876: ###
  878: []( provides released versions of
  879: NetBSD/amd64 as installation options.  Users can use pvgrub to boot
  880: their own kernel, and a small FAT32 /boot is encouraged.  See the
  881: [ NetBSD
  882: HOWTO](
  883: (which is in need of updating).
  885: ### Amazon
  887: See the [Amazon EC2 page](/amazon_ec2/).

CVSweb for NetBSD wikisrc <> software: FreeBSD-CVSweb