File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.197: download - view: text, annotated - select for diffs
Wed Mar 3 23:17:59 2021 UTC (2 months, 2 weeks ago) by gdt
Branches: MAIN
CVS tags: HEAD
xen: add todo about balloon

    1: [[!meta title="Xen HowTo"]]
    2: 
    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.
   10: 
   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](http://www.xenproject.org/).
   15: 
   16: [[!toc]]
   17: 
   18: # Overview
   19: 
   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.
   24: 
   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.
   29: 
   30: There are many choices one can make; the HOWTO recommends the standard
   31: approach and limits discussion of alternatives in many cases.
   32: 
   33: ## Guest Styles
   34: 
   35: Xen supports different styles of guests.
   36: 
   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: """]]
   44: 
   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](https://wiki.xen.org/wiki/Paravirtualization_(PV\)).
   49: 
   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.
   53: 
   54: In PVHVM mode, the guest runs as HVM, but additionally can use PV
   55: drivers for efficiency.
   56: See [PV on HVM](https://wiki.xen.org/wiki/PV_on_HVM).
   57: 
   58: There have been two PVH modes: original PVH and PVHv2.  Original PVH
   59: was based on PV mode and is no longer relevant at all.  PVHv2 is
   60: basically lightweight HVM with PV drivers.  A critical feature of it
   61: is that qemu is not needed; the hypervisor can do the emulation that
   62: is required.  Thus, a dom0 can be PVHv2.
   63: The source code uses PVH and config files use pvh; this refers to PVHv2.
   64: See [PVH(v2)](https://wiki.xenproject.org/wiki/PVH_(v2\)_Domu).
   65: 
   66: At system boot, the dom0 kernel is loaded as a module with Xen as the kernel.
   67: The dom0 can start one or more domUs.  (Booting is explained in detail
   68: in the dom0 section.)
   69: 
   70: ## CPU Architecture
   71: 
   72: Xen runs on x86_64 hardware (the NetBSD amd64 port).
   73: 
   74: There is a concept of Xen running on ARM, but there are no reports of this working with NetBSD.
   75: 
   76: The dom0 system should be amd64.  (Instructions for i386PAE dom0 have been removed from the HOWTO.)
   77: 
   78: The domU can be i386PAE or amd64.
   79: i386PAE at one point was considered as [faster](https://lists.xen.org/archives/html/xen-devel/2012-07/msg00085.html) than amd64.
   80: 
   81: ## Xen Versions
   82: 
   83: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
   84: xenkernel and xentools.  We will refer only to the kernel versions,
   85: but note that both packages must be installed together and must have
   86: matching versions.
   87: 
   88: Versions available in pkgsrc:
   89: 
   90: [[!table data="""
   91: Xen Version	|Package Name	|Xen CPU Support	|EOL'ed By Upstream
   92: 4.11		|xenkernel411	|x86_64			|No
   93: 4.13		|xenkernel413	|x86_64			|No
   94: """]]
   95: 
   96: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
   97: 
   98: Older Xen had a python-based management tool called xm, now replaced
   99: by xl.
  100: 
  101: ## NetBSD versions
  102: 
  103: Xen has been supported in NetBSD for a long time, at least since 2005.
  104: Initially Xen was PV only.
  105: 
  106: NetBSD 8 and up support PV and HVM modes.
  107: 
  108: Support for PVHVM and PVH is available only in NetBSD-current.
  109: 
  110: NetBSD up to and including NetBSD 9 as a dom0 does not run SMP,
  111: because some drivers are not yet safe for this.  NetBSD-current
  112: supports SMP in dom0.
  113: 
  114: NetBSD, when run as a domU, can and does typically run SMP.
  115: 
  116: Note that while Xen 4.13 is current, the kernel support is still
  117: called XEN3, because the hypercall interface has not changed
  118: significantly.
  119: 
  120: # Creating a NetBSD dom0
  121: 
  122: In order to install a NetBSD as a dom0, one first installs a normal
  123: NetBSD system, and then pivot the install to a dom0 install by
  124: changing the kernel and boot configuration.
  125: 
  126: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
  127: 512M it worked reliably.  This does not make sense, but if you see
  128: "not ELF" after Xen boots, try increasing dom0 RAM.
  129: 
  130: ## Installation of NetBSD
  131: 
  132: [Install NetBSD/amd64](/guide/inst/) just as you would if you were not
  133: using Xen.  Therefore, use the most recent release, or a build from
  134: the most recent stable branch.  Alternatively, use -current, being
  135: mindful of all the usual caveats of lower stability of current, and
  136: likely a bit more so.
  137: 
  138: ## Installation of Xen
  139: 
  140: ### Building Xen
  141: 
  142: Use the most recent version of Xen in pkgsrc, unless the DESCR says that it is not suitable.
  143: Therefore, choose 4.13.
  144: In the dom0, install xenkernel413 and xentools413 from pkgsrc.
  145: 
  146: Once this is done, copy the Xen kernel from where pkgsrc puts it to
  147: where the boot process will be able to find it:
  148: 
  149: [[!template id=programlisting text="""
  150: # cp -p /usr/pkg/xen413-kernel/xen.gz /
  151: """]]
  152: 
  153: Then, place a NetBSD XEN3_DOM0 kernel in the `/` directory. Such
  154: kernel can either be taken from a local release build.sh run, compiled
  155: manually, or downloaded from the NetBSD FTP, for example at:
  156: 
  157: [[!template id=programlisting text="""
  158: ftp.netbsd.org/pub/NetBSD/NetBSD-9.1/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
  159: """]]
  160: 
  161: ### Configuring booting
  162: 
  163: Read boot.cfg(8) carefully.  Add lines to /boot.cfg to boot Xen:
  164: 
  165: [[!template id=filecontent name="/boot.cfg" text="""
  166: menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
  167: menu=Xen single user:load /netbsd-XEN3_DOM0.gz console=pc -s;multiboot /xen.gz dom0_mem=512M
  168: """]]
  169: 
  170: This specifies that the dom0 should have 512MB of ram, leaving the rest
  171: to be allocated for domUs.
  172: 
  173: NB: This says add, not replace, so that you will be able to more
  174: easily boot a NetBSD kernel without Xen.  Once Xen boots ok, you may
  175: want to set it as default.  It is highly likely that you will have
  176: trouble at some point, and keeping an up-to-date GENERIC for use in
  177: fixing problems is the standard prudent approach.
  178: 
  179: \todo Explain why rndseed is not set with Xen as part of the dom0
  180: subconfiguration.
  181: 
  182: Beware that userconf statements must be attached to the dom0 load, and
  183: may not be at top-level, because then they would try to configure the
  184: hypervisor, if there is a way to pass them via multiboot .  It appears
  185: that adding `userconf=pckbc` to `/boot.cfg` causes Xen to crash very
  186: early with a heap overflow.
  187: 
  188: ### Console selection
  189: 
  190: See boot_console(8).  Understand that you should start from a place of
  191: having console setup correct for booting GENERIC before trying to
  192: configure Xen.
  193: 
  194: Generally for GENERIC, one sets the console in bootxx_ffsv1 or
  195: equivalent, and this is passed on to /boot (where one typically does
  196: not set the console).  This configuration of bootxx_ffsv1 should also
  197: be in place for Xen systems, to allow seeing messages from /boot and
  198: use of a keyboard to select a line from the menu.  And, one should
  199: have a working boot path to GENERIC for rescue situations.
  200: 
  201: With GENERIC, the boot options are passed on to /netbsd, but there is
  202: currently no mechanism to pass these via multiboot to the hypervisor.
  203: Thus, in addition to configuring the console in the boot blocks, one
  204: must also configure it for Xen.
  205: 
  206: By default, the hypervisor (Xen itself) will use some sort of vga
  207: device as the console, much like GENERIC uses by default.  The vga
  208: console is relinquished at the conclusion of hypervisor boot, before
  209: the dom0 is started.  Xen when using a vga console does not process
  210: console input.
  211: 
  212: The hypervisor can be configured to use a serial port console, e.g.
  213: [[!template id=filecontent name="/boot.cfg" text="""
  214: menu=Xen:losad /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
  215: """]]
  216: This example uses the first serial port (Xen counts from 1; this is
  217: what NetBSD would call com0), and sets speed and parity.  (The dom0 is
  218: then configured to use the same serial port in this example.)
  219: 
  220: With the hypervisor configured for a serial console, it can get input,
  221: and there is a notion of passing this input to the dom0.  \todo
  222: Explain why, if Xen has a serial console, the dom0 console is
  223: typically also configured to open that same serial port, instead of
  224: getting the passthrough input via the xen console.
  225: 
  226: One also configures the console for the dom0.  While one might expect
  227: console=pc to be default, following behavior of GENERIC, a hasty read
  228: of the code suggests there is no default and booting without a
  229: selected console might lead to a panic.  Also, there is merit in
  230: explicit configuration.  Therefore the standard approach is to place
  231: console=pc as part of the load statement for the dom0 kernel, or
  232: alternatively console=com0.
  233: 
  234: The NetBSD dom0 kernel will attach xencons(4) (the man page does not
  235: exist), but this is not used as a console.  It is used to obtain the
  236: messages from the hypervisor's console; run `xl dmesg` to see them.
  237: 
  238: ### Tuning
  239: 
  240: In an attempt to add performance, one can also add `dom0_max_vcpus=1 dom0_vcpus_pin`,
  241: to force only one vcpu to be provided (since NetBSD dom0 can't use
  242: more) and to pin that vcpu to a physical CPU. Xen has
  243: [many boot options](http://xenbits.xenproject.org/docs/4.13-testing/misc/xen-command-line.html),
  244: and other than dom0 memory and max_vcpus, they are generally not
  245: necessary.
  246: \todo Revisit this advice with current.
  247: \todo Explain if anyone has ever actually measured that this helps.
  248: 
  249: ### rc.conf
  250: 
  251: Ensure that the boot scripts installed in
  252: `/usr/pkg/share/examples/rc.d` are in `/etc/rc.d`, either because you
  253: have `PKG_RCD_SCRIPTS=yes`, or manually.  (This is not special to Xen,
  254: but a normal part of pkgsrc usage.)
  255: 
  256: Set `xencommons=YES` in rc.conf:
  257: 
  258: [[!template id=filecontent name="/etc/rc.conf" text="""
  259: xencommons=YES
  260: """]]
  261: 
  262: \todo Recommend for/against xen-watchdog.
  263: 
  264: ### Testing
  265: 
  266: Now, reboot so that you are running a DOM0 kernel under Xen, rather
  267: than GENERIC without Xen.
  268: 
  269: Once the reboot is done, use `xl` to inspect Xen's boot messages,
  270: available resources, and running domains.  For example:
  271: 
  272: [[!template id=programlisting text="""
  273: # xl dmesg
  274: ... xen's boot info ...
  275: # xl info
  276: ... available memory, etc ...
  277: # xl list
  278: Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  279: Domain-0           0       64    0  r----     58.1
  280: """]]
  281: 
  282: Xen logs will be in /var/log/xen.
  283: 
  284: ### Issues with xencommons
  285: 
  286: `xencommons` starts `xenstored`, which stores data on behalf of dom0 and
  287: domUs.  It does not currently work to stop and start xenstored.
  288: Certainly all domUs should be shutdown first, following the sort order
  289: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
  290: and is not notified when xenstored exits, leading to not recreating
  291: the state when the new xenstored starts.  Until there's a mechanism to
  292: make this work, one should not expect to be able to restart xenstored
  293: (and thus xencommons).  There is currently no reason to expect that
  294: this will get fixed any time soon.
  295: \todo Confirm if this is still true in 2020.
  296: 
  297: ## Xen-specific NetBSD issues
  298: 
  299: There are (at least) two additional things different about NetBSD as a
  300: dom0 kernel compared to hardware.
  301: 
  302: One is that through NetBSD 9 the module ABI is different because some
  303: of the #defines change, so there are separate sets of modules in
  304: /stand.  In NetBSD-current, there is only one set of modules.
  305: 
  306: The other difference is that XEN3_DOM0 does not have exactly the same
  307: options as GENERIC.  While it is debatable whether or not this is a
  308: bug, users should be aware of this and can simply add missing config
  309: items if desired.
  310: 
  311: Finally, there have been occasional reports of trouble with X11
  312: servers in NetBSD as a dom0.
  313: 
  314: ## Updating Xen in a dom0
  315: 
  316: Basically, update the xenkernel and xentools packages and copy the new
  317: Xen kernel into place, and reboot.  This procedure should be usable to
  318: update to a new Xen release, but the reader is reminded that having a
  319: non-Xen boot methods was recommended earlier.
  320: 
  321: ## Updating NetBSD in a dom0
  322: 
  323: This is just like updating NetBSD on bare hardware, assuming the new
  324: version supports the version of Xen you are running.  Generally, one
  325: replaces the kernel and reboots, and then overlays userland binaries
  326: and adjusts `/etc`.
  327: 
  328: Note that one should update both the non-Xen kernel typically used for
  329: rescue purposes, as well as the DOM0 kernel used with Xen.
  330: 
  331: ## anita (for testing NetBSD)
  332: 
  333: With a NetBSD dom0, even without any domUs, one should be able to run
  334: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
  335: root, because anita must create a domU):
  336: 
  337: [[!template id=programlisting text="""
  338: anita --vmm=xl test file:///usr/obj/i386/
  339: """]]
  340: 
  341: # Unprivileged domains (domU)
  342: 
  343: This section describes general concepts about domUs.  It does not
  344: address specific domU operating systems or how to install them.  The
  345: config files for domUs are typically in `/usr/pkg/etc/xen`, and are
  346: typically named so that the file name, domU name and the domU's host
  347: name match.
  348: 
  349: The domU is provided with CPU and memory by Xen, configured by the
  350: dom0.  The domU is provided with disk and network by the dom0,
  351: mediated by Xen, and configured in the dom0.
  352: 
  353: Entropy in domUs can be an issue; physical disks and network are on
  354: the dom0.  NetBSD's /dev/random system works, but is often challenged.
  355: 
  356: ## Config files
  357: 
  358: See /usr/pkg/share/examples/xen/xlexample*
  359: for a small number of well-commented examples, mostly for running
  360: GNU/Linux.
  361: 
  362: The following is an example minimal domain configuration file. The domU
  363: serves as a network file server.
  364: 
  365: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
  366: name = "domU-id"
  367: kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
  368: memory = 1024
  369: vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
  370: disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
  371:          'file:/n0/xen/foo-wd1,0x1,w' ]
  372: """]]
  373: 
  374: The domain will have name given in the `name` setting.  The kernel has the
  375: host/domU name in it, so that on the dom0 one can update the various
  376: domUs independently.  The `vif` line causes an interface to be provided,
  377: with a specific mac address (do not reuse MAC addresses!), in bridge
  378: mode.  Two disks are provided, and they are both writable; the bits
  379: are stored in files and Xen attaches them to a vnd(4) device in the
  380: dom0 on domain creation.  The system treats xbd0 as the boot device
  381: without needing explicit configuration.
  382: 
  383: There is not type line; that implicitly defines a pv domU.
  384: 
  385: By convention, domain config files are kept in `/usr/pkg/etc/xen`.  Note
  386: that "xl create" takes the name of a config file, while other commands
  387: take the name of a domain.
  388: 
  389: Examples of commands:
  390: 
  391: [[!template id=programlisting text="""
  392: xl create /usr/pkg/etc/xen/foo
  393: xl console domU-id
  394: xl create -c /usr/pkg/etc/xen/foo
  395: xl shutdown domU-id
  396: xl list
  397: """]]
  398: 
  399: Typing `^]` will exit the console session.  Shutting down a domain is
  400: equivalent to pushing the power button; a NetBSD domU will receive a
  401: power-press event and do a clean shutdown.  Shutting down the dom0
  402: will trigger controlled shutdowns of all configured domUs.
  403: 
  404: ## CPU and memory
  405: 
  406: A domain is provided with some number of vcpus, up to the number
  407: of CPUs seen by the hypervisor. For a domU, it is controlled
  408: from the config file by the "vcpus = N" directive.
  409: 
  410: A domain is provided with memory; this is controlled in the config
  411: file by "memory = N" (in megabytes).  In the straightforward case, the
  412: sum of the the memory allocated to the dom0 and all domUs must be less
  413: than the available memory.
  414: 
  415: ## Balloon driver
  416: 
  417: Xen provides a `balloon` driver, which can be used to let domains use
  418: more memory temporarily.
  419: 
  420: \todo Explain how to set up a aystem to use the balloon scheme in a
  421: useful manner.
  422: 
  423: ## Virtual disks
  424: 
  425: In domU config files, the disks are defined as a sequence of 3-tuples:
  426: 
  427:  * The first element is "method:/path/to/disk". Common methods are
  428:    "file:" for a file-backed vnd, and "phy:" for something that is already
  429:    a device, such as an LVM logical volume.
  430: 
  431:  * The second element is an artifact of how virtual disks are passed to
  432:    Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
  433:    are given a device name to associate with the disk, and values like
  434:    "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
  435:    as xbd0, the second as xbd1, and so on.  However, xl demands a
  436:    second argument.  The name given is converted to a major/minor by
  437:    calling stat(2) on the name in /dev and this is passed to the domU.
  438:    In the general case, the dom0 and domU can be different operating
  439:    systems, and it is an unwarranted assumption that they have consistent
  440:    numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
  441:    as both dom0 and domU, using values of 0x0 for the first disk and 0x1
  442:    for the second works fine and avoids this issue.  For a GNU/Linux
  443:    guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
  444:    /dev/hda1.
  445: 
  446:  * The third element is "w" for writable disks, and "r" for read-only
  447:    disks.
  448: 
  449: Example:
  450: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
  451: disk = [ 'file:/n0/xen/foo-wd0,0x0,w' ]
  452: """]]
  453: 
  454: Note that NetBSD by default creates only vnd[0123].  If you need more
  455: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
  456: dom0.
  457: 
  458: ## Virtual Networking
  459: 
  460: Xen provides virtual Ethernets, each of which connects the dom0 and a
  461: domU.  For each virtual network, there is an interface "xvifN.M" in
  462: the dom0, and a matching interface xennetM (NetBSD name) in domU index N.
  463: The interfaces behave as if there is an Ethernet with two
  464: adapters connected.  From this primitive, one can construct various
  465: configurations.  We focus on two common and useful cases for which
  466: there are existing scripts: bridging and NAT.
  467: 
  468: With bridging (in the example above), the domU perceives itself to be
  469: on the same network as the dom0.  For server virtualization, this is
  470: usually best.  Bridging is accomplished by creating a bridge(4) device
  471: and adding the dom0's physical interface and the various xvifN.0
  472: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
  473: config file.  The bridge must be set up already in the dom0; an
  474: example /etc/ifconfig.bridge0 is:
  475: 
  476: [[!template id=filecontent name="/etc/ifconfig.bridge0" text="""
  477: create
  478: up
  479: !brconfig bridge0 add wm0
  480: """]]
  481: 
  482: With NAT, the domU perceives itself to be behind a NAT running on the
  483: dom0.  This is often appropriate when running Xen on a workstation.
  484: TODO: NAT appears to be configured by "vif = [ '' ]".
  485: 
  486: The MAC address specified is the one used for the interface in the new
  487: domain.  The interface in dom0 will use this address XOR'd with
  488: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
  489: 
  490: ## Starting domains automatically
  491: 
  492: To start domains `domU-netbsd` and `domU-linux` at boot and shut them
  493: down cleanly on dom0 shutdown, add the following in rc.conf:
  494: 
  495: [[!template id=filecontent name="/etc/rc.conf" text="""
  496: xendomains="domU-netbsd domU-linux"
  497: """]]
  498: 
  499: # domU setup for specific systems
  500: 
  501: Creating domUs is almost entirely independent of operating system.  We
  502: have already presented the basics of config files in the previous system.
  503: 
  504: Of course, this section presumes that you have a working dom0.
  505: 
  506: ## Creating a NetBSD PV domU
  507: 
  508: See the earlier config file, and adjust memory.  Decide on how much
  509: storage you will provide, and prepare it (file or LVM).
  510: 
  511: While the kernel will be obtained from the dom0 file system, the same
  512: file should be present in the domU as /netbsd so that tools like
  513: savecore(8) can work.   (This is helpful but not necessary.)
  514: 
  515: The kernel must be specifically built for Xen, to use PV interfacesas
  516: a domU.  NetBSD release builds provide the following kernels:
  517: 
  518:         i386 XEN3PAE_DOMU
  519:         amd64 XEN3_DOMU
  520: 
  521: This will boot NetBSD, but this is not that useful if the disk is
  522: empty.  One approach is to unpack sets onto the disk outside of Xen
  523: (by mounting it, just as you would prepare a physical disk for a
  524: system you can't run the installer on).
  525: 
  526: A second approach is to run an INSTALL kernel, which has a miniroot
  527: and can load sets from the network.  To do this, copy the INSTALL
  528: kernel to / and change the kernel line in the config file to:
  529: 
  530:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  531: 
  532: Then, start the domain as "xl create -c configfile".
  533: 
  534: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
  535: line should be used in the config file.
  536: 
  537:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  538: 
  539: After booting the domain, the option to install via CDROM may be
  540: selected.  The CDROM device should be changed to `xbd1d`.
  541: 
  542: Once done installing, "halt -p" the new domain (don't reboot or halt:
  543: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
  544: config file), switch the config file back to the XEN3_DOMU kernel,
  545: and start the new domain again. Now it should be able to use "root on
  546: xbd0a" and you should have a functional NetBSD domU.
  547: 
  548: TODO: check if this is still accurate.
  549: When the new domain is booting you'll see some warnings about *wscons*
  550: and the pseudo-terminals. These can be fixed by editing the files
  551: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  552: `/etc/ttys`, except *console*, like this:
  553: 
  554:     console "/usr/libexec/getty Pc"         vt100   on secure
  555:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  556:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  557:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  558:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  559: 
  560: Finally, all screens must be commented out from `/etc/wscons.conf`.
  561: 
  562: One should also run `powerd` in a domU, but this should not need
  563: configuring.  With powerd, the domain will run a controlled shutdown
  564: if `xl shutdown -R` or `xl shutdown -H` is used on the dom0, via
  565: receiving a synthetic `power button pressed` signal.  In 9 and
  566: current, `powerd` is run by default under Xen kernels (or if ACPI is
  567: present), and it can be added to rc.conf if not.
  568: 
  569: It is not strictly necessary to have a kernel (as /netbsd) in the domU
  570: file system.  However, various programs (e.g. netstat) will use that
  571: kernel to look up symbols to read from kernel virtual memory.  If
  572: /netbsd is not the running kernel, those lookups will fail.  (This is
  573: not really a Xen-specific issue, but because the domU kernel is
  574: obtained from the dom0, it is far more likely to be out of sync or
  575: missing with Xen.)
  576: 
  577: Note that NetBSD by default creates only xbd[0123].  If you need more
  578: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
  579: 
  580: ## Creating a Linux PV domU
  581: 
  582: Creating unprivileged Linux domains isn't much different from
  583: unprivileged NetBSD domains, but there are some details to know.
  584: 
  585: First, the second parameter passed to the disk declaration (the '0x1' in
  586: the example below)
  587: 
  588:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  589: 
  590: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  591: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
  592: So, hda1 which has major 3 and minor 1 on a Linux system will have
  593: device number 0x301.  Alternatively, devices names can be used (hda,
  594: hdb, ...)  as xentools has a table to map these names to devices
  595: numbers.  To export a partition to a Linux guest we can use:
  596: 
  597:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
  598:         root = "/dev/hda1 ro"
  599: 
  600: and it will appear as /dev/hda on the Linux system, and be used as root
  601: partition.
  602: 
  603: To install the Linux system on the partition to be exported to the
  604: guest domain, the following method can be used: install
  605: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
  606: that will be the root partition of your Linux domain, and mount it.
  607: Then copy the files from a working Linux system, make adjustments in
  608: `/etc` (fstab, network config).  It should also be possible to extract
  609: binary packages such as .rpm or .deb directly to the mounted partition
  610: using the appropriate tool, possibly running under NetBSD's Linux
  611: emulation.  Once the file system has been populated, umount it.  If
  612: desirable, the file system can be converted to ext3 using tune2fs -j.
  613: It should now be possible to boot the Linux guest domain, using one of
  614: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
  615: 
  616: To get the Linux console right, you need to add:
  617: 
  618:     extra = "xencons=tty1"
  619: 
  620: to your configuration since not all Linux distributions auto-attach a
  621: tty to the xen console.
  622: 
  623: ## Creating a NetBSD HVM domU
  624: 
  625: Use type='hvm', probably.  Use a GENERIC kernel within the disk image.
  626: 
  627: ## Creating a NetBSD PVH domU
  628: 
  629: This only works with a current kernel in the domU.
  630: 
  631: Use type='pvh'.  Probably, use a GENERIC kernel within the disk image,
  632: which in current has PV support.
  633: 
  634: \todo Verify.
  635: 
  636: \todo Verify if one can have current PVH domU on a 9 dom0.
  637: 
  638: ## Creating a Solaris domU
  639: 
  640: See possibly outdated
  641: [Solaris domU instructions](/ports/xen/howto-solaris/).
  642: 
  643: ## PCI passthrough: Using PCI devices in guest domains
  644: 
  645: NB: PCI passthrough only works on some Xen versions and as of 2020 it
  646: is not clear that it works on any version in pkgsrc.  \todo Reports
  647: confirming or denying this notion should be sent to port-xen@.
  648: 
  649: The dom0 can give other domains access to selected PCI
  650: devices. This can allow, for example, a non-privileged domain to have
  651: access to a physical network interface or disk controller.  However,
  652: keep in mind that giving a domain access to a PCI device most likely
  653: will give the domain read/write access to the whole physical memory,
  654: as PCs don't have an IOMMU to restrict memory access to DMA-capable
  655: device.  Also, it's not possible to export ISA devices to non-dom0
  656: domains, which means that the primary VGA adapter can't be exported.
  657: A guest domain trying to access the VGA registers will panic.
  658: 
  659: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
  660: not been ported to later versions at this time.
  661: 
  662: For a PCI device to be exported to a domU, is has to be attached to
  663: the "pciback" driver in dom0.  Devices passed to the dom0 via the
  664: pciback.hide boot parameter will attach to "pciback" instead of the
  665: usual driver.  The list of devices is specified as "(bus:dev.func)",
  666: where bus and dev are 2-digit hexadecimal numbers, and func a
  667: single-digit number:
  668: 
  669:         pciback.hide=(00:0a.0)(00:06.0)
  670: 
  671: pciback devices should show up in the dom0's boot messages, and the
  672: devices should be listed in the `/kern/xen/pci` directory.
  673: 
  674: PCI devices to be exported to a domU are listed in the "pci" array of
  675: the domU's config file, with the format "0000:bus:dev.func".
  676: 
  677:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
  678: 
  679: In the domU an "xpci" device will show up, to which one or more pci
  680: buses will attach.  Then the PCI drivers will attach to PCI buses as
  681: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
  682: or any PCI drivers built in by default; you have to build your own
  683: kernel to use PCI devices in a domU.  Here's a kernel config example;
  684: note that only the "xpci" lines are unusual.
  685: 
  686:         include         "arch/i386/conf/XEN3_DOMU"
  687: 
  688:         # Add support for PCI buses to the XEN3_DOMU kernel
  689:         xpci* at xenbus ?
  690:         pci* at xpci ?
  691: 
  692:         # PCI USB controllers
  693:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
  694: 
  695:         # USB bus support
  696:         usb*    at uhci?
  697: 
  698:         # USB Hubs
  699:         uhub*   at usb?
  700:         uhub*   at uhub? port ? configuration ? interface ?
  701: 
  702:         # USB Mass Storage
  703:         umass*  at uhub? port ? configuration ? interface ?
  704:         wd*     at umass?
  705:         # SCSI controllers
  706:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
  707: 
  708:         # SCSI bus support (for both ahc and umass)
  709:         scsibus* at scsi?
  710: 
  711:         # SCSI devices
  712:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
  713:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
  714: 
  715: 
  716: # Miscellaneous Information
  717: 
  718: ## Nesting under Linux KVM
  719: 
  720: It is possible to run a Xen and a NetBSD dom0 under Linux KVM.  One
  721: can enable virtio in the dom0 for greater speed.
  722: 
  723: ## Other nesting
  724: 
  725: In theory, any full emulation should be able to run Xen and a NetBSD
  726: dom0.  The HOWTO does not currently have information about Xen XVM
  727: mode, nvmm, qemu, Virtualbox, etc.
  728: 
  729: ## NetBSD 5 as domU
  730: 
  731: [NetBSD 5 is known to panic.](http://mail-index.netbsd.org/port-xen/2018/04/17/msg009181.html)
  732: (However, NetBSD 5 systems should be updated to a supported version.)
  733: 
  734: # NetBSD as a domU in a VPS
  735: 
  736: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
  737: hardware.  This section explains how to deal with Xen in a domU as a
  738: virtual private server where you do not control or have access to the
  739: dom0.  This is not intended to be an exhaustive list of VPS providers;
  740: only a few are mentioned that specifically support NetBSD.
  741: 
  742: VPS operators provide varying degrees of access and mechanisms for
  743: configuration.  The big issue is usually how one controls which kernel
  744: is booted, because the kernel is nominally in the dom0 file system (to
  745: which VPS users do not normally have access).  A second issue is how
  746: to install NetBSD.
  747: A VPS user may want to compile a kernel for security updates, to run
  748: npf, run IPsec, or any other reason why someone would want to change
  749: their kernel.
  750: 
  751: One approach is to have an administrative interface to upload a kernel,
  752: or to select from a prepopulated list.  Other approaches are pygrub
  753: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
  754: kernel from the domU file system.  This is closer to a regular physical
  755: computer, where someone who controls a machine can replace the kernel.
  756: 
  757: A second issue is multiple CPUs.  With NetBSD 6, domUs support
  758: multiple vcpus, and it is typical for VPS providers to enable multiple
  759: CPUs for NetBSD domUs.
  760: 
  761: ## Complexities due to Xen changes
  762: 
  763: Xen has many security advisories and people running Xen systems make
  764: different choices.
  765: 
  766: ### stub domains
  767: 
  768: Some (Linux only?) dom0 systems use something called "stub domains" to
  769: isolate qemu from the dom0 system, as a security and reliabilty
  770: mechanism when running HVM domUs.  Somehow, NetBSD's GENERIC kernel
  771: ends up using PIO for disks rather than DMA.  Of course, all of this
  772: is emulated, but emulated PIO is unusably slow.  This problem is not
  773: currently understood.
  774: 
  775: ### Grant tables
  776: 
  777: There are multiple versions of using grant tables, and some security
  778: advisories have suggested disabling some versions.  Some versions of
  779: NetBSD apparently only use specific versions and this can lead to
  780: "NetBSD current doesn't run on hosting provider X" situations.
  781: 
  782: \todo Explain better.
  783: 
  784: ## Boot methods
  785: 
  786: ### pvgrub
  787: 
  788: pvgrub is a version of grub that uses PV operations instead of BIOS
  789: calls.  It is booted from the dom0 as the domU kernel, and then reads
  790: /grub/menu.lst and loads a kernel from the domU file system.
  791: 
  792: [Panix](http://www.panix.com/) lets users use pvgrub.  Panix reports
  793: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
  794: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
  795: page](http://www.panix.com/v-colo/grub.html), which describes only
  796: Linux but should be updated to cover NetBSD :-).
  797: 
  798: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
  799: their own kernel.  See then [prgmr.com NetBSD
  800: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
  801: (which is in need of updating).
  802: 
  803: It appears that [grub's FFS
  804: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
  805: does not support all aspects of modern FFS, but there are also reports
  806: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
  807: partition for the kernel with the intent that grub can understand it,
  808: which leads to /netbsd not being the actual kernel.  One must remember
  809: to update the special boot partition.
  810: 
  811: ### pygrub
  812: 
  813: pygrub runs in the dom0 and looks into the domU file system.  This
  814: implies that the domU must have a kernel in a file system in a format
  815: known to pygrub.
  816: 
  817: pygrub doesn't seem to work to load Linux images under NetBSD dom0,
  818: and is inherently less secure than pvgrub due to running inside dom0. For both these
  819: reasons, pygrub should not be used, and is only still present so that
  820: historical DomU images using it still work.
  821: 
  822: As of 2014, pygrub seems to be of mostly historical
  823: interest. New DomUs should use pvgrub.
  824: 
  825: ## Specific Providers
  826: 
  827: ### Amazon
  828: 
  829: See the [Amazon EC2 page](/amazon_ec2/).

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