Annotation of wikisrc/ports/xen/howto.mdwn, revision 1.176

1.144     maxv        1: [[!meta title="Xen HowTo"]]
                      2: 
1.157     maxv        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.
1.12      gdt        10: 
1.172     gdt        11: Xen supports different styles of guests; see [PV on HVM](https://wiki.xen.org/wiki/PV_on_HVM) and [PVH(v2)](https://wiki.xenproject.org/wiki/PVH_(v2)_Domu) for upstream documentation.
1.149     maxv       12: 
                     13: [[!table data="""
                     14: Style of guest |Supported by NetBSD
1.150     maxv       15: PV             |Yes (dom0, domU)
                     16: HVM            |Yes (domU)
1.171     gdt        17: PVHVM          |current-only (domU)
1.176   ! gdt        18: PVHv2          |current-only (domU, dom0 not yet)
1.149     maxv       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
1.170     gdt        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: 
1.176   ! gdt        28: In PVHVM mode, the guest runs as HVM, but additionally can use PV
        !            29: drivers for efficiency.
1.170     gdt        30: 
1.176   ! gdt        31: In PVHv2H mode, operation is similar to PVHVM, except that qemu is not
        !            32: run and thus the PV interfaces for console, disks, networking are the
        !            33: only way to access these resources.
1.149     maxv       34: 
1.27      jnemeth    35: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
1.12      gdt        36: The dom0 can start one or more domUs.  (Booting is explained in detail
                     37: in the dom0 section.)
                     38: 
                     39: This HOWTO presumes a basic familiarity with the Xen system
1.123     gdt        40: architecture, with installing NetBSD on i386/amd64 hardware, and with
                     41: installing software from pkgsrc.  See also the [Xen
                     42: website](http://www.xenproject.org/).
1.1       mspo       43: 
1.160     maxv       44: [[!toc]]
1.150     maxv       45: 
1.174     gdt        46: # Versions and Support
1.15      gdt        47: 
1.111     wiz        48: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
1.15      gdt        49: xenkernel and xentools.  We will refer only to the kernel versions,
                     50: but note that both packages must be installed together and must have
                     51: matching versions.
                     52: 
1.145     maxv       53: Versions available in pkgsrc:
1.85      gdt        54: 
1.145     maxv       55: [[!table data="""
1.166     gdt        56: Xen Version    |Package Name   |Xen CPU Support        |xm?    |EOL'ed By Upstream
1.167     gdt        57: 4.11           |xenkernel411   |x86_64                 |       |No
1.176   ! gdt        58: 4.13           |xenkernel413   |x86_64                 |       |No
1.145     maxv       59: """]]
1.113     gdt        60: 
1.96      gdt        61: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
                     62: 
1.157     maxv       63: Multiprocessor (SMP) support in NetBSD differs depending on the domain:
1.145     maxv       64: 
1.157     maxv       65: [[!table data="""
                     66: Domain         |Supports SMP
                     67: dom0           |No
                     68: domU           |Yes
                     69: """]]
1.15      gdt        70: 
1.167     gdt        71: Note: NetBSD support is called XEN3.  However, it does support Xen 4,
1.147     maxv       72: because the hypercall interface has remained identical.
                     73: 
1.166     gdt        74: Older Xen had a python-based management tool called xm, now replaced
1.176   ! gdt        75: by xl.
1.166     gdt        76: 
1.18      gdt        77: Architecture
                     78: ------------
                     79: 
1.167     gdt        80: Xen 4.5 and later runs on x86_64 hardware (the NetBSD amd64 port).
1.176   ! gdt        81: There is a concept of Xen running on ARM, but there are no reports of this working with NetBSD.
1.157     maxv       82: 
1.176   ! gdt        83: The dom0 system should be amd64.  (Instructions for i386PAE dom0 have been removed from the HOWTO.)
1.157     maxv       84: 
1.176   ! gdt        85: The domU can be i386PAE or amd64.
        !            86: i386PAE at one point was considered as [faster](https://lists.xen.org/archives/html/xen-devel/2012-07/msg00085.html) than amd64.
1.29      gdt        87: 
1.166     gdt        88: # Creating a dom0
1.15      gdt        89: 
1.158     maxv       90: In order to install a NetBSD as a dom0, one must first install a normal
                     91: NetBSD system, and then pivot the install to a dom0 install by changing
                     92: the kernel and boot configuration.
1.15      gdt        93: 
1.142     gdt        94: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
                     95: 512M it worked reliably.  This does not make sense, but if you see
                     96: "not ELF" after Xen boots, try increasing dom0 RAM.
                     97: 
1.19      gdt        98: Installation of NetBSD
                     99: ----------------------
1.13      gdt       100: 
1.157     maxv      101: [Install NetBSD/amd64](/guide/inst/)
1.19      gdt       102: just as you would if you were not using Xen.
1.1       mspo      103: 
1.19      gdt       104: Installation of Xen
                    105: -------------------
1.1       mspo      106: 
1.155     maxv      107: We will consider that you chose to use Xen 4.8, with NetBSD/amd64 as
                    108: dom0. In the dom0, install xenkernel48 and xentools48 from pkgsrc.
                    109: 
                    110: Once this is done, install the Xen kernel itself:
                    111: 
                    112: [[!template id=programlisting text="""
                    113: # cp /usr/pkg/xen48-kernel/xen.gz /
                    114: """]]
                    115: 
1.159     maxv      116: Then, place a NetBSD XEN3_DOM0 kernel in the `/` directory. Such kernel
                    117: can either be compiled manually, or downloaded from the NetBSD FTP, for
                    118: example at:
                    119: 
                    120: [[!template id=programlisting text="""
                    121: ftp.netbsd.org/pub/NetBSD/NetBSD-8.0/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
                    122: """]]
1.20      gdt       123: 
1.155     maxv      124: Add a line to /boot.cfg to boot Xen:
1.20      gdt       125: 
1.157     maxv      126: [[!template id=filecontent name="/boot.cfg" text="""
1.152     maxv      127: menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
                    128: """]]
1.20      gdt       129: 
1.159     maxv      130: This specifies that the dom0 should have 512MB of ram, leaving the rest
                    131: to be allocated for domUs.  To use a serial console, use:
1.77      gdt       132: 
1.157     maxv      133: [[!template id=filecontent name="/boot.cfg" text="""
1.152     maxv      134: menu=Xen:load /netbsd-XEN3_DOM0.gz;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
                    135: """]]
1.77      gdt       136: 
                    137: which will use the first serial port for Xen (which counts starting
1.139     gson      138: from 1, unlike NetBSD which counts starting from 0), forcing
                    139: speed/parity.  Because the NetBSD command line lacks a
                    140: "console=pc" argument, it will use the default "xencons" console device,
                    141: which directs the console I/O through Xen to the same console device Xen
                    142: itself uses (in this case, the serial port).
                    143: 
1.157     maxv      144: In an attempt to add performance, one can also add `dom0_max_vcpus=1 dom0_vcpus_pin`,
1.37      gdt       145: to force only one vcpu to be provided (since NetBSD dom0 can't use
1.158     maxv      146: more) and to pin that vcpu to a physical CPU. Xen has
                    147: [many boot options](http://xenbits.xenproject.org/docs/4.8-testing/misc/xen-command-line.html),
1.111     wiz       148: and other than dom0 memory and max_vcpus, they are generally not
1.93      gdt       149: necessary.
                    150: 
1.159     maxv      151: Copy the boot scripts into `/etc/rc.d`:
                    152: 
                    153: [[!template id=programlisting text="""
                    154: # cp /usr/pkg/share/examples/rc.d/xen* /etc/rc.d/
                    155: """]]
                    156: 
1.158     maxv      157: Enable `xencommons`:
1.31      gdt       158: 
1.157     maxv      159: [[!template id=filecontent name="/etc/rc.conf" text="""
1.152     maxv      160: xencommons=YES
                    161: """]]
1.31      gdt       162: 
1.158     maxv      163: Now, reboot so that you are running a DOM0 kernel under Xen, rather
                    164: than GENERIC without Xen.
1.157     maxv      165: 
1.31      gdt       166: TODO: Recommend for/against xen-watchdog.
1.27      jnemeth   167: 
1.158     maxv      168: Once the reboot is done, use `xl` to inspect Xen's boot messages,
1.157     maxv      169: available resources, and running domains.  For example:
1.34      gdt       170: 
1.153     maxv      171: [[!template id=programlisting text="""
                    172: # xl dmesg
                    173: ... xen's boot info ...
                    174: # xl info
                    175: ... available memory, etc ...
                    176: # xl list
                    177: Name              Id  Mem(MB)  CPU  State  Time(s)  Console
                    178: Domain-0           0       64    0  r----     58.1
                    179: """]]
                    180: 
                    181: Xen logs will be in /var/log/xen.
1.33      gdt       182: 
1.88      gdt       183: ### Issues with xencommons
                    184: 
1.157     maxv      185: `xencommons` starts `xenstored`, which stores data on behalf of dom0 and
1.88      gdt       186: domUs.  It does not currently work to stop and start xenstored.
                    187: Certainly all domUs should be shutdown first, following the sort order
                    188: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
                    189: and is not notified when xenstored exits, leading to not recreating
                    190: the state when the new xenstored starts.  Until there's a mechanism to
                    191: make this work, one should not expect to be able to restart xenstored
                    192: (and thus xencommons).  There is currently no reason to expect that
                    193: this will get fixed any time soon.
                    194: 
1.41      gdt       195: anita (for testing NetBSD)
                    196: --------------------------
                    197: 
1.157     maxv      198: With the setup so far (assuming 4.8/xl), one should be able to run
1.82      gdt       199: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
                    200: root, because anita must create a domU):
                    201: 
1.153     maxv      202: [[!template id=programlisting text="""
                    203: anita --vmm=xl test file:///usr/obj/i386/
                    204: """]]
1.82      gdt       205: 
1.40      gdt       206: Xen-specific NetBSD issues
                    207: --------------------------
                    208: 
                    209: There are (at least) two additional things different about NetBSD as a
                    210: dom0 kernel compared to hardware.
                    211: 
1.111     wiz       212: One is that the module ABI is different because some of the #defines
1.109     gdt       213: change, so one must build modules for Xen.  As of netbsd-7, the build
1.157     maxv      214: system does this automatically.
1.40      gdt       215: 
                    216: The other difference is that XEN3_DOM0 does not have exactly the same
                    217: options as GENERIC.  While it is debatable whether or not this is a
                    218: bug, users should be aware of this and can simply add missing config
                    219: items if desired.
                    220: 
1.15      gdt       221: Updating NetBSD in a dom0
                    222: -------------------------
                    223: 
                    224: This is just like updating NetBSD on bare hardware, assuming the new
                    225: version supports the version of Xen you are running.  Generally, one
                    226: replaces the kernel and reboots, and then overlays userland binaries
1.157     maxv      227: and adjusts `/etc`.
1.15      gdt       228: 
                    229: Note that one must update both the non-Xen kernel typically used for
                    230: rescue purposes and the DOM0 kernel used with Xen.
                    231: 
1.55      gdt       232: Converting from grub to /boot
                    233: -----------------------------
                    234: 
1.157     maxv      235: These instructions were used to convert a system from
1.55      gdt       236: grub to /boot.  The system was originally installed in February of
                    237: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
                    238: over time.  Before these commands, it was running NetBSD 6 i386, Xen
                    239: 4.1 and grub, much like the message linked earlier in the grub
                    240: section.
                    241: 
1.157     maxv      242: [[!template id=programlisting text="""
                    243: # Install MBR bootblocks on both disks.
                    244: fdisk -i /dev/rwd0d
                    245: fdisk -i /dev/rwd1d
                    246: # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
                    247: installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
                    248: installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
                    249: # Install secondary boot loader
                    250: cp -p /usr/mdec/boot /
                    251: # Create boot.cfg following earlier guidance:
                    252: menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
                    253: menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=512M
                    254: menu=GENERIC:boot
                    255: menu=GENERIC single-user:boot -s
                    256: menu=GENERIC.ok:boot netbsd.ok
                    257: menu=GENERIC.ok single-user:boot netbsd.ok -s
                    258: menu=Drop to boot prompt:prompt
                    259: default=1
                    260: timeout=30
                    261: """]]
1.22      gdt       262: 
1.102     gdt       263: Upgrading Xen versions
1.15      gdt       264: ---------------------
                    265: 
1.110     gdt       266: Minor version upgrades are trivial.  Just rebuild/replace the
1.157     maxv      267: xenkernel version and copy the new xen.gz to `/` (where `/boot.cfg`
1.110     gdt       268: references it), and reboot.
                    269: 
1.160     maxv      270: #Unprivileged domains (domU)
1.28      gdt       271: 
                    272: This section describes general concepts about domUs.  It does not
1.33      gdt       273: address specific domU operating systems or how to install them.  The
1.157     maxv      274: config files for domUs are typically in `/usr/pkg/etc/xen`, and are
1.60      wiki      275: typically named so that the file name, domU name and the domU's host
1.33      gdt       276: name match.
                    277: 
1.111     wiz       278: The domU is provided with CPU and memory by Xen, configured by the
1.33      gdt       279: dom0.  The domU is provided with disk and network by the dom0,
                    280: mediated by Xen, and configured in the dom0.
                    281: 
                    282: Entropy in domUs can be an issue; physical disks and network are on
                    283: the dom0.  NetBSD's /dev/random system works, but is often challenged.
                    284: 
1.48      gdt       285: Config files
                    286: ------------
                    287: 
1.165     gson      288: See /usr/pkg/share/examples/xen/xlexample*
1.164     gson      289: for a small number of well-commented examples, mostly for running
1.48      gdt       290: GNU/Linux.
                    291: 
1.157     maxv      292: The following is an example minimal domain configuration file. The domU
                    293: serves as a network file server.
                    294: 
                    295: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
                    296: name = "domU-id"
                    297: kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
                    298: memory = 1024
                    299: vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
                    300: disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
                    301:          'file:/n0/xen/foo-wd1,0x1,w' ]
                    302: """]]
1.48      gdt       303: 
1.157     maxv      304: The domain will have name given in the `name` setting.  The kernel has the
1.48      gdt       305: host/domU name in it, so that on the dom0 one can update the various
1.157     maxv      306: domUs independently.  The `vif` line causes an interface to be provided,
1.48      gdt       307: with a specific mac address (do not reuse MAC addresses!), in bridge
                    308: mode.  Two disks are provided, and they are both writable; the bits
                    309: are stored in files and Xen attaches them to a vnd(4) device in the
1.111     wiz       310: dom0 on domain creation.  The system treats xbd0 as the boot device
1.48      gdt       311: without needing explicit configuration.
                    312: 
1.163     gson      313: By convention, domain config files are kept in `/usr/pkg/etc/xen`.  Note
1.157     maxv      314: that "xl create" takes the name of a config file, while other commands
                    315: take the name of a domain.
                    316: 
                    317: Examples of commands:
1.48      gdt       318: 
1.157     maxv      319: [[!template id=programlisting text="""
1.163     gson      320: xl create /usr/pkg/etc/xen/foo
                    321: xl console domU-id
                    322: xl create -c /usr/pkg/etc/xen/foo
                    323: xl shutdown domU-id
1.157     maxv      324: xl list
                    325: """]]
                    326: 
                    327: Typing `^]` will exit the console session.  Shutting down a domain is
1.48      gdt       328: equivalent to pushing the power button; a NetBSD domU will receive a
                    329: power-press event and do a clean shutdown.  Shutting down the dom0
                    330: will trigger controlled shutdowns of all configured domUs.
                    331: 
1.33      gdt       332: CPU and memory
                    333: --------------
                    334: 
1.169     plunky    335: A domain is provided with some number of vcpus, up to the number
1.157     maxv      336: of CPUs seen by the hypervisor. For a domU, it is controlled
1.48      gdt       337: from the config file by the "vcpus = N" directive.
                    338: 
                    339: A domain is provided with memory; this is controlled in the config
                    340: file by "memory = N" (in megabytes).  In the straightforward case, the
                    341: sum of the the memory allocated to the dom0 and all domUs must be less
1.33      gdt       342: than the available memory.
                    343: 
                    344: Xen also provides a "balloon" driver, which can be used to let domains
1.157     maxv      345: use more memory temporarily.
1.28      gdt       346: 
                    347: Virtual disks
                    348: -------------
                    349: 
1.158     maxv      350: In domU config files, the disks are defined as a sequence of 3-tuples:
                    351: 
                    352:  * The first element is "method:/path/to/disk". Common methods are
1.162     gutterid  353:    "file:" for a file-backed vnd, and "phy:" for something that is already
                    354:    a device, such as an LVM logical volume.
1.158     maxv      355: 
                    356:  * The second element is an artifact of how virtual disks are passed to
                    357:    Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
                    358:    are given a device name to associate with the disk, and values like
                    359:    "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
                    360:    as xbd0, the second as xbd1, and so on.  However, xl demands a
                    361:    second argument.  The name given is converted to a major/minor by
                    362:    calling stat(2) on the name in /dev and this is passed to the domU.
                    363:    In the general case, the dom0 and domU can be different operating
                    364:    systems, and it is an unwarranted assumption that they have consistent
                    365:    numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
                    366:    as both dom0 and domU, using values of 0x0 for the first disk and 0x1
                    367:    for the second works fine and avoids this issue.  For a GNU/Linux
                    368:    guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
                    369:    /dev/hda1.
1.48      gdt       370: 
1.158     maxv      371:  * The third element is "w" for writable disks, and "r" for read-only
                    372:    disks.
                    373: 
                    374: Example:
                    375: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
                    376: disk = [ 'file:/n0/xen/foo-wd0,0x0,w' ]
                    377: """]]
1.28      gdt       378: 
1.127     gdt       379: Note that NetBSD by default creates only vnd[0123].  If you need more
                    380: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
                    381: dom0.
                    382: 
                    383: Note that NetBSD by default creates only xbd[0123].  If you need more
                    384: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
                    385: 
1.28      gdt       386: Virtual Networking
                    387: ------------------
                    388: 
1.111     wiz       389: Xen provides virtual Ethernets, each of which connects the dom0 and a
1.46      gdt       390: domU.  For each virtual network, there is an interface "xvifN.M" in
1.158     maxv      391: the dom0, and a matching interface xennetM (NetBSD name) in domU index N.
                    392: The interfaces behave as if there is an Ethernet with two
1.111     wiz       393: adapters connected.  From this primitive, one can construct various
1.46      gdt       394: configurations.  We focus on two common and useful cases for which
                    395: there are existing scripts: bridging and NAT.
1.28      gdt       396: 
1.48      gdt       397: With bridging (in the example above), the domU perceives itself to be
                    398: on the same network as the dom0.  For server virtualization, this is
                    399: usually best.  Bridging is accomplished by creating a bridge(4) device
                    400: and adding the dom0's physical interface and the various xvifN.0
                    401: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
                    402: config file.  The bridge must be set up already in the dom0; an
                    403: example /etc/ifconfig.bridge0 is:
1.46      gdt       404: 
1.157     maxv      405: [[!template id=filecontent name="/etc/ifconfig.bridge0" text="""
                    406: create
                    407: up
                    408: !brconfig bridge0 add wm0
                    409: """]]
1.28      gdt       410: 
                    411: With NAT, the domU perceives itself to be behind a NAT running on the
                    412: dom0.  This is often appropriate when running Xen on a workstation.
1.48      gdt       413: TODO: NAT appears to be configured by "vif = [ '' ]".
1.28      gdt       414: 
1.49      gdt       415: The MAC address specified is the one used for the interface in the new
1.53      gdt       416: domain.  The interface in dom0 will use this address XOR'd with
1.49      gdt       417: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
                    418: 
1.48      gdt       419: Starting domains automatically
                    420: ------------------------------
1.28      gdt       421: 
1.158     maxv      422: To start domains `domU-netbsd` and `domU-linux` at boot and shut them
                    423: down cleanly on dom0 shutdown, add the following in rc.conf:
1.28      gdt       424: 
1.158     maxv      425: [[!template id=filecontent name="/etc/rc.conf" text="""
                    426: xendomains="domU-netbsd domU-linux"
                    427: """]]
1.28      gdt       428: 
1.160     maxv      429: #Creating a domU
1.14      gdt       430: 
                    431: Creating domUs is almost entirely independent of operating system.  We
1.49      gdt       432: have already presented the basics of config files.  Note that you must
1.158     maxv      433: have already completed the dom0 setup so that "xl list" works.
1.14      gdt       434: 
1.158     maxv      435: Creating a NetBSD domU
                    436: ----------------------
1.1       mspo      437: 
1.49      gdt       438: See the earlier config file, and adjust memory.  Decide on how much
1.162     gutterid  439: storage you will provide, and prepare it (file or LVM).
1.49      gdt       440: 
1.111     wiz       441: While the kernel will be obtained from the dom0 file system, the same
1.49      gdt       442: file should be present in the domU as /netbsd so that tools like
                    443: savecore(8) can work.   (This is helpful but not necessary.)
                    444: 
                    445: The kernel must be specifically for Xen and for use as a domU.  The
                    446: i386 and amd64 provide the following kernels:
                    447: 
                    448:         i386 XEN3PAE_DOMU
1.95      gdt       449:         amd64 XEN3_DOMU
1.5       mspo      450: 
1.49      gdt       451: This will boot NetBSD, but this is not that useful if the disk is
                    452: empty.  One approach is to unpack sets onto the disk outside of xen
                    453: (by mounting it, just as you would prepare a physical disk for a
                    454: system you can't run the installer on).
                    455: 
                    456: A second approach is to run an INSTALL kernel, which has a miniroot
                    457: and can load sets from the network.  To do this, copy the INSTALL
                    458: kernel to / and change the kernel line in the config file to:
1.5       mspo      459: 
1.49      gdt       460:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
1.5       mspo      461: 
1.163     gson      462: Then, start the domain as "xl create -c configfile".
1.1       mspo      463: 
1.49      gdt       464: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
                    465: line should be used in the config file.
1.1       mspo      466: 
1.3       mspo      467:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
1.1       mspo      468: 
                    469: After booting the domain, the option to install via CDROM may be
1.49      gdt       470: selected.  The CDROM device should be changed to `xbd1d`.
1.1       mspo      471: 
1.49      gdt       472: Once done installing, "halt -p" the new domain (don't reboot or halt,
                    473: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
                    474: config file), switch the config file back to the XEN3_DOMU kernel,
                    475: and start the new domain again. Now it should be able to use "root on
                    476: xbd0a" and you should have a, functional NetBSD domU.
1.1       mspo      477: 
1.49      gdt       478: TODO: check if this is still accurate.
1.1       mspo      479: When the new domain is booting you'll see some warnings about *wscons*
                    480: and the pseudo-terminals. These can be fixed by editing the files
1.5       mspo      481: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
                    482: `/etc/ttys`, except *console*, like this:
1.1       mspo      483: 
1.3       mspo      484:     console "/usr/libexec/getty Pc"         vt100   on secure
                    485:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
                    486:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
                    487:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
                    488:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
1.1       mspo      489: 
1.5       mspo      490: Finally, all screens must be commented out from `/etc/wscons.conf`.
1.1       mspo      491: 
                    492: It is also desirable to add
                    493: 
1.49      gdt       494:         powerd=YES
1.1       mspo      495: 
1.5       mspo      496: in rc.conf. This way, the domain will be properly shut down if
1.53      gdt       497: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
1.1       mspo      498: 
1.92      gdt       499: It is not strictly necessary to have a kernel (as /netbsd) in the domU
1.111     wiz       500: file system.  However, various programs (e.g. netstat) will use that
1.92      gdt       501: kernel to look up symbols to read from kernel virtual memory.  If
                    502: /netbsd is not the running kernel, those lookups will fail.  (This is
                    503: not really a Xen-specific issue, but because the domU kernel is
                    504: obtained from the dom0, it is far more likely to be out of sync or
                    505: missing with Xen.)
                    506: 
1.158     maxv      507: Creating a Linux domU
                    508: ---------------------
1.1       mspo      509: 
                    510: Creating unprivileged Linux domains isn't much different from
                    511: unprivileged NetBSD domains, but there are some details to know.
                    512: 
                    513: First, the second parameter passed to the disk declaration (the '0x1' in
                    514: the example below)
                    515: 
1.3       mspo      516:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
1.1       mspo      517: 
                    518: does matter to Linux. It wants a Linux device number here (e.g. 0x300
1.49      gdt       519: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
                    520: So, hda1 which has major 3 and minor 1 on a Linux system will have
                    521: device number 0x301.  Alternatively, devices names can be used (hda,
                    522: hdb, ...)  as xentools has a table to map these names to devices
                    523: numbers.  To export a partition to a Linux guest we can use:
1.1       mspo      524: 
1.49      gdt       525:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
                    526:         root = "/dev/hda1 ro"
1.1       mspo      527: 
                    528: and it will appear as /dev/hda on the Linux system, and be used as root
                    529: partition.
                    530: 
1.49      gdt       531: To install the Linux system on the partition to be exported to the
                    532: guest domain, the following method can be used: install
                    533: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
                    534: that will be the root partition of your Linux domain, and mount it.
                    535: Then copy the files from a working Linux system, make adjustments in
                    536: `/etc` (fstab, network config).  It should also be possible to extract
                    537: binary packages such as .rpm or .deb directly to the mounted partition
                    538: using the appropriate tool, possibly running under NetBSD's Linux
1.111     wiz       539: emulation.  Once the file system has been populated, umount it.  If
                    540: desirable, the file system can be converted to ext3 using tune2fs -j.
1.49      gdt       541: It should now be possible to boot the Linux guest domain, using one of
                    542: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
1.1       mspo      543: 
1.111     wiz       544: To get the Linux console right, you need to add:
1.1       mspo      545: 
1.3       mspo      546:     extra = "xencons=tty1"
1.1       mspo      547: 
1.111     wiz       548: to your configuration since not all Linux distributions auto-attach a
1.1       mspo      549: tty to the xen console.
                    550: 
1.158     maxv      551: Creating a Solaris domU
                    552: -----------------------
1.1       mspo      553: 
1.50      gdt       554: See possibly outdated
                    555: [Solaris domU instructions](/ports/xen/howto-solaris/).
1.5       mspo      556: 
1.1       mspo      557: 
1.52      gdt       558: PCI passthrough: Using PCI devices in guest domains
                    559: ---------------------------------------------------
1.1       mspo      560: 
1.53      gdt       561: The dom0 can give other domains access to selected PCI
1.52      gdt       562: devices. This can allow, for example, a non-privileged domain to have
                    563: access to a physical network interface or disk controller.  However,
                    564: keep in mind that giving a domain access to a PCI device most likely
                    565: will give the domain read/write access to the whole physical memory,
                    566: as PCs don't have an IOMMU to restrict memory access to DMA-capable
1.53      gdt       567: device.  Also, it's not possible to export ISA devices to non-dom0
1.52      gdt       568: domains, which means that the primary VGA adapter can't be exported.
                    569: A guest domain trying to access the VGA registers will panic.
                    570: 
1.53      gdt       571: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
1.52      gdt       572: not been ported to later versions at this time.
                    573: 
                    574: For a PCI device to be exported to a domU, is has to be attached to
                    575: the "pciback" driver in dom0.  Devices passed to the dom0 via the
                    576: pciback.hide boot parameter will attach to "pciback" instead of the
                    577: usual driver.  The list of devices is specified as "(bus:dev.func)",
1.5       mspo      578: where bus and dev are 2-digit hexadecimal numbers, and func a
                    579: single-digit number:
1.1       mspo      580: 
1.52      gdt       581:         pciback.hide=(00:0a.0)(00:06.0)
1.1       mspo      582: 
1.52      gdt       583: pciback devices should show up in the dom0's boot messages, and the
1.5       mspo      584: devices should be listed in the `/kern/xen/pci` directory.
1.1       mspo      585: 
1.52      gdt       586: PCI devices to be exported to a domU are listed in the "pci" array of
                    587: the domU's config file, with the format "0000:bus:dev.func".
1.1       mspo      588: 
1.52      gdt       589:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
1.1       mspo      590: 
1.52      gdt       591: In the domU an "xpci" device will show up, to which one or more pci
1.111     wiz       592: buses will attach.  Then the PCI drivers will attach to PCI buses as
1.52      gdt       593: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
                    594: or any PCI drivers built in by default; you have to build your own
                    595: kernel to use PCI devices in a domU.  Here's a kernel config example;
                    596: note that only the "xpci" lines are unusual.
                    597: 
                    598:         include         "arch/i386/conf/XEN3_DOMU"
                    599: 
1.111     wiz       600:         # Add support for PCI buses to the XEN3_DOMU kernel
1.52      gdt       601:         xpci* at xenbus ?
                    602:         pci* at xpci ?
                    603: 
                    604:         # PCI USB controllers
                    605:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
                    606: 
                    607:         # USB bus support
                    608:         usb*    at uhci?
                    609: 
                    610:         # USB Hubs
                    611:         uhub*   at usb?
                    612:         uhub*   at uhub? port ? configuration ? interface ?
                    613: 
                    614:         # USB Mass Storage
                    615:         umass*  at uhub? port ? configuration ? interface ?
                    616:         wd*     at umass?
                    617:         # SCSI controllers
                    618:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
                    619: 
                    620:         # SCSI bus support (for both ahc and umass)
                    621:         scsibus* at scsi?
                    622: 
                    623:         # SCSI devices
                    624:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
                    625:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
1.1       mspo      626: 
                    627: 
1.173     gdt       628: # Specific Issues
                    629: 
                    630: ## domU
                    631: 
1.176   ! gdt       632: [NetBSD 5 is known to panic.](http://mail-index.netbsd.org/port-xen/2018/04/17/msg009181.html)
1.175     gdt       633: (However, NetBSD 5 systems should be updated to a supported version.)
1.173     gdt       634: 
                    635: # NetBSD as a domU in a VPS
1.28      gdt       636: 
                    637: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
                    638: hardware.  This section explains how to deal with Xen in a domU as a
                    639: virtual private server where you do not control or have access to the
1.70      gdt       640: dom0.  This is not intended to be an exhaustive list of VPS providers;
                    641: only a few are mentioned that specifically support NetBSD.
1.28      gdt       642: 
1.52      gdt       643: VPS operators provide varying degrees of access and mechanisms for
                    644: configuration.  The big issue is usually how one controls which kernel
1.111     wiz       645: is booted, because the kernel is nominally in the dom0 file system (to
                    646: which VPS users do not normally have access).  A second issue is how
1.70      gdt       647: to install NetBSD.
1.52      gdt       648: A VPS user may want to compile a kernel for security updates, to run
                    649: npf, run IPsec, or any other reason why someone would want to change
                    650: their kernel.
                    651: 
1.111     wiz       652: One approach is to have an administrative interface to upload a kernel,
1.68      gdt       653: or to select from a prepopulated list.  Other approaches are pygrub
1.59      gdt       654: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
1.111     wiz       655: kernel from the domU file system.  This is closer to a regular physical
1.59      gdt       656: computer, where someone who controls a machine can replace the kernel.
1.52      gdt       657: 
1.74      gdt       658: A second issue is multiple CPUs.  With NetBSD 6, domUs support
                    659: multiple vcpus, and it is typical for VPS providers to enable multiple
                    660: CPUs for NetBSD domUs.
                    661: 
1.59      gdt       662: pvgrub
                    663: ------
                    664: 
                    665: pvgrub is a version of grub that uses PV operations instead of BIOS
                    666: calls.  It is booted from the dom0 as the domU kernel, and then reads
1.111     wiz       667: /grub/menu.lst and loads a kernel from the domU file system.
1.59      gdt       668: 
1.70      gdt       669: [Panix](http://www.panix.com/) lets users use pvgrub.  Panix reports
1.71      gdt       670: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
                    671: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
1.70      gdt       672: page](http://www.panix.com/v-colo/grub.html), which describes only
1.74      gdt       673: Linux but should be updated to cover NetBSD :-).
1.70      gdt       674: 
                    675: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
                    676: their own kernel.  See then [prgmr.com NetBSD
1.74      gdt       677: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
                    678: (which is in need of updating).
1.59      gdt       679: 
1.70      gdt       680: It appears that [grub's FFS
                    681: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
                    682: does not support all aspects of modern FFS, but there are also reports
1.72      gdt       683: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
1.70      gdt       684: partition for the kernel with the intent that grub can understand it,
                    685: which leads to /netbsd not being the actual kernel.  One must remember
1.111     wiz       686: to update the special boot partition.
1.59      gdt       687: 
1.168     wiki      688: pygrub
                    689: -------
                    690: 
                    691: pygrub runs in the dom0 and looks into the domU file system.  This
                    692: implies that the domU must have a kernel in a file system in a format
                    693: known to pygrub.
                    694: 
                    695: pygrub doesn't seem to work to load Linux images under NetBSD dom0,
                    696: and is inherently less secure than pvgrub due to running inside dom0. For both these
                    697: reasons, pygrub should not be used, and is only still present so that
                    698: historical DomU images using it still work.
                    699: 
                    700: As of 2014, pygrub seems to be of mostly historical
                    701: interest. New DomUs should use pvgrub.
                    702: 
1.59      gdt       703: Amazon
                    704: ------
                    705: 
1.143     wiki      706: See the [Amazon EC2 page](/amazon_ec2/).

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