Annotation of wikisrc/ports/xen/howto.mdwn, revision 1.203

1.203   ! gdt         1: [[!meta title="Xen Status and HowTo"]]
1.144     maxv        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: 
1.203   ! gdt        11: This document provides status on what Xen things work on NetBSD
        !            12: (upstream documentation might say something works if it works on some
        !            13: particular Linux system).
        !            14: 
        !            15: This document is also a HOWTO that presumes a basic familiarity with
        !            16: the Xen system architecture, with installing NetBSD on amd64 hardware,
        !            17: and with installing software from pkgsrc.  See also the [Xen
1.182     gdt        18: website](http://www.xenproject.org/).
                     19: 
1.203   ! gdt        20: If this document says that something works, and you find that it does
        !            21: not, it is best to ask on port-xen and if you are correct to file a
        !            22: PR.
1.201     gdt        23: 
1.182     gdt        24: [[!toc]]
                     25: 
1.183     gdt        26: # Overview
1.182     gdt        27: 
                     28: The basic concept of Xen is that the hypervisor (xenkernel) runs on
                     29: the hardware, and runs a privileged domain ("dom0") that can access
                     30: disks/networking/etc.  One then runs additonal unprivileged domains
                     31: (each a "domU"), presumably to do something useful.
                     32: 
                     33: This HOWTO addresses how to run a NetBSD dom0 (and hence also build
                     34: xen itself).  It also addresses how to run domUs in that environment,
                     35: and how to deal with having a domU in a Xen environment run by someone
                     36: else and/or not running NetBSD.
                     37: 
1.186     gdt        38: There are many choices one can make; the HOWTO recommends the standard
                     39: approach and limits discussion of alternatives in many cases.
                     40: 
1.182     gdt        41: ## Guest Styles
                     42: 
                     43: Xen supports different styles of guests.
1.149     maxv       44: 
                     45: [[!table data="""
1.203   ! gdt        46: Style of guest |description                            |NetBSD dom0?   |NetBSD domU?
        !            47: PV             |Paravirtualization (hypercalls)        |yes            |yes
        !            48: HVM            |unmodified guest - domU perceives HW   |N/A            |yes
        !            49: PVHVM          |Like HVM but also PV drivers           |N/A            |current only
        !            50: PVH            |                                       |not yet        |current only
1.149     maxv       51: """]]
                     52: 
                     53: In Para-Virtualized (PV) mode, the guest OS does not attempt to access
                     54: hardware directly, but instead makes hypercalls to the hypervisor; PV
1.170     gdt        55: guests must be specifically coded for Xen.
1.182     gdt        56: See [PV](https://wiki.xen.org/wiki/Paravirtualization_(PV\)).
1.170     gdt        57: 
                     58: In HVM mode, no guest modification is required; however, hardware
                     59: support is required, such as VT-x on Intel CPUs and SVM on AMD CPUs.
1.199     gdt        60: The dom0 runs qemu to emulate hardware.  It is therefore non-sensical
                     61: to have an HVM dom0.
1.170     gdt        62: 
1.176     gdt        63: In PVHVM mode, the guest runs as HVM, but additionally can use PV
1.199     gdt        64: drivers for efficiency.  Therefore it is non-sensical for to have a
                     65: PVHVM dom0.  See [PV on HVM](https://wiki.xen.org/wiki/PV_on_HVM).
1.170     gdt        66: 
1.180     gdt        67: There have been two PVH modes: original PVH and PVHv2.  Original PVH
1.203   ! gdt        68: was based on PV mode and is no longer relevant at all.  Therefore
        !            69: PVHv2 is abreviated PVH.  PVHv2 is basically lightweight HVM with PV
        !            70: drivers.  A critical feature of it is that qemu is not needed; the
        !            71: hypervisor can do the emulation that is required.  Thus, a dom0 can be
        !            72: PVHv2.  The source code uses PVH and config files use pvh, but NB that
        !            73: this refers to PVHv2.  See
        !            74: [PVH(v2)](https://wiki.xenproject.org/wiki/PVH_(v2\)_Domu).
1.180     gdt        75: 
1.182     gdt        76: At system boot, the dom0 kernel is loaded as a module with Xen as the kernel.
1.12      gdt        77: The dom0 can start one or more domUs.  (Booting is explained in detail
                     78: in the dom0 section.)
                     79: 
1.182     gdt        80: ## CPU Architecture
                     81: 
                     82: Xen runs on x86_64 hardware (the NetBSD amd64 port).
                     83: 
                     84: There is a concept of Xen running on ARM, but there are no reports of this working with NetBSD.
                     85: 
                     86: The dom0 system should be amd64.  (Instructions for i386PAE dom0 have been removed from the HOWTO.)
1.1       mspo       87: 
1.182     gdt        88: The domU can be i386PAE or amd64.
                     89: i386PAE at one point was considered as [faster](https://lists.xen.org/archives/html/xen-devel/2012-07/msg00085.html) than amd64.
1.150     maxv       90: 
1.182     gdt        91: ## Xen Versions
1.15      gdt        92: 
1.111     wiz        93: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
1.15      gdt        94: xenkernel and xentools.  We will refer only to the kernel versions,
                     95: but note that both packages must be installed together and must have
                     96: matching versions.
                     97: 
1.145     maxv       98: Versions available in pkgsrc:
1.85      gdt        99: 
1.145     maxv      100: [[!table data="""
1.184     gdt       101: Xen Version    |Package Name   |Xen CPU Support        |EOL'ed By Upstream
                    102: 4.11           |xenkernel411   |x86_64                 |No
                    103: 4.13           |xenkernel413   |x86_64                 |No
1.145     maxv      104: """]]
1.113     gdt       105: 
1.96      gdt       106: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
                    107: 
1.166     gdt       108: Older Xen had a python-based management tool called xm, now replaced
1.176     gdt       109: by xl.
1.166     gdt       110: 
1.183     gdt       111: ## NetBSD versions
                    112: 
                    113: Xen has been supported in NetBSD for a long time, at least since 2005.
                    114: Initially Xen was PV only.
                    115: 
1.203   ! gdt       116: NetBSD has supported PV for a very long time, in both dom0 and domU.
        !           117: NetBSD >=8 supports HVM mode as a domU (HVM as a dom0 does not make
        !           118: sense).
        !           119: 
        !           120: Support for PVHVM and PVH is available only in NetBSD-current; this is
        !           121: currently somewhat experimental, although PVHVM appears reasonably
        !           122: solid.
        !           123: 
        !           124: NetBSD up to and including NetBSD 9 as a dom0 cannot safely run SMP.
        !           125: Even if one added "options MULTIPROCESSOR" and configured multiple
        !           126: vcpus, the kernel is likely to crash because of drivers without
        !           127: adequate locking.
        !           128: 
        !           129: NetBSD-current supports SMP in dom0, and XEN3_DOM0 includes "options
        !           130: MULTIPROCESSOR".
        !           131: 
        !           132: NetBSD (since NetBSD 6), when run as a domU, can run SMP, using
        !           133: multiple CPUs if provided.  The XEN3_DOMU kernel is built
        !           134: with "options MULITPROCESSOR".
1.185     gdt       135: 
1.191     gdt       136: Note that while Xen 4.13 is current, the kernel support is still
                    137: called XEN3, because the hypercall interface has not changed
                    138: significantly.
1.185     gdt       139: 
1.187     gdt       140: # Creating a NetBSD dom0
1.15      gdt       141: 
1.191     gdt       142: In order to install a NetBSD as a dom0, one first installs a normal
                    143: NetBSD system, and then pivot the install to a dom0 install by
                    144: changing the kernel and boot configuration.
1.15      gdt       145: 
1.142     gdt       146: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
                    147: 512M it worked reliably.  This does not make sense, but if you see
                    148: "not ELF" after Xen boots, try increasing dom0 RAM.
                    149: 
1.181     gdt       150: ## Installation of NetBSD
1.13      gdt       151: 
1.191     gdt       152: [Install NetBSD/amd64](/guide/inst/) just as you would if you were not
                    153: using Xen.  Therefore, use the most recent release, or a build from
                    154: the most recent stable branch.  Alternatively, use -current, being
                    155: mindful of all the usual caveats of lower stability of current, and
                    156: likely a bit more so.
1.1       mspo      157: 
1.181     gdt       158: ## Installation of Xen
1.1       mspo      159: 
1.191     gdt       160: ### Building Xen
                    161: 
1.187     gdt       162: Use the most recent version of Xen in pkgsrc, unless the DESCR says that it is not suitable.
1.186     gdt       163: Therefore, choose 4.13.
                    164: In the dom0, install xenkernel413 and xentools413 from pkgsrc.
1.155     maxv      165: 
1.186     gdt       166: Once this is done, copy the Xen kernel from where pkgsrc puts it to
                    167: where the boot process will be able to find it:
1.155     maxv      168: 
                    169: [[!template id=programlisting text="""
1.186     gdt       170: # cp -p /usr/pkg/xen413-kernel/xen.gz /
1.155     maxv      171: """]]
                    172: 
1.191     gdt       173: Then, place a NetBSD XEN3_DOM0 kernel in the `/` directory. Such
                    174: kernel can either be taken from a local release build.sh run, compiled
                    175: manually, or downloaded from the NetBSD FTP, for example at:
1.159     maxv      176: 
                    177: [[!template id=programlisting text="""
1.186     gdt       178: ftp.netbsd.org/pub/NetBSD/NetBSD-9.1/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
1.159     maxv      179: """]]
1.20      gdt       180: 
1.191     gdt       181: ### Configuring booting
                    182: 
1.198     gdt       183: Read boot.cfg(8) carefully.  Add lines to /boot.cfg to boot Xen,
                    184: adjusting for your root filesystem:
1.20      gdt       185: 
1.157     maxv      186: [[!template id=filecontent name="/boot.cfg" text="""
1.198     gdt       187: menu=Xen:load /netbsd-XEN3_DOM0.gz root=wd0a console=pc;multiboot /xen.gz dom0_mem=512M
                    188: menu=Xen single user:load /netbsd-XEN3_DOM0.gz root=wd0a console=pc -s;multiboot /xen.gz dom0_mem=512M
1.152     maxv      189: """]]
1.20      gdt       190: 
1.159     maxv      191: This specifies that the dom0 should have 512MB of ram, leaving the rest
1.187     gdt       192: to be allocated for domUs.
                    193: 
1.191     gdt       194: NB: This says add, not replace, so that you will be able to more
                    195: easily boot a NetBSD kernel without Xen.  Once Xen boots ok, you may
                    196: want to set it as default.  It is highly likely that you will have
                    197: trouble at some point, and keeping an up-to-date GENERIC for use in
                    198: fixing problems is the standard prudent approach.
                    199: 
1.194     gdt       200: \todo Explain why rndseed is not set with Xen as part of the dom0
                    201: subconfiguration.
                    202: 
1.198     gdt       203: Note that you are likely to have to set root= because the boot device
                    204: from /boot is not passed via Xen to the dom0 kernel.  With one disk,
                    205: it will work, but e.g. plugging in USB disk to a machine with root on
                    206: wd0a causes boot to fail.
                    207: 
1.195     gdt       208: Beware that userconf statements must be attached to the dom0 load, and
                    209: may not be at top-level, because then they would try to configure the
                    210: hypervisor, if there is a way to pass them via multiboot .  It appears
                    211: that adding `userconf=pckbc` to `/boot.cfg` causes Xen to crash very
                    212: early with a heap overflow.
                    213: 
1.191     gdt       214: ### Console selection
                    215: 
                    216: See boot_console(8).  Understand that you should start from a place of
                    217: having console setup correct for booting GENERIC before trying to
                    218: configure Xen.
                    219: 
1.193     gdt       220: Generally for GENERIC, one sets the console in bootxx_ffsv1 or
                    221: equivalent, and this is passed on to /boot (where one typically does
                    222: not set the console).  This configuration of bootxx_ffsv1 should also
                    223: be in place for Xen systems, to allow seeing messages from /boot and
                    224: use of a keyboard to select a line from the menu.  And, one should
                    225: have a working boot path to GENERIC for rescue situations.
                    226: 
                    227: With GENERIC, the boot options are passed on to /netbsd, but there is
                    228: currently no mechanism to pass these via multiboot to the hypervisor.
                    229: Thus, in addition to configuring the console in the boot blocks, one
                    230: must also configure it for Xen.
                    231: 
                    232: By default, the hypervisor (Xen itself) will use some sort of vga
                    233: device as the console, much like GENERIC uses by default.  The vga
                    234: console is relinquished at the conclusion of hypervisor boot, before
1.194     gdt       235: the dom0 is started.  Xen when using a vga console does not process
                    236: console input.
1.77      gdt       237: 
1.193     gdt       238: The hypervisor can be configured to use a serial port console, e.g.
1.157     maxv      239: [[!template id=filecontent name="/boot.cfg" text="""
1.202     gson      240: menu=Xen:load /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
1.152     maxv      241: """]]
1.194     gdt       242: This example uses the first serial port (Xen counts from 1; this is
1.193     gdt       243: what NetBSD would call com0), and sets speed and parity.  (The dom0 is
                    244: then configured to use the same serial port in this example.)
                    245: 
1.194     gdt       246: With the hypervisor configured for a serial console, it can get input,
                    247: and there is a notion of passing this input to the dom0.  \todo
                    248: Explain why, if Xen has a serial console, the dom0 console is
                    249: typically also configured to open that same serial port, instead of
                    250: getting the passthrough input via the xen console.
                    251: 
1.193     gdt       252: One also configures the console for the dom0.  While one might expect
                    253: console=pc to be default, following behavior of GENERIC, a hasty read
                    254: of the code suggests there is no default and booting without a
                    255: selected console might lead to a panic.  Also, there is merit in
                    256: explicit configuration.  Therefore the standard approach is to place
                    257: console=pc as part of the load statement for the dom0 kernel, or
                    258: alternatively console=com0.
                    259: 
                    260: The NetBSD dom0 kernel will attach xencons(4) (the man page does not
                    261: exist), but this is not used as a console.  It is used to obtain the
                    262: messages from the hypervisor's console; run `xl dmesg` to see them.
1.191     gdt       263: 
                    264: ### Tuning
                    265: 
1.157     maxv      266: In an attempt to add performance, one can also add `dom0_max_vcpus=1 dom0_vcpus_pin`,
1.37      gdt       267: to force only one vcpu to be provided (since NetBSD dom0 can't use
1.158     maxv      268: more) and to pin that vcpu to a physical CPU. Xen has
1.178     gdt       269: [many boot options](http://xenbits.xenproject.org/docs/4.13-testing/misc/xen-command-line.html),
1.111     wiz       270: and other than dom0 memory and max_vcpus, they are generally not
1.93      gdt       271: necessary.
1.191     gdt       272: \todo Revisit this advice with current.
                    273: \todo Explain if anyone has ever actually measured that this helps.
                    274: 
                    275: ### rc.conf
1.93      gdt       276: 
1.186     gdt       277: Ensure that the boot scripts installed in
                    278: `/usr/pkg/share/examples/rc.d` are in `/etc/rc.d`, either because you
                    279: have `PKG_RCD_SCRIPTS=yes`, or manually.  (This is not special to Xen,
                    280: but a normal part of pkgsrc usage.)
1.159     maxv      281: 
1.186     gdt       282: Set `xencommons=YES` in rc.conf:
1.31      gdt       283: 
1.157     maxv      284: [[!template id=filecontent name="/etc/rc.conf" text="""
1.152     maxv      285: xencommons=YES
                    286: """]]
1.31      gdt       287: 
1.187     gdt       288: \todo Recommend for/against xen-watchdog.
                    289: 
1.191     gdt       290: ### Testing
                    291: 
1.158     maxv      292: Now, reboot so that you are running a DOM0 kernel under Xen, rather
                    293: than GENERIC without Xen.
1.157     maxv      294: 
1.158     maxv      295: Once the reboot is done, use `xl` to inspect Xen's boot messages,
1.157     maxv      296: available resources, and running domains.  For example:
1.34      gdt       297: 
1.153     maxv      298: [[!template id=programlisting text="""
                    299: # xl dmesg
                    300: ... xen's boot info ...
                    301: # xl info
                    302: ... available memory, etc ...
                    303: # xl list
                    304: Name              Id  Mem(MB)  CPU  State  Time(s)  Console
                    305: Domain-0           0       64    0  r----     58.1
                    306: """]]
                    307: 
                    308: Xen logs will be in /var/log/xen.
1.33      gdt       309: 
1.88      gdt       310: ### Issues with xencommons
                    311: 
1.157     maxv      312: `xencommons` starts `xenstored`, which stores data on behalf of dom0 and
1.88      gdt       313: domUs.  It does not currently work to stop and start xenstored.
                    314: Certainly all domUs should be shutdown first, following the sort order
                    315: of the rc.d scripts.  However, the dom0 sets up state with xenstored,
                    316: and is not notified when xenstored exits, leading to not recreating
                    317: the state when the new xenstored starts.  Until there's a mechanism to
                    318: make this work, one should not expect to be able to restart xenstored
                    319: (and thus xencommons).  There is currently no reason to expect that
                    320: this will get fixed any time soon.
1.187     gdt       321: \todo Confirm if this is still true in 2020.
1.82      gdt       322: 
1.181     gdt       323: ## Xen-specific NetBSD issues
1.40      gdt       324: 
                    325: There are (at least) two additional things different about NetBSD as a
                    326: dom0 kernel compared to hardware.
                    327: 
1.191     gdt       328: One is that through NetBSD 9 the module ABI is different because some
                    329: of the #defines change, so there are separate sets of modules in
                    330: /stand.  In NetBSD-current, there is only one set of modules.
1.40      gdt       331: 
                    332: The other difference is that XEN3_DOM0 does not have exactly the same
                    333: options as GENERIC.  While it is debatable whether or not this is a
                    334: bug, users should be aware of this and can simply add missing config
                    335: items if desired.
                    336: 
1.187     gdt       337: Finally, there have been occasional reports of trouble with X11
                    338: servers in NetBSD as a dom0.
                    339: 
                    340: ## Updating Xen in a dom0
                    341: 
                    342: Basically, update the xenkernel and xentools packages and copy the new
                    343: Xen kernel into place, and reboot.  This procedure should be usable to
                    344: update to a new Xen release, but the reader is reminded that having a
                    345: non-Xen boot methods was recommended earlier.
                    346: 
1.181     gdt       347: ## Updating NetBSD in a dom0
1.15      gdt       348: 
                    349: This is just like updating NetBSD on bare hardware, assuming the new
                    350: version supports the version of Xen you are running.  Generally, one
                    351: replaces the kernel and reboots, and then overlays userland binaries
1.157     maxv      352: and adjusts `/etc`.
1.15      gdt       353: 
1.191     gdt       354: Note that one should update both the non-Xen kernel typically used for
                    355: rescue purposes, as well as the DOM0 kernel used with Xen.
1.15      gdt       356: 
1.187     gdt       357: ## anita (for testing NetBSD)
                    358: 
                    359: With a NetBSD dom0, even without any domUs, one should be able to run
                    360: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
                    361: root, because anita must create a domU):
                    362: 
                    363: [[!template id=programlisting text="""
                    364: anita --vmm=xl test file:///usr/obj/i386/
                    365: """]]
                    366: 
                    367: # Unprivileged domains (domU)
1.28      gdt       368: 
                    369: This section describes general concepts about domUs.  It does not
1.33      gdt       370: address specific domU operating systems or how to install them.  The
1.157     maxv      371: config files for domUs are typically in `/usr/pkg/etc/xen`, and are
1.60      wiki      372: typically named so that the file name, domU name and the domU's host
1.33      gdt       373: name match.
                    374: 
1.111     wiz       375: The domU is provided with CPU and memory by Xen, configured by the
1.33      gdt       376: dom0.  The domU is provided with disk and network by the dom0,
                    377: mediated by Xen, and configured in the dom0.
                    378: 
                    379: Entropy in domUs can be an issue; physical disks and network are on
                    380: the dom0.  NetBSD's /dev/random system works, but is often challenged.
                    381: 
1.181     gdt       382: ## Config files
1.48      gdt       383: 
1.200     gdt       384: See /usr/pkg/share/examples/xen/xlexample* for a very small number of
                    385: examples for running GNU/Linux.
1.48      gdt       386: 
1.157     maxv      387: The following is an example minimal domain configuration file. The domU
                    388: serves as a network file server.
                    389: 
                    390: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
                    391: name = "domU-id"
                    392: kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
                    393: memory = 1024
                    394: vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
                    395: disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
                    396:          'file:/n0/xen/foo-wd1,0x1,w' ]
                    397: """]]
1.48      gdt       398: 
1.157     maxv      399: The domain will have name given in the `name` setting.  The kernel has the
1.48      gdt       400: host/domU name in it, so that on the dom0 one can update the various
1.157     maxv      401: domUs independently.  The `vif` line causes an interface to be provided,
1.48      gdt       402: with a specific mac address (do not reuse MAC addresses!), in bridge
                    403: mode.  Two disks are provided, and they are both writable; the bits
                    404: are stored in files and Xen attaches them to a vnd(4) device in the
1.111     wiz       405: dom0 on domain creation.  The system treats xbd0 as the boot device
1.48      gdt       406: without needing explicit configuration.
                    407: 
1.200     gdt       408: There is not type line; that implicitly defines a pv domU.  Otherwise,
                    409: one sets type to the lower-case version of the domU type in the table
1.201     gdt       410: above, e.g. `type = "hvm"`.
1.187     gdt       411: 
1.163     gson      412: By convention, domain config files are kept in `/usr/pkg/etc/xen`.  Note
1.157     maxv      413: that "xl create" takes the name of a config file, while other commands
                    414: take the name of a domain.
                    415: 
                    416: Examples of commands:
1.48      gdt       417: 
1.157     maxv      418: [[!template id=programlisting text="""
1.163     gson      419: xl create /usr/pkg/etc/xen/foo
                    420: xl console domU-id
                    421: xl create -c /usr/pkg/etc/xen/foo
                    422: xl shutdown domU-id
1.157     maxv      423: xl list
                    424: """]]
                    425: 
                    426: Typing `^]` will exit the console session.  Shutting down a domain is
1.48      gdt       427: equivalent to pushing the power button; a NetBSD domU will receive a
                    428: power-press event and do a clean shutdown.  Shutting down the dom0
                    429: will trigger controlled shutdowns of all configured domUs.
                    430: 
1.181     gdt       431: ## CPU and memory
1.33      gdt       432: 
1.169     plunky    433: A domain is provided with some number of vcpus, up to the number
1.157     maxv      434: of CPUs seen by the hypervisor. For a domU, it is controlled
1.48      gdt       435: from the config file by the "vcpus = N" directive.
                    436: 
                    437: A domain is provided with memory; this is controlled in the config
                    438: file by "memory = N" (in megabytes).  In the straightforward case, the
                    439: sum of the the memory allocated to the dom0 and all domUs must be less
1.33      gdt       440: than the available memory.
                    441: 
1.197     gdt       442: ## Balloon driver
                    443: 
                    444: Xen provides a `balloon` driver, which can be used to let domains use
                    445: more memory temporarily.
                    446: 
                    447: \todo Explain how to set up a aystem to use the balloon scheme in a
                    448: useful manner.
1.28      gdt       449: 
1.181     gdt       450: ## Virtual disks
1.28      gdt       451: 
1.158     maxv      452: In domU config files, the disks are defined as a sequence of 3-tuples:
                    453: 
                    454:  * The first element is "method:/path/to/disk". Common methods are
1.162     gutterid  455:    "file:" for a file-backed vnd, and "phy:" for something that is already
                    456:    a device, such as an LVM logical volume.
1.158     maxv      457: 
                    458:  * The second element is an artifact of how virtual disks are passed to
                    459:    Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
                    460:    are given a device name to associate with the disk, and values like
                    461:    "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
                    462:    as xbd0, the second as xbd1, and so on.  However, xl demands a
                    463:    second argument.  The name given is converted to a major/minor by
                    464:    calling stat(2) on the name in /dev and this is passed to the domU.
                    465:    In the general case, the dom0 and domU can be different operating
                    466:    systems, and it is an unwarranted assumption that they have consistent
                    467:    numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
                    468:    as both dom0 and domU, using values of 0x0 for the first disk and 0x1
                    469:    for the second works fine and avoids this issue.  For a GNU/Linux
                    470:    guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
                    471:    /dev/hda1.
1.48      gdt       472: 
1.158     maxv      473:  * The third element is "w" for writable disks, and "r" for read-only
                    474:    disks.
                    475: 
                    476: Example:
                    477: [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
                    478: disk = [ 'file:/n0/xen/foo-wd0,0x0,w' ]
                    479: """]]
1.28      gdt       480: 
1.127     gdt       481: Note that NetBSD by default creates only vnd[0123].  If you need more
                    482: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
                    483: dom0.
                    484: 
1.187     gdt       485: ## Virtual Networking
1.28      gdt       486: 
1.111     wiz       487: Xen provides virtual Ethernets, each of which connects the dom0 and a
1.46      gdt       488: domU.  For each virtual network, there is an interface "xvifN.M" in
1.158     maxv      489: the dom0, and a matching interface xennetM (NetBSD name) in domU index N.
                    490: The interfaces behave as if there is an Ethernet with two
1.111     wiz       491: adapters connected.  From this primitive, one can construct various
1.46      gdt       492: configurations.  We focus on two common and useful cases for which
                    493: there are existing scripts: bridging and NAT.
1.28      gdt       494: 
1.48      gdt       495: With bridging (in the example above), the domU perceives itself to be
                    496: on the same network as the dom0.  For server virtualization, this is
                    497: usually best.  Bridging is accomplished by creating a bridge(4) device
                    498: and adding the dom0's physical interface and the various xvifN.0
                    499: interfaces to the bridge.  One specifies "bridge=bridge0" in the domU
                    500: config file.  The bridge must be set up already in the dom0; an
                    501: example /etc/ifconfig.bridge0 is:
1.46      gdt       502: 
1.157     maxv      503: [[!template id=filecontent name="/etc/ifconfig.bridge0" text="""
                    504: create
                    505: up
                    506: !brconfig bridge0 add wm0
                    507: """]]
1.28      gdt       508: 
                    509: With NAT, the domU perceives itself to be behind a NAT running on the
                    510: dom0.  This is often appropriate when running Xen on a workstation.
1.48      gdt       511: TODO: NAT appears to be configured by "vif = [ '' ]".
1.28      gdt       512: 
1.49      gdt       513: The MAC address specified is the one used for the interface in the new
1.53      gdt       514: domain.  The interface in dom0 will use this address XOR'd with
1.49      gdt       515: 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
                    516: 
1.187     gdt       517: ## Starting domains automatically
1.28      gdt       518: 
1.158     maxv      519: To start domains `domU-netbsd` and `domU-linux` at boot and shut them
                    520: down cleanly on dom0 shutdown, add the following in rc.conf:
1.28      gdt       521: 
1.158     maxv      522: [[!template id=filecontent name="/etc/rc.conf" text="""
                    523: xendomains="domU-netbsd domU-linux"
                    524: """]]
1.28      gdt       525: 
1.188     gdt       526: # domU setup for specific systems
1.14      gdt       527: 
                    528: Creating domUs is almost entirely independent of operating system.  We
1.188     gdt       529: have already presented the basics of config files in the previous system.
                    530: 
                    531: Of course, this section presumes that you have a working dom0.
1.14      gdt       532: 
1.187     gdt       533: ## Creating a NetBSD PV domU
                    534: 
1.49      gdt       535: See the earlier config file, and adjust memory.  Decide on how much
1.162     gutterid  536: storage you will provide, and prepare it (file or LVM).
1.49      gdt       537: 
1.111     wiz       538: While the kernel will be obtained from the dom0 file system, the same
1.49      gdt       539: file should be present in the domU as /netbsd so that tools like
                    540: savecore(8) can work.   (This is helpful but not necessary.)
                    541: 
1.188     gdt       542: The kernel must be specifically built for Xen, to use PV interfacesas
                    543: a domU.  NetBSD release builds provide the following kernels:
1.49      gdt       544: 
                    545:         i386 XEN3PAE_DOMU
1.95      gdt       546:         amd64 XEN3_DOMU
1.5       mspo      547: 
1.49      gdt       548: This will boot NetBSD, but this is not that useful if the disk is
1.188     gdt       549: empty.  One approach is to unpack sets onto the disk outside of Xen
1.49      gdt       550: (by mounting it, just as you would prepare a physical disk for a
                    551: system you can't run the installer on).
                    552: 
                    553: A second approach is to run an INSTALL kernel, which has a miniroot
                    554: and can load sets from the network.  To do this, copy the INSTALL
                    555: kernel to / and change the kernel line in the config file to:
1.5       mspo      556: 
1.49      gdt       557:         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
1.5       mspo      558: 
1.163     gson      559: Then, start the domain as "xl create -c configfile".
1.1       mspo      560: 
1.49      gdt       561: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
                    562: line should be used in the config file.
1.1       mspo      563: 
1.3       mspo      564:     disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
1.1       mspo      565: 
                    566: After booting the domain, the option to install via CDROM may be
1.49      gdt       567: selected.  The CDROM device should be changed to `xbd1d`.
1.1       mspo      568: 
1.188     gdt       569: Once done installing, "halt -p" the new domain (don't reboot or halt:
1.49      gdt       570: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
                    571: config file), switch the config file back to the XEN3_DOMU kernel,
                    572: and start the new domain again. Now it should be able to use "root on
1.188     gdt       573: xbd0a" and you should have a functional NetBSD domU.
1.1       mspo      574: 
1.49      gdt       575: TODO: check if this is still accurate.
1.1       mspo      576: When the new domain is booting you'll see some warnings about *wscons*
                    577: and the pseudo-terminals. These can be fixed by editing the files
1.5       mspo      578: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
                    579: `/etc/ttys`, except *console*, like this:
1.1       mspo      580: 
1.3       mspo      581:     console "/usr/libexec/getty Pc"         vt100   on secure
                    582:     ttyE0   "/usr/libexec/getty Pc"         vt220   off secure
                    583:     ttyE1   "/usr/libexec/getty Pc"         vt220   off secure
                    584:     ttyE2   "/usr/libexec/getty Pc"         vt220   off secure
                    585:     ttyE3   "/usr/libexec/getty Pc"         vt220   off secure
1.1       mspo      586: 
1.5       mspo      587: Finally, all screens must be commented out from `/etc/wscons.conf`.
1.1       mspo      588: 
1.192     gdt       589: One should also run `powerd` in a domU, but this should not need
                    590: configuring.  With powerd, the domain will run a controlled shutdown
                    591: if `xl shutdown -R` or `xl shutdown -H` is used on the dom0, via
                    592: receiving a synthetic `power button pressed` signal.  In 9 and
                    593: current, `powerd` is run by default under Xen kernels (or if ACPI is
                    594: present), and it can be added to rc.conf if not.
1.1       mspo      595: 
1.92      gdt       596: It is not strictly necessary to have a kernel (as /netbsd) in the domU
1.111     wiz       597: file system.  However, various programs (e.g. netstat) will use that
1.92      gdt       598: kernel to look up symbols to read from kernel virtual memory.  If
                    599: /netbsd is not the running kernel, those lookups will fail.  (This is
                    600: not really a Xen-specific issue, but because the domU kernel is
                    601: obtained from the dom0, it is far more likely to be out of sync or
                    602: missing with Xen.)
                    603: 
1.187     gdt       604: Note that NetBSD by default creates only xbd[0123].  If you need more
                    605: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
                    606: 
1.196     gdt       607: ## Creating a Linux PV domU
1.1       mspo      608: 
                    609: Creating unprivileged Linux domains isn't much different from
                    610: unprivileged NetBSD domains, but there are some details to know.
                    611: 
                    612: First, the second parameter passed to the disk declaration (the '0x1' in
                    613: the example below)
                    614: 
1.3       mspo      615:     disk = [ 'phy:/dev/wd0e,0x1,w' ]
1.1       mspo      616: 
                    617: does matter to Linux. It wants a Linux device number here (e.g. 0x300
1.49      gdt       618: for hda).  Linux builds device numbers as: (major \<\< 8 + minor).
                    619: So, hda1 which has major 3 and minor 1 on a Linux system will have
                    620: device number 0x301.  Alternatively, devices names can be used (hda,
                    621: hdb, ...)  as xentools has a table to map these names to devices
                    622: numbers.  To export a partition to a Linux guest we can use:
1.1       mspo      623: 
1.49      gdt       624:         disk = [ 'phy:/dev/wd0e,0x300,w' ]
                    625:         root = "/dev/hda1 ro"
1.1       mspo      626: 
                    627: and it will appear as /dev/hda on the Linux system, and be used as root
                    628: partition.
                    629: 
1.49      gdt       630: To install the Linux system on the partition to be exported to the
                    631: guest domain, the following method can be used: install
                    632: sysutils/e2fsprogs from pkgsrc.  Use mke2fs to format the partition
                    633: that will be the root partition of your Linux domain, and mount it.
                    634: Then copy the files from a working Linux system, make adjustments in
                    635: `/etc` (fstab, network config).  It should also be possible to extract
                    636: binary packages such as .rpm or .deb directly to the mounted partition
                    637: using the appropriate tool, possibly running under NetBSD's Linux
1.111     wiz       638: emulation.  Once the file system has been populated, umount it.  If
                    639: desirable, the file system can be converted to ext3 using tune2fs -j.
1.49      gdt       640: It should now be possible to boot the Linux guest domain, using one of
                    641: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
1.1       mspo      642: 
1.111     wiz       643: To get the Linux console right, you need to add:
1.1       mspo      644: 
1.3       mspo      645:     extra = "xencons=tty1"
1.1       mspo      646: 
1.111     wiz       647: to your configuration since not all Linux distributions auto-attach a
1.1       mspo      648: tty to the xen console.
                    649: 
1.180     gdt       650: ## Creating a NetBSD HVM domU
                    651: 
1.196     gdt       652: Use type='hvm', probably.  Use a GENERIC kernel within the disk image.
1.180     gdt       653: 
                    654: ## Creating a NetBSD PVH domU
                    655: 
1.196     gdt       656: This only works with a current kernel in the domU.
1.180     gdt       657: 
1.196     gdt       658: Use type='pvh'.  Probably, use a GENERIC kernel within the disk image,
                    659: which in current has PV support.
                    660: 
                    661: \todo Verify.
                    662: 
                    663: \todo Verify if one can have current PVH domU on a 9 dom0.
1.180     gdt       664: 
1.187     gdt       665: ## Creating a Solaris domU
1.1       mspo      666: 
1.50      gdt       667: See possibly outdated
                    668: [Solaris domU instructions](/ports/xen/howto-solaris/).
1.5       mspo      669: 
1.187     gdt       670: ## PCI passthrough: Using PCI devices in guest domains
1.1       mspo      671: 
1.180     gdt       672: NB: PCI passthrough only works on some Xen versions and as of 2020 it
1.196     gdt       673: is not clear that it works on any version in pkgsrc.  \todo Reports
1.180     gdt       674: confirming or denying this notion should be sent to port-xen@.
                    675: 
1.53      gdt       676: The dom0 can give other domains access to selected PCI
1.52      gdt       677: devices. This can allow, for example, a non-privileged domain to have
                    678: access to a physical network interface or disk controller.  However,
                    679: keep in mind that giving a domain access to a PCI device most likely
                    680: will give the domain read/write access to the whole physical memory,
                    681: as PCs don't have an IOMMU to restrict memory access to DMA-capable
1.53      gdt       682: device.  Also, it's not possible to export ISA devices to non-dom0
1.52      gdt       683: domains, which means that the primary VGA adapter can't be exported.
                    684: A guest domain trying to access the VGA registers will panic.
                    685: 
1.53      gdt       686: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
1.52      gdt       687: not been ported to later versions at this time.
                    688: 
                    689: For a PCI device to be exported to a domU, is has to be attached to
                    690: the "pciback" driver in dom0.  Devices passed to the dom0 via the
                    691: pciback.hide boot parameter will attach to "pciback" instead of the
                    692: usual driver.  The list of devices is specified as "(bus:dev.func)",
1.5       mspo      693: where bus and dev are 2-digit hexadecimal numbers, and func a
                    694: single-digit number:
1.1       mspo      695: 
1.52      gdt       696:         pciback.hide=(00:0a.0)(00:06.0)
1.1       mspo      697: 
1.52      gdt       698: pciback devices should show up in the dom0's boot messages, and the
1.5       mspo      699: devices should be listed in the `/kern/xen/pci` directory.
1.1       mspo      700: 
1.52      gdt       701: PCI devices to be exported to a domU are listed in the "pci" array of
                    702: the domU's config file, with the format "0000:bus:dev.func".
1.1       mspo      703: 
1.52      gdt       704:         pci = [ '0000:00:06.0', '0000:00:0a.0' ]
1.1       mspo      705: 
1.52      gdt       706: In the domU an "xpci" device will show up, to which one or more pci
1.111     wiz       707: buses will attach.  Then the PCI drivers will attach to PCI buses as
1.52      gdt       708: usual.  Note that the default NetBSD DOMU kernels do not have "xpci"
                    709: or any PCI drivers built in by default; you have to build your own
                    710: kernel to use PCI devices in a domU.  Here's a kernel config example;
                    711: note that only the "xpci" lines are unusual.
                    712: 
                    713:         include         "arch/i386/conf/XEN3_DOMU"
                    714: 
1.111     wiz       715:         # Add support for PCI buses to the XEN3_DOMU kernel
1.52      gdt       716:         xpci* at xenbus ?
                    717:         pci* at xpci ?
                    718: 
                    719:         # PCI USB controllers
                    720:         uhci*   at pci? dev ? function ?        # Universal Host Controller (Intel)
                    721: 
                    722:         # USB bus support
                    723:         usb*    at uhci?
                    724: 
                    725:         # USB Hubs
                    726:         uhub*   at usb?
                    727:         uhub*   at uhub? port ? configuration ? interface ?
                    728: 
                    729:         # USB Mass Storage
                    730:         umass*  at uhub? port ? configuration ? interface ?
                    731:         wd*     at umass?
                    732:         # SCSI controllers
                    733:         ahc*    at pci? dev ? function ?        # Adaptec [23]94x, aic78x0 SCSI
                    734: 
                    735:         # SCSI bus support (for both ahc and umass)
                    736:         scsibus* at scsi?
                    737: 
                    738:         # SCSI devices
                    739:         sd*     at scsibus? target ? lun ?      # SCSI disk drives
                    740:         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
1.1       mspo      741: 
                    742: 
1.189     gdt       743: # Miscellaneous Information
1.173     gdt       744: 
1.189     gdt       745: ## Nesting under Linux KVM
                    746: 
                    747: It is possible to run a Xen and a NetBSD dom0 under Linux KVM.  One
                    748: can enable virtio in the dom0 for greater speed.
                    749: 
                    750: ## Other nesting
                    751: 
                    752: In theory, any full emulation should be able to run Xen and a NetBSD
                    753: dom0.  The HOWTO does not currently have information about Xen XVM
                    754: mode, nvmm, qemu, Virtualbox, etc.
                    755: 
                    756: ## NetBSD 5 as domU
1.173     gdt       757: 
1.176     gdt       758: [NetBSD 5 is known to panic.](http://mail-index.netbsd.org/port-xen/2018/04/17/msg009181.html)
1.175     gdt       759: (However, NetBSD 5 systems should be updated to a supported version.)
1.173     gdt       760: 
                    761: # NetBSD as a domU in a VPS
1.28      gdt       762: 
                    763: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
                    764: hardware.  This section explains how to deal with Xen in a domU as a
                    765: virtual private server where you do not control or have access to the
1.70      gdt       766: dom0.  This is not intended to be an exhaustive list of VPS providers;
                    767: only a few are mentioned that specifically support NetBSD.
1.28      gdt       768: 
1.52      gdt       769: VPS operators provide varying degrees of access and mechanisms for
                    770: configuration.  The big issue is usually how one controls which kernel
1.111     wiz       771: is booted, because the kernel is nominally in the dom0 file system (to
                    772: which VPS users do not normally have access).  A second issue is how
1.70      gdt       773: to install NetBSD.
1.52      gdt       774: A VPS user may want to compile a kernel for security updates, to run
                    775: npf, run IPsec, or any other reason why someone would want to change
                    776: their kernel.
                    777: 
1.111     wiz       778: One approach is to have an administrative interface to upload a kernel,
1.68      gdt       779: or to select from a prepopulated list.  Other approaches are pygrub
1.59      gdt       780: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
1.111     wiz       781: kernel from the domU file system.  This is closer to a regular physical
1.59      gdt       782: computer, where someone who controls a machine can replace the kernel.
1.52      gdt       783: 
1.74      gdt       784: A second issue is multiple CPUs.  With NetBSD 6, domUs support
                    785: multiple vcpus, and it is typical for VPS providers to enable multiple
                    786: CPUs for NetBSD domUs.
                    787: 
1.180     gdt       788: ## Complexities due to Xen changes
                    789: 
                    790: Xen has many security advisories and people running Xen systems make
                    791: different choices.
                    792: 
                    793: ### stub domains
                    794: 
                    795: Some (Linux only?) dom0 systems use something called "stub domains" to
                    796: isolate qemu from the dom0 system, as a security and reliabilty
                    797: mechanism when running HVM domUs.  Somehow, NetBSD's GENERIC kernel
                    798: ends up using PIO for disks rather than DMA.  Of course, all of this
                    799: is emulated, but emulated PIO is unusably slow.  This problem is not
                    800: currently understood.
                    801: 
                    802: ### Grant tables
                    803: 
                    804: There are multiple versions of using grant tables, and some security
                    805: advisories have suggested disabling some versions.  Some versions of
                    806: NetBSD apparently only use specific versions and this can lead to
                    807: "NetBSD current doesn't run on hosting provider X" situations.
                    808: 
                    809: \todo Explain better.
                    810: 
1.181     gdt       811: ## Boot methods
                    812: 
                    813: ### pvgrub
1.59      gdt       814: 
                    815: pvgrub is a version of grub that uses PV operations instead of BIOS
                    816: calls.  It is booted from the dom0 as the domU kernel, and then reads
1.111     wiz       817: /grub/menu.lst and loads a kernel from the domU file system.
1.59      gdt       818: 
1.70      gdt       819: [Panix](http://www.panix.com/) lets users use pvgrub.  Panix reports
1.71      gdt       820: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
                    821: (and hence with defaults from "newfs -O 2").  See [Panix's pvgrub
1.70      gdt       822: page](http://www.panix.com/v-colo/grub.html), which describes only
1.74      gdt       823: Linux but should be updated to cover NetBSD :-).
1.70      gdt       824: 
                    825: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
                    826: their own kernel.  See then [prgmr.com NetBSD
1.74      gdt       827: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
                    828: (which is in need of updating).
1.59      gdt       829: 
1.70      gdt       830: It appears that [grub's FFS
                    831: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
                    832: does not support all aspects of modern FFS, but there are also reports
1.72      gdt       833: that FFSv2 works fine.  At prgmr, typically one has an ext2 or FAT
1.70      gdt       834: partition for the kernel with the intent that grub can understand it,
                    835: which leads to /netbsd not being the actual kernel.  One must remember
1.111     wiz       836: to update the special boot partition.
1.59      gdt       837: 
1.181     gdt       838: ### pygrub
1.168     wiki      839: 
                    840: pygrub runs in the dom0 and looks into the domU file system.  This
                    841: implies that the domU must have a kernel in a file system in a format
                    842: known to pygrub.
                    843: 
                    844: pygrub doesn't seem to work to load Linux images under NetBSD dom0,
                    845: and is inherently less secure than pvgrub due to running inside dom0. For both these
                    846: reasons, pygrub should not be used, and is only still present so that
                    847: historical DomU images using it still work.
                    848: 
                    849: As of 2014, pygrub seems to be of mostly historical
                    850: interest. New DomUs should use pvgrub.
                    851: 
1.181     gdt       852: ## Specific Providers
                    853: 
                    854: ### Amazon
1.59      gdt       855: 
1.143     wiki      856: See the [Amazon EC2 page](/amazon_ec2/).

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