File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.200: download - view: text, annotated - select for diffs
Sun Mar 7 01:06:00 2021 UTC (17 months, 1 week ago) by gdt
Branches: MAIN
CVS tags: HEAD
xen: explain type line in config

    1: [[!meta title="Xen 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 HOWTO presumes a basic familiarity with the Xen system
   12: architecture, with installing NetBSD on amd64 hardware, and with
   13: installing software from pkgsrc.  See also the [Xen
   14: website](
   16: [[!toc]]
   18: # Overview
   20: The basic concept of Xen is that the hypervisor (xenkernel) runs on
   21: the hardware, and runs a privileged domain ("dom0") that can access
   22: disks/networking/etc.  One then runs additonal unprivileged domains
   23: (each a "domU"), presumably to do something useful.
   25: This HOWTO addresses how to run a NetBSD dom0 (and hence also build
   26: xen itself).  It also addresses how to run domUs in that environment,
   27: and how to deal with having a domU in a Xen environment run by someone
   28: else and/or not running NetBSD.
   30: There are many choices one can make; the HOWTO recommends the standard
   31: approach and limits discussion of alternatives in many cases.
   33: ## Guest Styles
   35: Xen supports different styles of guests.
   37: [[!table data="""
   38: Style of guest	|Supported by NetBSD
   39: PV		|Yes (dom0, domU)
   40: HVM		|Yes (domU)
   41: PVHVM		|current-only (domU)
   42: PVH		|current-only (domU, dom0 not yet)
   43: """]]
   45: In Para-Virtualized (PV) mode, the guest OS does not attempt to access
   46: hardware directly, but instead makes hypercalls to the hypervisor; PV
   47: guests must be specifically coded for Xen.
   48: See [PV](\)).
   50: In HVM mode, no guest modification is required; however, hardware
   51: support is required, such as VT-x on Intel CPUs and SVM on AMD CPUs.
   52: The dom0 runs qemu to emulate hardware.  It is therefore non-sensical
   53: to have an HVM dom0.
   55: In PVHVM mode, the guest runs as HVM, but additionally can use PV
   56: drivers for efficiency.  Therefore it is non-sensical for to have a
   57: PVHVM dom0.  See [PV on HVM](
   59: There have been two PVH modes: original PVH and PVHv2.  Original PVH
   60: was based on PV mode and is no longer relevant at all.  PVHv2 is
   61: basically lightweight HVM with PV drivers.  A critical feature of it
   62: is that qemu is not needed; the hypervisor can do the emulation that
   63: is required.  Thus, a dom0 can be PVHv2.
   64: The source code uses PVH and config files use pvh; this refers to PVHv2.
   65: See [PVH(v2)](\)_Domu).
   67: At system boot, the dom0 kernel is loaded as a module with Xen as the kernel.
   68: The dom0 can start one or more domUs.  (Booting is explained in detail
   69: in the dom0 section.)
   71: ## CPU Architecture
   73: Xen runs on x86_64 hardware (the NetBSD amd64 port).
   75: There is a concept of Xen running on ARM, but there are no reports of this working with NetBSD.
   77: The dom0 system should be amd64.  (Instructions for i386PAE dom0 have been removed from the HOWTO.)
   79: The domU can be i386PAE or amd64.
   80: i386PAE at one point was considered as [faster]( than amd64.
   82: ## Xen Versions
   84: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
   85: xenkernel and xentools.  We will refer only to the kernel versions,
   86: but note that both packages must be installed together and must have
   87: matching versions.
   89: Versions available in pkgsrc:
   91: [[!table data="""
   92: Xen Version	|Package Name	|Xen CPU Support	|EOL'ed By Upstream
   93: 4.11		|xenkernel411	|x86_64			|No
   94: 4.13		|xenkernel413	|x86_64			|No
   95: """]]
   97: See also the [Xen Security Advisory page](
   99: Older Xen had a python-based management tool called xm, now replaced
  100: by xl.
  102: ## NetBSD versions
  104: Xen has been supported in NetBSD for a long time, at least since 2005.
  105: Initially Xen was PV only.
  107: NetBSD 8 and up support PV and HVM modes.
  109: Support for PVHVM and PVH is available only in NetBSD-current.
  111: NetBSD up to and including NetBSD 9 as a dom0 does not run SMP,
  112: because some drivers are not yet safe for this.  NetBSD-current
  113: supports SMP in dom0.
  115: NetBSD, when run as a domU, can and does typically run SMP.
  117: Note that while Xen 4.13 is current, the kernel support is still
  118: called XEN3, because the hypercall interface has not changed
  119: significantly.
  121: # Creating a NetBSD dom0
  123: In order to install a NetBSD as a dom0, one first installs a normal
  124: NetBSD system, and then pivot the install to a dom0 install by
  125: changing the kernel and boot configuration.
  127: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
  128: 512M it worked reliably.  This does not make sense, but if you see
  129: "not ELF" after Xen boots, try increasing dom0 RAM.
  131: ## Installation of NetBSD
  133: [Install NetBSD/amd64](/guide/inst/) just as you would if you were not
  134: using Xen.  Therefore, use the most recent release, or a build from
  135: the most recent stable branch.  Alternatively, use -current, being
  136: mindful of all the usual caveats of lower stability of current, and
  137: likely a bit more so.
  139: ## Installation of Xen
  141: ### Building Xen
  143: Use the most recent version of Xen in pkgsrc, unless the DESCR says that it is not suitable.
  144: Therefore, choose 4.13.
  145: In the dom0, install xenkernel413 and xentools413 from pkgsrc.
  147: Once this is done, copy the Xen kernel from where pkgsrc puts it to
  148: where the boot process will be able to find it:
  150: [[!template id=programlisting text="""
  151: # cp -p /usr/pkg/xen413-kernel/xen.gz /
  152: """]]
  154: Then, place a NetBSD XEN3_DOM0 kernel in the `/` directory. Such
  155: kernel can either be taken from a local release run, compiled
  156: manually, or downloaded from the NetBSD FTP, for example at:
  158: [[!template id=programlisting text="""
  160: """]]
  162: ### Configuring booting
  164: Read boot.cfg(8) carefully.  Add lines to /boot.cfg to boot Xen,
  165: adjusting for your root filesystem:
  167: [[!template id=filecontent name="/boot.cfg" text="""
  168: menu=Xen:load /netbsd-XEN3_DOM0.gz root=wd0a console=pc;multiboot /xen.gz dom0_mem=512M
  169: menu=Xen single user:load /netbsd-XEN3_DOM0.gz root=wd0a console=pc -s;multiboot /xen.gz dom0_mem=512M
  170: """]]
  172: This specifies that the dom0 should have 512MB of ram, leaving the rest
  173: to be allocated for domUs.
  175: NB: This says add, not replace, so that you will be able to more
  176: easily boot a NetBSD kernel without Xen.  Once Xen boots ok, you may
  177: want to set it as default.  It is highly likely that you will have
  178: trouble at some point, and keeping an up-to-date GENERIC for use in
  179: fixing problems is the standard prudent approach.
  181: \todo Explain why rndseed is not set with Xen as part of the dom0
  182: subconfiguration.
  184: Note that you are likely to have to set root= because the boot device
  185: from /boot is not passed via Xen to the dom0 kernel.  With one disk,
  186: it will work, but e.g. plugging in USB disk to a machine with root on
  187: wd0a causes boot to fail.
  189: Beware that userconf statements must be attached to the dom0 load, and
  190: may not be at top-level, because then they would try to configure the
  191: hypervisor, if there is a way to pass them via multiboot .  It appears
  192: that adding `userconf=pckbc` to `/boot.cfg` causes Xen to crash very
  193: early with a heap overflow.
  195: ### Console selection
  197: See boot_console(8).  Understand that you should start from a place of
  198: having console setup correct for booting GENERIC before trying to
  199: configure Xen.
  201: Generally for GENERIC, one sets the console in bootxx_ffsv1 or
  202: equivalent, and this is passed on to /boot (where one typically does
  203: not set the console).  This configuration of bootxx_ffsv1 should also
  204: be in place for Xen systems, to allow seeing messages from /boot and
  205: use of a keyboard to select a line from the menu.  And, one should
  206: have a working boot path to GENERIC for rescue situations.
  208: With GENERIC, the boot options are passed on to /netbsd, but there is
  209: currently no mechanism to pass these via multiboot to the hypervisor.
  210: Thus, in addition to configuring the console in the boot blocks, one
  211: must also configure it for Xen.
  213: By default, the hypervisor (Xen itself) will use some sort of vga
  214: device as the console, much like GENERIC uses by default.  The vga
  215: console is relinquished at the conclusion of hypervisor boot, before
  216: the dom0 is started.  Xen when using a vga console does not process
  217: console input.
  219: The hypervisor can be configured to use a serial port console, e.g.
  220: [[!template id=filecontent name="/boot.cfg" text="""
  221: menu=Xen:losad /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
  222: """]]
  223: This example uses the first serial port (Xen counts from 1; this is
  224: what NetBSD would call com0), and sets speed and parity.  (The dom0 is
  225: then configured to use the same serial port in this example.)
  227: With the hypervisor configured for a serial console, it can get input,
  228: and there is a notion of passing this input to the dom0.  \todo
  229: Explain why, if Xen has a serial console, the dom0 console is
  230: typically also configured to open that same serial port, instead of
  231: getting the passthrough input via the xen console.
  233: One also configures the console for the dom0.  While one might expect
  234: console=pc to be default, following behavior of GENERIC, a hasty read
  235: of the code suggests there is no default and booting without a
  236: selected console might lead to a panic.  Also, there is merit in
  237: explicit configuration.  Therefore the standard approach is to place
  238: console=pc as part of the load statement for the dom0 kernel, or
  239: alternatively console=com0.
  241: The NetBSD dom0 kernel will attach xencons(4) (the man page does not
  242: exist), but this is not used as a console.  It is used to obtain the
  243: messages from the hypervisor's console; run `xl dmesg` to see them.
  245: ### Tuning
  247: In an attempt to add performance, one can also add `dom0_max_vcpus=1 dom0_vcpus_pin`,
  248: to force only one vcpu to be provided (since NetBSD dom0 can't use
  249: more) and to pin that vcpu to a physical CPU. Xen has
  250: [many boot options](,
  251: and other than dom0 memory and max_vcpus, they are generally not
  252: necessary.
  253: \todo Revisit this advice with current.
  254: \todo Explain if anyone has ever actually measured that this helps.
  256: ### rc.conf
  258: Ensure that the boot scripts installed in
  259: `/usr/pkg/share/examples/rc.d` are in `/etc/rc.d`, either because you
  260: have `PKG_RCD_SCRIPTS=yes`, or manually.  (This is not special to Xen,
  261: but a normal part of pkgsrc usage.)
  263: Set `xencommons=YES` in rc.conf:
  265: [[!template id=filecontent name="/etc/rc.conf" text="""
  266: xencommons=YES
  267: """]]
  269: \todo Recommend for/against xen-watchdog.
  271: ### Testing
  273: Now, reboot so that you are running a DOM0 kernel under Xen, rather
  274: than GENERIC without Xen.
  276: Once the reboot is done, use `xl` to inspect Xen's boot messages,
  277: available resources, and running domains.  For example:
  279: [[!template id=programlisting text="""
  280: # xl dmesg
  281: ... xen's boot info ...
  282: # xl info
  283: ... available memory, etc ...
  284: # xl list
  285: Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  286: Domain-0           0       64    0  r----     58.1
  287: """]]
  289: Xen logs will be in /var/log/xen.
  291: ### Issues with xencommons
  293: `xencommons` starts `xenstored`, which stores data on behalf of dom0 and
  294: domUs.  It does not currently work to stop and start xenstored.
  295: Certainly all domUs should be shutdown first, following the sort order
  296: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
  297: and is not notified when xenstored exits, leading to not recreating
  298: the state when the new xenstored starts.  Until there's a mechanism to
  299: make this work, one should not expect to be able to restart xenstored
  300: (and thus xencommons).  There is currently no reason to expect that
  301: this will get fixed any time soon.
  302: \todo Confirm if this is still true in 2020.
  304: ## Xen-specific NetBSD issues
  306: There are (at least) two additional things different about NetBSD as a
  307: dom0 kernel compared to hardware.
  309: One is that through NetBSD 9 the module ABI is different because some
  310: of the #defines change, so there are separate sets of modules in
  311: /stand.  In NetBSD-current, there is only one set of modules.
  313: The other difference is that XEN3_DOM0 does not have exactly the same
  314: options as GENERIC.  While it is debatable whether or not this is a
  315: bug, users should be aware of this and can simply add missing config
  316: items if desired.
  318: Finally, there have been occasional reports of trouble with X11
  319: servers in NetBSD as a dom0.
  321: ## Updating Xen in a dom0
  323: Basically, update the xenkernel and xentools packages and copy the new
  324: Xen kernel into place, and reboot.  This procedure should be usable to
  325: update to a new Xen release, but the reader is reminded that having a
  326: non-Xen boot methods was recommended earlier.
  328: ## Updating NetBSD in a dom0
  330: This is just like updating NetBSD on bare hardware, assuming the new
  331: version supports the version of Xen you are running.  Generally, one
  332: replaces the kernel and reboots, and then overlays userland binaries
  333: and adjusts `/etc`.
  335: Note that one should update both the non-Xen kernel typically used for
  336: rescue purposes, as well as the DOM0 kernel used with Xen.
  338: ## anita (for testing NetBSD)
  340: With a NetBSD dom0, even without any domUs, one should be able to run
  341: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
  342: root, because anita must create a domU):
  344: [[!template id=programlisting text="""
  345: anita --vmm=xl test file:///usr/obj/i386/
  346: """]]
  348: # Unprivileged domains (domU)
  350: This section describes general concepts about domUs.  It does not
  351: address specific domU operating systems or how to install them.  The
  352: config files for domUs are typically in `/usr/pkg/etc/xen`, and are
  353: typically named so that the file name, domU name and the domU's host
  354: name match.
  356: The domU is provided with CPU and memory by Xen, configured by the
  357: dom0.  The domU is provided with disk and network by the dom0,
  358: mediated by Xen, and configured in the dom0.
  360: Entropy in domUs can be an issue; physical disks and network are on
  361: the dom0.  NetBSD's /dev/random system works, but is often challenged.
  363: ## Config files
  365: See /usr/pkg/share/examples/xen/xlexample* for a very small number of
  366: examples for running GNU/Linux.
  368: The following is an example minimal domain configuration file. The domU
  369: serves as a network file server.
  371: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
  372: name = "domU-id"
  373: kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
  374: memory = 1024
  375: vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
  376: disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
  377:          'file:/n0/xen/foo-wd1,0x1,w' ]
  378: """]]
  380: The domain will have name given in the `name` setting.  The kernel has the
  381: host/domU name in it, so that on the dom0 one can update the various
  382: domUs independently.  The `vif` line causes an interface to be provided,
  383: with a specific mac address (do not reuse MAC addresses!), in bridge
  384: mode.  Two disks are provided, and they are both writable; the bits
  385: are stored in files and Xen attaches them to a vnd(4) device in the
  386: dom0 on domain creation.  The system treats xbd0 as the boot device
  387: without needing explicit configuration.
  389: There is not type line; that implicitly defines a pv domU.  Otherwise,
  390: one sets type to the lower-case version of the domU type in the table
  391: above, e.g.  `type = "hvm"`.
  393: By convention, domain config files are kept in `/usr/pkg/etc/xen`.  Note
  394: that "xl create" takes the name of a config file, while other commands
  395: take the name of a domain.
  397: Examples of commands:
  399: [[!template id=programlisting text="""
  400: xl create /usr/pkg/etc/xen/foo
  401: xl console domU-id
  402: xl create -c /usr/pkg/etc/xen/foo
  403: xl shutdown domU-id
  404: xl list
  405: """]]
  407: Typing `^]` will exit the console session.  Shutting down a domain is
  408: equivalent to pushing the power button; a NetBSD domU will receive a
  409: power-press event and do a clean shutdown.  Shutting down the dom0
  410: will trigger controlled shutdowns of all configured domUs.
  412: ## CPU and memory
  414: A domain is provided with some number of vcpus, up to the number
  415: of CPUs seen by the hypervisor. For a domU, it is controlled
  416: from the config file by the "vcpus = N" directive.
  418: A domain is provided with memory; this is controlled in the config
  419: file by "memory = N" (in megabytes).  In the straightforward case, the
  420: sum of the the memory allocated to the dom0 and all domUs must be less
  421: than the available memory.
  423: ## Balloon driver
  425: Xen provides a `balloon` driver, which can be used to let domains use
  426: more memory temporarily.
  428: \todo Explain how to set up a aystem to use the balloon scheme in a
  429: useful manner.
  431: ## Virtual disks
  433: In domU config files, the disks are defined as a sequence of 3-tuples:
  435:  * The first element is "method:/path/to/disk". Common methods are
  436:    "file:" for a file-backed vnd, and "phy:" for something that is already
  437:    a device, such as an LVM logical volume.
  439:  * The second element is an artifact of how virtual disks are passed to
  440:    Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
  441:    are given a device name to associate with the disk, and values like
  442:    "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
  443:    as xbd0, the second as xbd1, and so on.  However, xl demands a
  444:    second argument.  The name given is converted to a major/minor by
  445:    calling stat(2) on the name in /dev and this is passed to the domU.
  446:    In the general case, the dom0 and domU can be different operating
  447:    systems, and it is an unwarranted assumption that they have consistent
  448:    numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
  449:    as both dom0 and domU, using values of 0x0 for the first disk and 0x1
  450:    for the second works fine and avoids this issue.  For a GNU/Linux
  451:    guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
  452:    /dev/hda1.
  454:  * The third element is "w" for writable disks, and "r" for read-only
  455:    disks.
  457: Example:
  458: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
  459: disk = [ 'file:/n0/xen/foo-wd0,0x0,w' ]
  460: """]]
  462: Note that NetBSD by default creates only vnd[0123].  If you need more
  463: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
  464: dom0.
  466: ## Virtual Networking
  468: Xen provides virtual Ethernets, each of which connects the dom0 and a
  469: domU.  For each virtual network, there is an interface "xvifN.M" in
  470: the dom0, and a matching interface xennetM (NetBSD name) in domU index N.
  471: The interfaces behave as if there is an Ethernet with two
  472: adapters connected.  From this primitive, one can construct various
  473: configurations.  We focus on two common and useful cases for which
  474: there are existing scripts: bridging and NAT.
  476: With bridging (in the example above), the domU perceives itself to be
  477: on the same network as the dom0.  For server virtualization, this is
  478: usually best.  Bridging is accomplished by creating a bridge(4) device
  479: and adding the dom0's physical interface and the various xvifN.0
  480: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
  481: config file.  The bridge must be set up already in the dom0; an
  482: example /etc/ifconfig.bridge0 is:
  484: [[!template id=filecontent name="/etc/ifconfig.bridge0" text="""
  485: create
  486: up
  487: !brconfig bridge0 add wm0
  488: """]]
  490: With NAT, the domU perceives itself to be behind a NAT running on the
  491: dom0.  This is often appropriate when running Xen on a workstation.
  492: TODO: NAT appears to be configured by "vif = [ '' ]".
  494: The MAC address specified is the one used for the interface in the new
  495: domain.  The interface in dom0 will use this address XOR'd with
  496: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
  498: ## Starting domains automatically
  500: To start domains `domU-netbsd` and `domU-linux` at boot and shut them
  501: down cleanly on dom0 shutdown, add the following in rc.conf:
  503: [[!template id=filecontent name="/etc/rc.conf" text="""
  504: xendomains="domU-netbsd domU-linux"
  505: """]]
  507: # domU setup for specific systems
  509: Creating domUs is almost entirely independent of operating system.  We
  510: have already presented the basics of config files in the previous system.
  512: Of course, this section presumes that you have a working dom0.
  514: ## Creating a NetBSD PV domU
  516: See the earlier config file, and adjust memory.  Decide on how much
  517: storage you will provide, and prepare it (file or LVM).
  519: While the kernel will be obtained from the dom0 file system, the same
  520: file should be present in the domU as /netbsd so that tools like
  521: savecore(8) can work.   (This is helpful but not necessary.)
  523: The kernel must be specifically built for Xen, to use PV interfacesas
  524: a domU.  NetBSD release builds provide the following kernels:
  526:         i386 XEN3PAE_DOMU
  527:         amd64 XEN3_DOMU
  529: This will boot NetBSD, but this is not that useful if the disk is
  530: empty.  One approach is to unpack sets onto the disk outside of Xen
  531: (by mounting it, just as you would prepare a physical disk for a
  532: system you can't run the installer on).
  534: A second approach is to run an INSTALL kernel, which has a miniroot
  535: and can load sets from the network.  To do this, copy the INSTALL
  536: kernel to / and change the kernel line in the config file to:
  538:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  540: Then, start the domain as "xl create -c configfile".
  542: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
  543: line should be used in the config file.
  545:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  547: After booting the domain, the option to install via CDROM may be
  548: selected.  The CDROM device should be changed to `xbd1d`.
  550: Once done installing, "halt -p" the new domain (don't reboot or halt:
  551: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
  552: config file), switch the config file back to the XEN3_DOMU kernel,
  553: and start the new domain again. Now it should be able to use "root on
  554: xbd0a" and you should have a functional NetBSD domU.
  556: TODO: check if this is still accurate.
  557: When the new domain is booting you'll see some warnings about *wscons*
  558: and the pseudo-terminals. These can be fixed by editing the files
  559: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  560: `/etc/ttys`, except *console*, like this:
  562:     console "/usr/libexec/getty Pc"         vt100   on secure
  563:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  564:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  565:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  566:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  568: Finally, all screens must be commented out from `/etc/wscons.conf`.
  570: One should also run `powerd` in a domU, but this should not need
  571: configuring.  With powerd, the domain will run a controlled shutdown
  572: if `xl shutdown -R` or `xl shutdown -H` is used on the dom0, via
  573: receiving a synthetic `power button pressed` signal.  In 9 and
  574: current, `powerd` is run by default under Xen kernels (or if ACPI is
  575: present), and it can be added to rc.conf if not.
  577: It is not strictly necessary to have a kernel (as /netbsd) in the domU
  578: file system.  However, various programs (e.g. netstat) will use that
  579: kernel to look up symbols to read from kernel virtual memory.  If
  580: /netbsd is not the running kernel, those lookups will fail.  (This is
  581: not really a Xen-specific issue, but because the domU kernel is
  582: obtained from the dom0, it is far more likely to be out of sync or
  583: missing with Xen.)
  585: Note that NetBSD by default creates only xbd[0123].  If you need more
  586: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
  588: ## Creating a Linux PV domU
  590: Creating unprivileged Linux domains isn't much different from
  591: unprivileged NetBSD domains, but there are some details to know.
  593: First, the second parameter passed to the disk declaration (the '0x1' in
  594: the example below)
  596:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  598: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  599: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
  600: So, hda1 which has major 3 and minor 1 on a Linux system will have
  601: device number 0x301.  Alternatively, devices names can be used (hda,
  602: hdb, ...)  as xentools has a table to map these names to devices
  603: numbers.  To export a partition to a Linux guest we can use:
  605:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
  606:         root = "/dev/hda1 ro"
  608: and it will appear as /dev/hda on the Linux system, and be used as root
  609: partition.
  611: To install the Linux system on the partition to be exported to the
  612: guest domain, the following method can be used: install
  613: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
  614: that will be the root partition of your Linux domain, and mount it.
  615: Then copy the files from a working Linux system, make adjustments in
  616: `/etc` (fstab, network config).  It should also be possible to extract
  617: binary packages such as .rpm or .deb directly to the mounted partition
  618: using the appropriate tool, possibly running under NetBSD's Linux
  619: emulation.  Once the file system has been populated, umount it.  If
  620: desirable, the file system can be converted to ext3 using tune2fs -j.
  621: It should now be possible to boot the Linux guest domain, using one of
  622: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
  624: To get the Linux console right, you need to add:
  626:     extra = "xencons=tty1"
  628: to your configuration since not all Linux distributions auto-attach a
  629: tty to the xen console.
  631: ## Creating a NetBSD HVM domU
  633: Use type='hvm', probably.  Use a GENERIC kernel within the disk image.
  635: ## Creating a NetBSD PVH domU
  637: This only works with a current kernel in the domU.
  639: Use type='pvh'.  Probably, use a GENERIC kernel within the disk image,
  640: which in current has PV support.
  642: \todo Verify.
  644: \todo Verify if one can have current PVH domU on a 9 dom0.
  646: ## Creating a Solaris domU
  648: See possibly outdated
  649: [Solaris domU instructions](/ports/xen/howto-solaris/).
  651: ## PCI passthrough: Using PCI devices in guest domains
  653: NB: PCI passthrough only works on some Xen versions and as of 2020 it
  654: is not clear that it works on any version in pkgsrc.  \todo Reports
  655: confirming or denying this notion should be sent to port-xen@.
  657: The dom0 can give other domains access to selected PCI
  658: devices. This can allow, for example, a non-privileged domain to have
  659: access to a physical network interface or disk controller.  However,
  660: keep in mind that giving a domain access to a PCI device most likely
  661: will give the domain read/write access to the whole physical memory,
  662: as PCs don't have an IOMMU to restrict memory access to DMA-capable
  663: device.  Also, it's not possible to export ISA devices to non-dom0
  664: domains, which means that the primary VGA adapter can't be exported.
  665: A guest domain trying to access the VGA registers will panic.
  667: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
  668: not been ported to later versions at this time.
  670: For a PCI device to be exported to a domU, is has to be attached to
  671: the "pciback" driver in dom0.  Devices passed to the dom0 via the
  672: pciback.hide boot parameter will attach to "pciback" instead of the
  673: usual driver.  The list of devices is specified as "(bus:dev.func)",
  674: where bus and dev are 2-digit hexadecimal numbers, and func a
  675: single-digit number:
  677:         pciback.hide=(00:0a.0)(00:06.0)
  679: pciback devices should show up in the dom0's boot messages, and the
  680: devices should be listed in the `/kern/xen/pci` directory.
  682: PCI devices to be exported to a domU are listed in the "pci" array of
  683: the domU's config file, with the format "0000:bus:dev.func".
  685:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
  687: In the domU an "xpci" device will show up, to which one or more pci
  688: buses will attach.  Then the PCI drivers will attach to PCI buses as
  689: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
  690: or any PCI drivers built in by default; you have to build your own
  691: kernel to use PCI devices in a domU.  Here's a kernel config example;
  692: note that only the "xpci" lines are unusual.
  694:         include         "arch/i386/conf/XEN3_DOMU"
  696:         # Add support for PCI buses to the XEN3_DOMU kernel
  697:         xpci* at xenbus ?
  698:         pci* at xpci ?
  700:         # PCI USB controllers
  701:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
  703:         # USB bus support
  704:         usb*    at uhci?
  706:         # USB Hubs
  707:         uhub*   at usb?
  708:         uhub*   at uhub? port ? configuration ? interface ?
  710:         # USB Mass Storage
  711:         umass*  at uhub? port ? configuration ? interface ?
  712:         wd*     at umass?
  713:         # SCSI controllers
  714:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
  716:         # SCSI bus support (for both ahc and umass)
  717:         scsibus* at scsi?
  719:         # SCSI devices
  720:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
  721:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
  724: # Miscellaneous Information
  726: ## Nesting under Linux KVM
  728: It is possible to run a Xen and a NetBSD dom0 under Linux KVM.  One
  729: can enable virtio in the dom0 for greater speed.
  731: ## Other nesting
  733: In theory, any full emulation should be able to run Xen and a NetBSD
  734: dom0.  The HOWTO does not currently have information about Xen XVM
  735: mode, nvmm, qemu, Virtualbox, etc.
  737: ## NetBSD 5 as domU
  739: [NetBSD 5 is known to panic.](
  740: (However, NetBSD 5 systems should be updated to a supported version.)
  742: # NetBSD as a domU in a VPS
  744: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
  745: hardware.  This section explains how to deal with Xen in a domU as a
  746: virtual private server where you do not control or have access to the
  747: dom0.  This is not intended to be an exhaustive list of VPS providers;
  748: only a few are mentioned that specifically support NetBSD.
  750: VPS operators provide varying degrees of access and mechanisms for
  751: configuration.  The big issue is usually how one controls which kernel
  752: is booted, because the kernel is nominally in the dom0 file system (to
  753: which VPS users do not normally have access).  A second issue is how
  754: to install NetBSD.
  755: A VPS user may want to compile a kernel for security updates, to run
  756: npf, run IPsec, or any other reason why someone would want to change
  757: their kernel.
  759: One approach is to have an administrative interface to upload a kernel,
  760: or to select from a prepopulated list.  Other approaches are pygrub
  761: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
  762: kernel from the domU file system.  This is closer to a regular physical
  763: computer, where someone who controls a machine can replace the kernel.
  765: A second issue is multiple CPUs.  With NetBSD 6, domUs support
  766: multiple vcpus, and it is typical for VPS providers to enable multiple
  767: CPUs for NetBSD domUs.
  769: ## Complexities due to Xen changes
  771: Xen has many security advisories and people running Xen systems make
  772: different choices.
  774: ### stub domains
  776: Some (Linux only?) dom0 systems use something called "stub domains" to
  777: isolate qemu from the dom0 system, as a security and reliabilty
  778: mechanism when running HVM domUs.  Somehow, NetBSD's GENERIC kernel
  779: ends up using PIO for disks rather than DMA.  Of course, all of this
  780: is emulated, but emulated PIO is unusably slow.  This problem is not
  781: currently understood.
  783: ### Grant tables
  785: There are multiple versions of using grant tables, and some security
  786: advisories have suggested disabling some versions.  Some versions of
  787: NetBSD apparently only use specific versions and this can lead to
  788: "NetBSD current doesn't run on hosting provider X" situations.
  790: \todo Explain better.
  792: ## Boot methods
  794: ### pvgrub
  796: pvgrub is a version of grub that uses PV operations instead of BIOS
  797: calls.  It is booted from the dom0 as the domU kernel, and then reads
  798: /grub/menu.lst and loads a kernel from the domU file system.
  800: [Panix]( lets users use pvgrub.  Panix reports
  801: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
  802: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
  803: page](, which describes only
  804: Linux but should be updated to cover NetBSD :-).
  806: []( also lets users with pvgrub to boot
  807: their own kernel.  See then [ NetBSD
  808: HOWTO](
  809: (which is in need of updating).
  811: It appears that [grub's FFS
  812: code](
  813: does not support all aspects of modern FFS, but there are also reports
  814: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
  815: partition for the kernel with the intent that grub can understand it,
  816: which leads to /netbsd not being the actual kernel.  One must remember
  817: to update the special boot partition.
  819: ### pygrub
  821: pygrub runs in the dom0 and looks into the domU file system.  This
  822: implies that the domU must have a kernel in a file system in a format
  823: known to pygrub.
  825: pygrub doesn't seem to work to load Linux images under NetBSD dom0,
  826: and is inherently less secure than pvgrub due to running inside dom0. For both these
  827: reasons, pygrub should not be used, and is only still present so that
  828: historical DomU images using it still work.
  830: As of 2014, pygrub seems to be of mostly historical
  831: interest. New DomUs should use pvgrub.
  833: ## Specific Providers
  835: ### Amazon
  837: See the [Amazon EC2 page](/amazon_ec2/).

CVSweb for NetBSD wikisrc <> software: FreeBSD-CVSweb