File:  [NetBSD Developer Wiki] / wikisrc / ports / xen / howto.mdwn
Revision 1.170: download - view: text, annotated - select for diffs
Sat Oct 31 15:29:23 2020 UTC (6 months, 2 weeks ago) by gdt
Branches: MAIN
CVS tags: HEAD
xen howto: Update PVHVM/PVH for current

    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: Xen supports different styles of guest:
   12: 
   13: [[!table data="""
   14: Style of guest	|Supported by NetBSD
   15: PV		|Yes (dom0, domU)
   16: HVM		|Yes (domU)
   17: PVHVM		|current-only (dom0, domU)
   18: PVH		|current-only (dom0, domU)
   19: """]]
   20: 
   21: In Para-Virtualized (PV) mode, the guest OS does not attempt to access
   22: hardware directly, but instead makes hypercalls to the hypervisor; PV
   23: guests must be specifically coded for Xen.
   24: 
   25: In HVM mode, no guest modification is required; however, hardware
   26: support is required, such as VT-x on Intel CPUs and SVM on AMD CPUs.
   27: 
   28: In PVHVM mode, the guest mostly runs as HVM, but can use PV drivers for efficiency.  See https://wiki.xen.org/wiki/PV_on_HVM
   29: 
   30: In PVH mode, the guest uses PV interfaces for IO and native interfaces for everything else.   This is also called PVHv2; see https://wiki.xenproject.org/wiki/PVH_(v2)_Domu
   31: 
   32: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
   33: The dom0 can start one or more domUs.  (Booting is explained in detail
   34: in the dom0 section.)
   35: 
   36: This HOWTO presumes a basic familiarity with the Xen system
   37: architecture, with installing NetBSD on i386/amd64 hardware, and with
   38: installing software from pkgsrc.  See also the [Xen
   39: website](http://www.xenproject.org/).
   40: 
   41: [[!toc]]
   42: 
   43: #Versions and Support
   44: 
   45: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
   46: xenkernel and xentools.  We will refer only to the kernel versions,
   47: but note that both packages must be installed together and must have
   48: matching versions.
   49: 
   50: Versions available in pkgsrc:
   51: 
   52: [[!table data="""
   53: Xen Version	|Package Name	|Xen CPU Support	|xm?	|EOL'ed By Upstream
   54: 4.2		|xenkernel42	|i386 x86_64		|yes	|Yes
   55: 4.5		|xenkernel45	|x86_64			|	|Yes
   56: 4.6		|xenkernel46	|x86_64			|	|Yes
   57: 4.8		|xenkernel48	|x86_64			|	|Yes
   58: 4.11		|xenkernel411	|x86_64			|	|No
   59: """]]
   60: 
   61: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
   62: 
   63: Multiprocessor (SMP) support in NetBSD differs depending on the domain:
   64: 
   65: [[!table data="""
   66: Domain		|Supports SMP
   67: dom0		|No
   68: domU		|Yes
   69: """]]
   70: 
   71: Note: NetBSD support is called XEN3.  However, it does support Xen 4,
   72: because the hypercall interface has remained identical.
   73: 
   74: Older Xen had a python-based management tool called xm, now replaced
   75: by xl.  xm is obsolete, but 4.2 remains in pkgsrc because migrating
   76: from xm to xl is not always trivial, and because 4.2 is the last
   77: version to run on an i386 dom0.
   78: 
   79: Architecture
   80: ------------
   81: 
   82: Xen 4.5 and later runs on x86_64 hardware (the NetBSD amd64 port).
   83: Xen 4.2 can in theory use i386 hardware, but we do not have
   84: recent reports of success.
   85: 
   86: The dom0 system, plus each domU, can be either i386PAE or amd64.
   87: i386 without PAE is not supported.
   88: 
   89: The standard approach is to use NetBSD/amd64 for the dom0.
   90: 
   91: To use an i386PAE dom0 (other than on 4.2), one must build or obtain a
   92: 64bit Xen kernel and install it on the system.
   93: 
   94: For domUs, i386PAE is considered as
   95: [faster](https://lists.xen.org/archives/html/xen-devel/2012-07/msg00085.html)
   96: than amd64.
   97: 
   98: # Creating a dom0
   99: 
  100: In order to install a NetBSD as a dom0, one must first install a normal
  101: NetBSD system, and then pivot the install to a dom0 install by changing
  102: the kernel and boot configuration.
  103: 
  104: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
  105: 512M it worked reliably.  This does not make sense, but if you see
  106: "not ELF" after Xen boots, try increasing dom0 RAM.
  107: 
  108: Installation of NetBSD
  109: ----------------------
  110: 
  111: [Install NetBSD/amd64](/guide/inst/)
  112: just as you would if you were not using Xen.
  113: 
  114: Installation of Xen
  115: -------------------
  116: 
  117: We will consider that you chose to use Xen 4.8, with NetBSD/amd64 as
  118: dom0. In the dom0, install xenkernel48 and xentools48 from pkgsrc.
  119: 
  120: Once this is done, install the Xen kernel itself:
  121: 
  122: [[!template id=programlisting text="""
  123: # cp /usr/pkg/xen48-kernel/xen.gz /
  124: """]]
  125: 
  126: Then, place a NetBSD XEN3_DOM0 kernel in the `/` directory. Such kernel
  127: can either be compiled manually, or downloaded from the NetBSD FTP, for
  128: example at:
  129: 
  130: [[!template id=programlisting text="""
  131: ftp.netbsd.org/pub/NetBSD/NetBSD-8.0/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
  132: """]]
  133: 
  134: Add a line to /boot.cfg to boot Xen:
  135: 
  136: [[!template id=filecontent name="/boot.cfg" text="""
  137: menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
  138: """]]
  139: 
  140: This specifies that the dom0 should have 512MB of ram, leaving the rest
  141: to be allocated for domUs.  To use a serial console, use:
  142: 
  143: [[!template id=filecontent name="/boot.cfg" text="""
  144: menu=Xen:load /netbsd-XEN3_DOM0.gz;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
  145: """]]
  146: 
  147: which will use the first serial port for Xen (which counts starting
  148: from 1, unlike NetBSD which counts starting from 0), forcing
  149: speed/parity.  Because the NetBSD command line lacks a
  150: "console=pc" argument, it will use the default "xencons" console device,
  151: which directs the console I/O through Xen to the same console device Xen
  152: itself uses (in this case, the serial port).
  153: 
  154: In an attempt to add performance, one can also add `dom0_max_vcpus=1 dom0_vcpus_pin`,
  155: to force only one vcpu to be provided (since NetBSD dom0 can't use
  156: more) and to pin that vcpu to a physical CPU. Xen has
  157: [many boot options](http://xenbits.xenproject.org/docs/4.8-testing/misc/xen-command-line.html),
  158: and other than dom0 memory and max_vcpus, they are generally not
  159: necessary.
  160: 
  161: Copy the boot scripts into `/etc/rc.d`:
  162: 
  163: [[!template id=programlisting text="""
  164: # cp /usr/pkg/share/examples/rc.d/xen* /etc/rc.d/
  165: """]]
  166: 
  167: Enable `xencommons`:
  168: 
  169: [[!template id=filecontent name="/etc/rc.conf" text="""
  170: xencommons=YES
  171: """]]
  172: 
  173: Now, reboot so that you are running a DOM0 kernel under Xen, rather
  174: than GENERIC without Xen.
  175: 
  176: TODO: Recommend for/against xen-watchdog.
  177: 
  178: Once the reboot is done, use `xl` to inspect Xen's boot messages,
  179: available resources, and running domains.  For example:
  180: 
  181: [[!template id=programlisting text="""
  182: # xl dmesg
  183: ... xen's boot info ...
  184: # xl info
  185: ... available memory, etc ...
  186: # xl list
  187: Name              Id  Mem(MB)  CPU  State  Time(s)  Console
  188: Domain-0           0       64    0  r----     58.1
  189: """]]
  190: 
  191: Xen logs will be in /var/log/xen.
  192: 
  193: ### Issues with xencommons
  194: 
  195: `xencommons` starts `xenstored`, which stores data on behalf of dom0 and
  196: domUs.  It does not currently work to stop and start xenstored.
  197: Certainly all domUs should be shutdown first, following the sort order
  198: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
  199: and is not notified when xenstored exits, leading to not recreating
  200: the state when the new xenstored starts.  Until there's a mechanism to
  201: make this work, one should not expect to be able to restart xenstored
  202: (and thus xencommons).  There is currently no reason to expect that
  203: this will get fixed any time soon.
  204: 
  205: anita (for testing NetBSD)
  206: --------------------------
  207: 
  208: With the setup so far (assuming 4.8/xl), one should be able to run
  209: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
  210: root, because anita must create a domU):
  211: 
  212: [[!template id=programlisting text="""
  213: anita --vmm=xl test file:///usr/obj/i386/
  214: """]]
  215: 
  216: Xen-specific NetBSD issues
  217: --------------------------
  218: 
  219: There are (at least) two additional things different about NetBSD as a
  220: dom0 kernel compared to hardware.
  221: 
  222: One is that the module ABI is different because some of the #defines
  223: change, so one must build modules for Xen.  As of netbsd-7, the build
  224: system does this automatically.
  225: 
  226: The other difference is that XEN3_DOM0 does not have exactly the same
  227: options as GENERIC.  While it is debatable whether or not this is a
  228: bug, users should be aware of this and can simply add missing config
  229: items if desired.
  230: 
  231: Updating NetBSD in a dom0
  232: -------------------------
  233: 
  234: This is just like updating NetBSD on bare hardware, assuming the new
  235: version supports the version of Xen you are running.  Generally, one
  236: replaces the kernel and reboots, and then overlays userland binaries
  237: and adjusts `/etc`.
  238: 
  239: Note that one must update both the non-Xen kernel typically used for
  240: rescue purposes and the DOM0 kernel used with Xen.
  241: 
  242: Converting from grub to /boot
  243: -----------------------------
  244: 
  245: These instructions were used to convert a system from
  246: grub to /boot.  The system was originally installed in February of
  247: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
  248: over time.  Before these commands, it was running NetBSD 6 i386, Xen
  249: 4.1 and grub, much like the message linked earlier in the grub
  250: section.
  251: 
  252: [[!template id=programlisting text="""
  253: # Install MBR bootblocks on both disks.
  254: fdisk -i /dev/rwd0d
  255: fdisk -i /dev/rwd1d
  256: # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
  257: installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
  258: installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
  259: # Install secondary boot loader
  260: cp -p /usr/mdec/boot /
  261: # Create boot.cfg following earlier guidance:
  262: menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
  263: menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=512M
  264: menu=GENERIC:boot
  265: menu=GENERIC single-user:boot -s
  266: menu=GENERIC.ok:boot netbsd.ok
  267: menu=GENERIC.ok single-user:boot netbsd.ok -s
  268: menu=Drop to boot prompt:prompt
  269: default=1
  270: timeout=30
  271: """]]
  272: 
  273: Upgrading Xen versions
  274: ---------------------
  275: 
  276: Minor version upgrades are trivial.  Just rebuild/replace the
  277: xenkernel version and copy the new xen.gz to `/` (where `/boot.cfg`
  278: references it), and reboot.
  279: 
  280: #Unprivileged domains (domU)
  281: 
  282: This section describes general concepts about domUs.  It does not
  283: address specific domU operating systems or how to install them.  The
  284: config files for domUs are typically in `/usr/pkg/etc/xen`, and are
  285: typically named so that the file name, domU name and the domU's host
  286: name match.
  287: 
  288: The domU is provided with CPU and memory by Xen, configured by the
  289: dom0.  The domU is provided with disk and network by the dom0,
  290: mediated by Xen, and configured in the dom0.
  291: 
  292: Entropy in domUs can be an issue; physical disks and network are on
  293: the dom0.  NetBSD's /dev/random system works, but is often challenged.
  294: 
  295: Config files
  296: ------------
  297: 
  298: See /usr/pkg/share/examples/xen/xlexample*
  299: for a small number of well-commented examples, mostly for running
  300: GNU/Linux.
  301: 
  302: The following is an example minimal domain configuration file. The domU
  303: serves as a network file server.
  304: 
  305: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
  306: name = "domU-id"
  307: kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
  308: memory = 1024
  309: vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
  310: disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
  311:          'file:/n0/xen/foo-wd1,0x1,w' ]
  312: """]]
  313: 
  314: The domain will have name given in the `name` setting.  The kernel has the
  315: host/domU name in it, so that on the dom0 one can update the various
  316: domUs independently.  The `vif` line causes an interface to be provided,
  317: with a specific mac address (do not reuse MAC addresses!), in bridge
  318: mode.  Two disks are provided, and they are both writable; the bits
  319: are stored in files and Xen attaches them to a vnd(4) device in the
  320: dom0 on domain creation.  The system treats xbd0 as the boot device
  321: without needing explicit configuration.
  322: 
  323: By convention, domain config files are kept in `/usr/pkg/etc/xen`.  Note
  324: that "xl create" takes the name of a config file, while other commands
  325: take the name of a domain.
  326: 
  327: Examples of commands:
  328: 
  329: [[!template id=programlisting text="""
  330: xl create /usr/pkg/etc/xen/foo
  331: xl console domU-id
  332: xl create -c /usr/pkg/etc/xen/foo
  333: xl shutdown domU-id
  334: xl list
  335: """]]
  336: 
  337: Typing `^]` will exit the console session.  Shutting down a domain is
  338: equivalent to pushing the power button; a NetBSD domU will receive a
  339: power-press event and do a clean shutdown.  Shutting down the dom0
  340: will trigger controlled shutdowns of all configured domUs.
  341: 
  342: CPU and memory
  343: --------------
  344: 
  345: A domain is provided with some number of vcpus, up to the number
  346: of CPUs seen by the hypervisor. For a domU, it is controlled
  347: from the config file by the "vcpus = N" directive.
  348: 
  349: A domain is provided with memory; this is controlled in the config
  350: file by "memory = N" (in megabytes).  In the straightforward case, the
  351: sum of the the memory allocated to the dom0 and all domUs must be less
  352: than the available memory.
  353: 
  354: Xen also provides a "balloon" driver, which can be used to let domains
  355: use more memory temporarily.
  356: 
  357: Virtual disks
  358: -------------
  359: 
  360: In domU config files, the disks are defined as a sequence of 3-tuples:
  361: 
  362:  * The first element is "method:/path/to/disk". Common methods are
  363:    "file:" for a file-backed vnd, and "phy:" for something that is already
  364:    a device, such as an LVM logical volume.
  365: 
  366:  * The second element is an artifact of how virtual disks are passed to
  367:    Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
  368:    are given a device name to associate with the disk, and values like
  369:    "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
  370:    as xbd0, the second as xbd1, and so on.  However, xl demands a
  371:    second argument.  The name given is converted to a major/minor by
  372:    calling stat(2) on the name in /dev and this is passed to the domU.
  373:    In the general case, the dom0 and domU can be different operating
  374:    systems, and it is an unwarranted assumption that they have consistent
  375:    numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
  376:    as both dom0 and domU, using values of 0x0 for the first disk and 0x1
  377:    for the second works fine and avoids this issue.  For a GNU/Linux
  378:    guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
  379:    /dev/hda1.
  380: 
  381:  * The third element is "w" for writable disks, and "r" for read-only
  382:    disks.
  383: 
  384: Example:
  385: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
  386: disk = [ 'file:/n0/xen/foo-wd0,0x0,w' ]
  387: """]]
  388: 
  389: Note that NetBSD by default creates only vnd[0123].  If you need more
  390: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
  391: dom0.
  392: 
  393: Note that NetBSD by default creates only xbd[0123].  If you need more
  394: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
  395: 
  396: Virtual Networking
  397: ------------------
  398: 
  399: Xen provides virtual Ethernets, each of which connects the dom0 and a
  400: domU.  For each virtual network, there is an interface "xvifN.M" in
  401: the dom0, and a matching interface xennetM (NetBSD name) in domU index N.
  402: The interfaces behave as if there is an Ethernet with two
  403: adapters connected.  From this primitive, one can construct various
  404: configurations.  We focus on two common and useful cases for which
  405: there are existing scripts: bridging and NAT.
  406: 
  407: With bridging (in the example above), the domU perceives itself to be
  408: on the same network as the dom0.  For server virtualization, this is
  409: usually best.  Bridging is accomplished by creating a bridge(4) device
  410: and adding the dom0's physical interface and the various xvifN.0
  411: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
  412: config file.  The bridge must be set up already in the dom0; an
  413: example /etc/ifconfig.bridge0 is:
  414: 
  415: [[!template id=filecontent name="/etc/ifconfig.bridge0" text="""
  416: create
  417: up
  418: !brconfig bridge0 add wm0
  419: """]]
  420: 
  421: With NAT, the domU perceives itself to be behind a NAT running on the
  422: dom0.  This is often appropriate when running Xen on a workstation.
  423: TODO: NAT appears to be configured by "vif = [ '' ]".
  424: 
  425: The MAC address specified is the one used for the interface in the new
  426: domain.  The interface in dom0 will use this address XOR'd with
  427: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
  428: 
  429: Starting domains automatically
  430: ------------------------------
  431: 
  432: To start domains `domU-netbsd` and `domU-linux` at boot and shut them
  433: down cleanly on dom0 shutdown, add the following in rc.conf:
  434: 
  435: [[!template id=filecontent name="/etc/rc.conf" text="""
  436: xendomains="domU-netbsd domU-linux"
  437: """]]
  438: 
  439: #Creating a domU
  440: 
  441: Creating domUs is almost entirely independent of operating system.  We
  442: have already presented the basics of config files.  Note that you must
  443: have already completed the dom0 setup so that "xl list" works.
  444: 
  445: Creating a NetBSD domU
  446: ----------------------
  447: 
  448: See the earlier config file, and adjust memory.  Decide on how much
  449: storage you will provide, and prepare it (file or LVM).
  450: 
  451: While the kernel will be obtained from the dom0 file system, the same
  452: file should be present in the domU as /netbsd so that tools like
  453: savecore(8) can work.   (This is helpful but not necessary.)
  454: 
  455: The kernel must be specifically for Xen and for use as a domU.  The
  456: i386 and amd64 provide the following kernels:
  457: 
  458:         i386 XEN3PAE_DOMU
  459:         amd64 XEN3_DOMU
  460: 
  461: This will boot NetBSD, but this is not that useful if the disk is
  462: empty.  One approach is to unpack sets onto the disk outside of xen
  463: (by mounting it, just as you would prepare a physical disk for a
  464: system you can't run the installer on).
  465: 
  466: A second approach is to run an INSTALL kernel, which has a miniroot
  467: and can load sets from the network.  To do this, copy the INSTALL
  468: kernel to / and change the kernel line in the config file to:
  469: 
  470:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
  471: 
  472: Then, start the domain as "xl create -c configfile".
  473: 
  474: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
  475: line should be used in the config file.
  476: 
  477:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
  478: 
  479: After booting the domain, the option to install via CDROM may be
  480: selected.  The CDROM device should be changed to `xbd1d`.
  481: 
  482: Once done installing, "halt -p" the new domain (don't reboot or halt,
  483: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
  484: config file), switch the config file back to the XEN3_DOMU kernel,
  485: and start the new domain again. Now it should be able to use "root on
  486: xbd0a" and you should have a, functional NetBSD domU.
  487: 
  488: TODO: check if this is still accurate.
  489: When the new domain is booting you'll see some warnings about *wscons*
  490: and the pseudo-terminals. These can be fixed by editing the files
  491: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
  492: `/etc/ttys`, except *console*, like this:
  493: 
  494:     console "/usr/libexec/getty Pc"         vt100   on secure
  495:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
  496:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
  497:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
  498:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
  499: 
  500: Finally, all screens must be commented out from `/etc/wscons.conf`.
  501: 
  502: It is also desirable to add
  503: 
  504:         powerd=YES
  505: 
  506: in rc.conf. This way, the domain will be properly shut down if
  507: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
  508: 
  509: It is not strictly necessary to have a kernel (as /netbsd) in the domU
  510: file system.  However, various programs (e.g. netstat) will use that
  511: kernel to look up symbols to read from kernel virtual memory.  If
  512: /netbsd is not the running kernel, those lookups will fail.  (This is
  513: not really a Xen-specific issue, but because the domU kernel is
  514: obtained from the dom0, it is far more likely to be out of sync or
  515: missing with Xen.)
  516: 
  517: Creating a Linux domU
  518: ---------------------
  519: 
  520: Creating unprivileged Linux domains isn't much different from
  521: unprivileged NetBSD domains, but there are some details to know.
  522: 
  523: First, the second parameter passed to the disk declaration (the '0x1' in
  524: the example below)
  525: 
  526:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
  527: 
  528: does matter to Linux. It wants a Linux device number here (e.g. 0x300
  529: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
  530: So, hda1 which has major 3 and minor 1 on a Linux system will have
  531: device number 0x301.  Alternatively, devices names can be used (hda,
  532: hdb, ...)  as xentools has a table to map these names to devices
  533: numbers.  To export a partition to a Linux guest we can use:
  534: 
  535:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
  536:         root = "/dev/hda1 ro"
  537: 
  538: and it will appear as /dev/hda on the Linux system, and be used as root
  539: partition.
  540: 
  541: To install the Linux system on the partition to be exported to the
  542: guest domain, the following method can be used: install
  543: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
  544: that will be the root partition of your Linux domain, and mount it.
  545: Then copy the files from a working Linux system, make adjustments in
  546: `/etc` (fstab, network config).  It should also be possible to extract
  547: binary packages such as .rpm or .deb directly to the mounted partition
  548: using the appropriate tool, possibly running under NetBSD's Linux
  549: emulation.  Once the file system has been populated, umount it.  If
  550: desirable, the file system can be converted to ext3 using tune2fs -j.
  551: It should now be possible to boot the Linux guest domain, using one of
  552: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
  553: 
  554: To get the Linux console right, you need to add:
  555: 
  556:     extra = "xencons=tty1"
  557: 
  558: to your configuration since not all Linux distributions auto-attach a
  559: tty to the xen console.
  560: 
  561: Creating a Solaris domU
  562: -----------------------
  563: 
  564: See possibly outdated
  565: [Solaris domU instructions](/ports/xen/howto-solaris/).
  566: 
  567: 
  568: PCI passthrough: Using PCI devices in guest domains
  569: ---------------------------------------------------
  570: 
  571: The dom0 can give other domains access to selected PCI
  572: devices. This can allow, for example, a non-privileged domain to have
  573: access to a physical network interface or disk controller.  However,
  574: keep in mind that giving a domain access to a PCI device most likely
  575: will give the domain read/write access to the whole physical memory,
  576: as PCs don't have an IOMMU to restrict memory access to DMA-capable
  577: device.  Also, it's not possible to export ISA devices to non-dom0
  578: domains, which means that the primary VGA adapter can't be exported.
  579: A guest domain trying to access the VGA registers will panic.
  580: 
  581: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
  582: not been ported to later versions at this time.
  583: 
  584: For a PCI device to be exported to a domU, is has to be attached to
  585: the "pciback" driver in dom0.  Devices passed to the dom0 via the
  586: pciback.hide boot parameter will attach to "pciback" instead of the
  587: usual driver.  The list of devices is specified as "(bus:dev.func)",
  588: where bus and dev are 2-digit hexadecimal numbers, and func a
  589: single-digit number:
  590: 
  591:         pciback.hide=(00:0a.0)(00:06.0)
  592: 
  593: pciback devices should show up in the dom0's boot messages, and the
  594: devices should be listed in the `/kern/xen/pci` directory.
  595: 
  596: PCI devices to be exported to a domU are listed in the "pci" array of
  597: the domU's config file, with the format "0000:bus:dev.func".
  598: 
  599:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
  600: 
  601: In the domU an "xpci" device will show up, to which one or more pci
  602: buses will attach.  Then the PCI drivers will attach to PCI buses as
  603: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
  604: or any PCI drivers built in by default; you have to build your own
  605: kernel to use PCI devices in a domU.  Here's a kernel config example;
  606: note that only the "xpci" lines are unusual.
  607: 
  608:         include         "arch/i386/conf/XEN3_DOMU"
  609: 
  610:         # Add support for PCI buses to the XEN3_DOMU kernel
  611:         xpci* at xenbus ?
  612:         pci* at xpci ?
  613: 
  614:         # PCI USB controllers
  615:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
  616: 
  617:         # USB bus support
  618:         usb*    at uhci?
  619: 
  620:         # USB Hubs
  621:         uhub*   at usb?
  622:         uhub*   at uhub? port ? configuration ? interface ?
  623: 
  624:         # USB Mass Storage
  625:         umass*  at uhub? port ? configuration ? interface ?
  626:         wd*     at umass?
  627:         # SCSI controllers
  628:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
  629: 
  630:         # SCSI bus support (for both ahc and umass)
  631:         scsibus* at scsi?
  632: 
  633:         # SCSI devices
  634:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
  635:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
  636: 
  637: 
  638: #NetBSD as a domU in a VPS
  639: 
  640: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
  641: hardware.  This section explains how to deal with Xen in a domU as a
  642: virtual private server where you do not control or have access to the
  643: dom0.  This is not intended to be an exhaustive list of VPS providers;
  644: only a few are mentioned that specifically support NetBSD.
  645: 
  646: VPS operators provide varying degrees of access and mechanisms for
  647: configuration.  The big issue is usually how one controls which kernel
  648: is booted, because the kernel is nominally in the dom0 file system (to
  649: which VPS users do not normally have access).  A second issue is how
  650: to install NetBSD.
  651: A VPS user may want to compile a kernel for security updates, to run
  652: npf, run IPsec, or any other reason why someone would want to change
  653: their kernel.
  654: 
  655: One approach is to have an administrative interface to upload a kernel,
  656: or to select from a prepopulated list.  Other approaches are pygrub
  657: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
  658: kernel from the domU file system.  This is closer to a regular physical
  659: computer, where someone who controls a machine can replace the kernel.
  660: 
  661: A second issue is multiple CPUs.  With NetBSD 6, domUs support
  662: multiple vcpus, and it is typical for VPS providers to enable multiple
  663: CPUs for NetBSD domUs.
  664: 
  665: pvgrub
  666: ------
  667: 
  668: pvgrub is a version of grub that uses PV operations instead of BIOS
  669: calls.  It is booted from the dom0 as the domU kernel, and then reads
  670: /grub/menu.lst and loads a kernel from the domU file system.
  671: 
  672: [Panix](http://www.panix.com/) lets users use pvgrub.  Panix reports
  673: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
  674: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
  675: page](http://www.panix.com/v-colo/grub.html), which describes only
  676: Linux but should be updated to cover NetBSD :-).
  677: 
  678: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
  679: their own kernel.  See then [prgmr.com NetBSD
  680: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
  681: (which is in need of updating).
  682: 
  683: It appears that [grub's FFS
  684: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
  685: does not support all aspects of modern FFS, but there are also reports
  686: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
  687: partition for the kernel with the intent that grub can understand it,
  688: which leads to /netbsd not being the actual kernel.  One must remember
  689: to update the special boot partition.
  690: 
  691: pygrub
  692: -------
  693: 
  694: pygrub runs in the dom0 and looks into the domU file system.  This
  695: implies that the domU must have a kernel in a file system in a format
  696: known to pygrub.
  697: 
  698: pygrub doesn't seem to work to load Linux images under NetBSD dom0,
  699: and is inherently less secure than pvgrub due to running inside dom0. For both these
  700: reasons, pygrub should not be used, and is only still present so that
  701: historical DomU images using it still work.
  702: 
  703: As of 2014, pygrub seems to be of mostly historical
  704: interest. New DomUs should use pvgrub.
  705: 
  706: Amazon
  707: ------
  708: 
  709: See the [Amazon EC2 page](/amazon_ec2/).

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