Diff for /wikisrc/ports/xen/howto.mdwn between versions 1.157 and 1.195

version 1.157, 2018/08/27 16:54:42 version 1.195, 2021/03/03 15:03:13
Line 8  systems which operate in an unprivileged Line 8  systems which operate in an unprivileged
 from the domU systems are forwarded by the Xen hypervisor to the dom0 to be  from the domU systems are forwarded by the Xen hypervisor to the dom0 to be
 fulfilled.  fulfilled.
   
 Xen supports different styles of guest:  This HOWTO presumes a basic familiarity with the Xen system
   architecture, with installing NetBSD on amd64 hardware, and with
   installing software from pkgsrc.  See also the [Xen
   website](http://www.xenproject.org/).
   
   [[!toc]]
   
   # Overview
   
   The basic concept of Xen is that the hypervisor (xenkernel) runs on
   the hardware, and runs a privileged domain ("dom0") that can access
   disks/networking/etc.  One then runs additonal unprivileged domains
   (each a "domU"), presumably to do something useful.
   
   This HOWTO addresses how to run a NetBSD dom0 (and hence also build
   xen itself).  It also addresses how to run domUs in that environment,
   and how to deal with having a domU in a Xen environment run by someone
   else and/or not running NetBSD.
   
   There are many choices one can make; the HOWTO recommends the standard
   approach and limits discussion of alternatives in many cases.
   
   ## Guest Styles
   
   Xen supports different styles of guests.
   
 [[!table data="""  [[!table data="""
 Style of guest  |Supported by NetBSD  Style of guest  |Supported by NetBSD
 PV              |Yes (dom0, domU)  PV              |Yes (dom0, domU)
 HVM             |Yes (domU)  HVM             |Yes (domU)
 PVHVM           |No  PVHVM           |current-only (domU)
 PVH             |No  PVH             |current-only (domU, dom0 not yet)
 """]]  """]]
   
 In Para-Virtualized (PV) mode, the guest OS does not attempt to access  In Para-Virtualized (PV) mode, the guest OS does not attempt to access
 hardware directly, but instead makes hypercalls to the hypervisor; PV  hardware directly, but instead makes hypercalls to the hypervisor; PV
 guests must be specifically coded for Xen. In HVM mode, no guest  guests must be specifically coded for Xen.
 modification is required; however, hardware support is required, such  See [PV](https://wiki.xen.org/wiki/Paravirtualization_(PV\)).
 as VT-x on Intel CPUs and SVM on AMD CPUs.  
   In HVM mode, no guest modification is required; however, hardware
   support is required, such as VT-x on Intel CPUs and SVM on AMD CPUs.
   The dom0 runs qemu to emulate hardware.
   
   In PVHVM mode, the guest runs as HVM, but additionally can use PV
   drivers for efficiency.
   See [PV on HVM](https://wiki.xen.org/wiki/PV_on_HVM).
   
   There have been two PVH modes: original PVH and PVHv2.  Original PVH
   was based on PV mode and is no longer relevant at all.  PVHv2 is
   basically lightweight HVM with PV drivers.  A critical feature of it
   is that qemu is not needed; the hypervisor can do the emulation that
   is required.  Thus, a dom0 can be PVHv2.
   The source code uses PVH and config files use pvh; this refers to PVHv2.
   See [PVH(v2)](https://wiki.xenproject.org/wiki/PVH_(v2\)_Domu).
   
 At boot, the dom0 kernel is loaded as a module with Xen as the kernel.  At system boot, the dom0 kernel is loaded as a module with Xen as the kernel.
 The dom0 can start one or more domUs.  (Booting is explained in detail  The dom0 can start one or more domUs.  (Booting is explained in detail
 in the dom0 section.)  in the dom0 section.)
   
 This HOWTO presumes a basic familiarity with the Xen system  ## CPU Architecture
 architecture, with installing NetBSD on i386/amd64 hardware, and with  
 installing software from pkgsrc.  See also the [Xen  Xen runs on x86_64 hardware (the NetBSD amd64 port).
 website](http://www.xenproject.org/).  
   
 This HOWTO attempts to address both the case of running a NetBSD dom0  There is a concept of Xen running on ARM, but there are no reports of this working with NetBSD.
 on hardware and running domUs under it (NetBSD and other), and also  
 running NetBSD as a domU in a VPS.  
   
 Versions and Support  The dom0 system should be amd64.  (Instructions for i386PAE dom0 have been removed from the HOWTO.)
 ====================  
   The domU can be i386PAE or amd64.
   i386PAE at one point was considered as [faster](https://lists.xen.org/archives/html/xen-devel/2012-07/msg00085.html) than amd64.
   
   ## Xen Versions
   
 In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages  In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
 xenkernel and xentools.  We will refer only to the kernel versions,  xenkernel and xentools.  We will refer only to the kernel versions,
Line 49  Versions available in pkgsrc: Line 89  Versions available in pkgsrc:
   
 [[!table data="""  [[!table data="""
 Xen Version     |Package Name   |Xen CPU Support        |EOL'ed By Upstream  Xen Version     |Package Name   |Xen CPU Support        |EOL'ed By Upstream
 4.2             |xenkernel42    |32bit, 64bit           |Yes  4.11            |xenkernel411   |x86_64                 |No
 4.5             |xenkernel45    |64bit                  |Yes  4.13            |xenkernel413   |x86_64                 |No
 4.6             |xenkernel46    |64bit                  |Partially  
 4.8             |xenkernel48    |64bit                  |No  
 4.11            |xenkernel411   |64bit                  |No  
 """]]  """]]
   
 See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).  See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
   
 Multiprocessor (SMP) support in NetBSD differs depending on the domain:  Older Xen had a python-based management tool called xm, now replaced
   by xl.
 [[!table data="""  
 Domain          |Supports SMP  
 dom0            |No  
 domU            |Yes  
 """]]  
   
 Note: NetBSD support is called XEN3. However, it does support Xen 4,  
 because the hypercall interface has remained identical.  
   
 Architecture  ## NetBSD versions
 ------------  
   
 Xen itself runs on x86_64 hardware.  Xen has been supported in NetBSD for a long time, at least since 2005.
   Initially Xen was PV only.
   
 The dom0 system, plus each domU, can be either i386PAE or amd64.  NetBSD 8 and up support PV and HVM modes.
 i386 without PAE is not supported.  
   
 The standard approach is to use NetBSD/amd64 for the dom0.  Support for PVHVM and PVH is available only in NetBSD-current.
   
 To use an i386PAE dom0, one must build or obtain a 64bit Xen kernel and  NetBSD up to and including NetBSD 9 as a dom0 does not run SMP,
 install it on the system.  because some drivers are not yet safe for this.  NetBSD-current
   supports SMP in dom0.
   
 For domUs, i386PAE is considered as  NetBSD, when run as a domU, can and does typically run SMP.
 [faster](https://lists.xen.org/archives/html/xen-devel/2012-07/msg00085.html)  
 than amd64.  
   
 NetBSD as a dom0  Note that while Xen 4.13 is current, the kernel support is still
 ================  called XEN3, because the hypercall interface has not changed
   significantly.
   
 NetBSD can be used as a dom0 and works very well.  The following  # Creating a NetBSD dom0
 sections address installation, updating NetBSD, and updating Xen.  
   
 Note that it doesn't make sense to talk about installing a dom0 OS  In order to install a NetBSD as a dom0, one first installs a normal
 without also installing Xen itself.  We first address installing  NetBSD system, and then pivot the install to a dom0 install by
 NetBSD, which is not yet a dom0, and then adding Xen, pivoting the  changing the kernel and boot configuration.
 NetBSD install to a dom0 install by just changing the kernel and boot  
 configuration.  
   
 In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with  In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
 512M it worked reliably.  This does not make sense, but if you see  512M it worked reliably.  This does not make sense, but if you see
 "not ELF" after Xen boots, try increasing dom0 RAM.  "not ELF" after Xen boots, try increasing dom0 RAM.
   
 Styles of dom0 operation  ## Installation of NetBSD
 ------------------------  
   
 There are two basic ways to use Xen.  The traditional method is for  [Install NetBSD/amd64](/guide/inst/) just as you would if you were not
 the dom0 to do absolutely nothing other than providing support to some  using Xen.  Therefore, use the most recent release, or a build from
 number of domUs.  Such a system was probably installed for the sole  the most recent stable branch.  Alternatively, use -current, being
 purpose of hosting domUs, and sits in a server room on a UPS.  mindful of all the usual caveats of lower stability of current, and
   likely a bit more so.
   
 The other way is to put Xen under a normal-usage computer, so that the  ## Installation of Xen
 dom0 is what the computer would have been without Xen, perhaps a  
 desktop or laptop.  Then, one can run domUs at will.  
   
 Installation of NetBSD  ### Building Xen
 ----------------------  
   
 [Install NetBSD/amd64](/guide/inst/)  Use the most recent version of Xen in pkgsrc, unless the DESCR says that it is not suitable.
 just as you would if you were not using Xen.  Therefore, choose 4.13.
   In the dom0, install xenkernel413 and xentools413 from pkgsrc.
   
 Installation of Xen  Once this is done, copy the Xen kernel from where pkgsrc puts it to
 -------------------  where the boot process will be able to find it:
   
 We will consider that you chose to use Xen 4.8, with NetBSD/amd64 as  [[!template id=programlisting text="""
 dom0. In the dom0, install xenkernel48 and xentools48 from pkgsrc.  # cp -p /usr/pkg/xen413-kernel/xen.gz /
 Ensure that your packages are recent.  """]]
   
 Once this is done, install the Xen kernel itself:  Then, place a NetBSD XEN3_DOM0 kernel in the `/` directory. Such
   kernel can either be taken from a local release build.sh run, compiled
   manually, or downloaded from the NetBSD FTP, for example at:
   
 [[!template id=programlisting text="""  [[!template id=programlisting text="""
 # cp /usr/pkg/xen48-kernel/xen.gz /  ftp.netbsd.org/pub/NetBSD/NetBSD-9.1/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
 """]]  """]]
   
 Then, place a NetBSD XEN3_DOM0 kernel in the `/` directory, copied from  ### Configuring booting
 `releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz` of a NetBSD build.  
   
 Add a line to /boot.cfg to boot Xen:  Read boot.cfg(8) carefully.  Add lines to /boot.cfg to boot Xen:
   
 [[!template id=filecontent name="/boot.cfg" text="""  [[!template id=filecontent name="/boot.cfg" text="""
 menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M  menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
   menu=Xen single user:load /netbsd-XEN3_DOM0.gz console=pc -s;multiboot /xen.gz dom0_mem=512M
 """]]  """]]
   
 which specifies that the dom0 should have 512MB of ram, leaving the rest  This specifies that the dom0 should have 512MB of ram, leaving the rest
 to be allocated for domUs.  To use a serial console, use  to be allocated for domUs.
   
   NB: This says add, not replace, so that you will be able to more
   easily boot a NetBSD kernel without Xen.  Once Xen boots ok, you may
   want to set it as default.  It is highly likely that you will have
   trouble at some point, and keeping an up-to-date GENERIC for use in
   fixing problems is the standard prudent approach.
   
   \todo Explain why rndseed is not set with Xen as part of the dom0
   subconfiguration.
   
   Beware that userconf statements must be attached to the dom0 load, and
   may not be at top-level, because then they would try to configure the
   hypervisor, if there is a way to pass them via multiboot .  It appears
   that adding `userconf=pckbc` to `/boot.cfg` causes Xen to crash very
   early with a heap overflow.
   
   ### Console selection
   
   See boot_console(8).  Understand that you should start from a place of
   having console setup correct for booting GENERIC before trying to
   configure Xen.
   
   Generally for GENERIC, one sets the console in bootxx_ffsv1 or
   equivalent, and this is passed on to /boot (where one typically does
   not set the console).  This configuration of bootxx_ffsv1 should also
   be in place for Xen systems, to allow seeing messages from /boot and
   use of a keyboard to select a line from the menu.  And, one should
   have a working boot path to GENERIC for rescue situations.
   
   With GENERIC, the boot options are passed on to /netbsd, but there is
   currently no mechanism to pass these via multiboot to the hypervisor.
   Thus, in addition to configuring the console in the boot blocks, one
   must also configure it for Xen.
   
   By default, the hypervisor (Xen itself) will use some sort of vga
   device as the console, much like GENERIC uses by default.  The vga
   console is relinquished at the conclusion of hypervisor boot, before
   the dom0 is started.  Xen when using a vga console does not process
   console input.
   
   The hypervisor can be configured to use a serial port console, e.g.
 [[!template id=filecontent name="/boot.cfg" text="""  [[!template id=filecontent name="/boot.cfg" text="""
 menu=Xen:load /netbsd-XEN3_DOM0.gz;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1  menu=Xen:losad /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
 """]]  """]]
   This example uses the first serial port (Xen counts from 1; this is
   what NetBSD would call com0), and sets speed and parity.  (The dom0 is
   then configured to use the same serial port in this example.)
   
   With the hypervisor configured for a serial console, it can get input,
   and there is a notion of passing this input to the dom0.  \todo
   Explain why, if Xen has a serial console, the dom0 console is
   typically also configured to open that same serial port, instead of
   getting the passthrough input via the xen console.
   
   One also configures the console for the dom0.  While one might expect
   console=pc to be default, following behavior of GENERIC, a hasty read
   of the code suggests there is no default and booting without a
   selected console might lead to a panic.  Also, there is merit in
   explicit configuration.  Therefore the standard approach is to place
   console=pc as part of the load statement for the dom0 kernel, or
   alternatively console=com0.
   
   The NetBSD dom0 kernel will attach xencons(4) (the man page does not
   exist), but this is not used as a console.  It is used to obtain the
   messages from the hypervisor's console; run `xl dmesg` to see them.
   
 which will use the first serial port for Xen (which counts starting  ### Tuning
 from 1, unlike NetBSD which counts starting from 0), forcing  
 speed/parity.  Because the NetBSD command line lacks a  
 "console=pc" argument, it will use the default "xencons" console device,  
 which directs the console I/O through Xen to the same console device Xen  
 itself uses (in this case, the serial port).  
   
 In an attempt to add performance, one can also add `dom0_max_vcpus=1 dom0_vcpus_pin`,  In an attempt to add performance, one can also add `dom0_max_vcpus=1 dom0_vcpus_pin`,
 to force only one vcpu to be provided (since NetBSD dom0 can't use  to force only one vcpu to be provided (since NetBSD dom0 can't use
 more) and to pin that vcpu to a physical CPU.  more) and to pin that vcpu to a physical CPU. Xen has
   [many boot options](http://xenbits.xenproject.org/docs/4.13-testing/misc/xen-command-line.html),
 Xen has [many boot  
 options](http://xenbits.xenproject.org/docs/4.5-testing/misc/xen-command-line.html),  
 and other than dom0 memory and max_vcpus, they are generally not  and other than dom0 memory and max_vcpus, they are generally not
 necessary.  necessary.
   \todo Revisit this advice with current.
   \todo Explain if anyone has ever actually measured that this helps.
   
 Now, reboot so that you are running a DOM0 kernel under Xen, rather  ### rc.conf
 than GENERIC without Xen.  
   
 Configuring Xen  
 ---------------  
   
 Now, you have a system that will boot Xen and the dom0 kernel, but not  
 do anything else special.  Make sure that you have rebooted into Xen.  
 There will be no domUs, and none can be started because you still have  
 to configure the dom0 daemons.  
   
 Since Xen 4.2, the tool which should be used is `xl`.  Ensure that the boot scripts installed in
   `/usr/pkg/share/examples/rc.d` are in `/etc/rc.d`, either because you
   have `PKG_RCD_SCRIPTS=yes`, or manually.  (This is not special to Xen,
   but a normal part of pkgsrc usage.)
   
 For 4.1 and up, you should enable `xencommons`:  Set `xencommons=YES` in rc.conf:
   
 [[!template id=filecontent name="/etc/rc.conf" text="""  [[!template id=filecontent name="/etc/rc.conf" text="""
 xencommons=YES  xencommons=YES
 """]]  """]]
   
 Not enabling xencommons will result in a hang; it is necessary to hit `^C` on  \todo Recommend for/against xen-watchdog.
 the console to let the machine finish booting:  
   ### Testing
   
 TODO: Recommend for/against xen-watchdog.  Now, reboot so that you are running a DOM0 kernel under Xen, rather
   than GENERIC without Xen.
   
 After you have configured the daemons and either started them (in the  Once the reboot is done, use `xl` to inspect Xen's boot messages,
 order given) or rebooted, use `xl` to inspect Xen's boot messages,  
 available resources, and running domains.  For example:  available resources, and running domains.  For example:
   
 [[!template id=programlisting text="""  [[!template id=programlisting text="""
Line 216  the state when the new xenstored starts. Line 292  the state when the new xenstored starts.
 make this work, one should not expect to be able to restart xenstored  make this work, one should not expect to be able to restart xenstored
 (and thus xencommons).  There is currently no reason to expect that  (and thus xencommons).  There is currently no reason to expect that
 this will get fixed any time soon.  this will get fixed any time soon.
   \todo Confirm if this is still true in 2020.
   
 anita (for testing NetBSD)  ## Xen-specific NetBSD issues
 --------------------------  
   
 With the setup so far (assuming 4.8/xl), one should be able to run  
 anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as  
 root, because anita must create a domU):  
   
 [[!template id=programlisting text="""  
 anita --vmm=xl test file:///usr/obj/i386/  
 """]]  
   
 Xen-specific NetBSD issues  
 --------------------------  
   
 There are (at least) two additional things different about NetBSD as a  There are (at least) two additional things different about NetBSD as a
 dom0 kernel compared to hardware.  dom0 kernel compared to hardware.
   
 One is that the module ABI is different because some of the #defines  One is that through NetBSD 9 the module ABI is different because some
 change, so one must build modules for Xen.  As of netbsd-7, the build  of the #defines change, so there are separate sets of modules in
 system does this automatically.  /stand.  In NetBSD-current, there is only one set of modules.
   
 The other difference is that XEN3_DOM0 does not have exactly the same  The other difference is that XEN3_DOM0 does not have exactly the same
 options as GENERIC.  While it is debatable whether or not this is a  options as GENERIC.  While it is debatable whether or not this is a
 bug, users should be aware of this and can simply add missing config  bug, users should be aware of this and can simply add missing config
 items if desired.  items if desired.
   
 Updating NetBSD in a dom0  Finally, there have been occasional reports of trouble with X11
 -------------------------  servers in NetBSD as a dom0.
   
   ## Updating Xen in a dom0
   
   Basically, update the xenkernel and xentools packages and copy the new
   Xen kernel into place, and reboot.  This procedure should be usable to
   update to a new Xen release, but the reader is reminded that having a
   non-Xen boot methods was recommended earlier.
   
   ## Updating NetBSD in a dom0
   
 This is just like updating NetBSD on bare hardware, assuming the new  This is just like updating NetBSD on bare hardware, assuming the new
 version supports the version of Xen you are running.  Generally, one  version supports the version of Xen you are running.  Generally, one
 replaces the kernel and reboots, and then overlays userland binaries  replaces the kernel and reboots, and then overlays userland binaries
 and adjusts `/etc`.  and adjusts `/etc`.
   
 Note that one must update both the non-Xen kernel typically used for  Note that one should update both the non-Xen kernel typically used for
 rescue purposes and the DOM0 kernel used with Xen.  rescue purposes, as well as the DOM0 kernel used with Xen.
   
 Converting from grub to /boot  ## anita (for testing NetBSD)
 -----------------------------  
   
 These instructions were used to convert a system from  With a NetBSD dom0, even without any domUs, one should be able to run
 grub to /boot.  The system was originally installed in February of  anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated  root, because anita must create a domU):
 over time.  Before these commands, it was running NetBSD 6 i386, Xen  
 4.1 and grub, much like the message linked earlier in the grub  
 section.  
   
 [[!template id=programlisting text="""  [[!template id=programlisting text="""
 # Install MBR bootblocks on both disks.  anita --vmm=xl test file:///usr/obj/i386/
 fdisk -i /dev/rwd0d  """]]
 fdisk -i /dev/rwd1d  
 # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.  
 installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1  
 installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1  
 # Install secondary boot loader  
 cp -p /usr/mdec/boot /  
 # Create boot.cfg following earlier guidance:  
 menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M  
 menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=512M  
 menu=GENERIC:boot  
 menu=GENERIC single-user:boot -s  
 menu=GENERIC.ok:boot netbsd.ok  
 menu=GENERIC.ok single-user:boot netbsd.ok -s  
 menu=Drop to boot prompt:prompt  
 default=1  
 timeout=30  
 """]]  
   
 Upgrading Xen versions  
 ---------------------  
   
 Minor version upgrades are trivial.  Just rebuild/replace the  
 xenkernel version and copy the new xen.gz to `/` (where `/boot.cfg`  
 references it), and reboot.  
   
 Unprivileged domains (domU)  # Unprivileged domains (domU)
 ===========================  
   
 This section describes general concepts about domUs.  It does not  This section describes general concepts about domUs.  It does not
 address specific domU operating systems or how to install them.  The  address specific domU operating systems or how to install them.  The
Line 308  mediated by Xen, and configured in the d Line 353  mediated by Xen, and configured in the d
 Entropy in domUs can be an issue; physical disks and network are on  Entropy in domUs can be an issue; physical disks and network are on
 the dom0.  NetBSD's /dev/random system works, but is often challenged.  the dom0.  NetBSD's /dev/random system works, but is often challenged.
   
 Config files  ## Config files
 ------------  
   
 There is no good order to present config files and the concepts  See /usr/pkg/share/examples/xen/xlexample*
 surrounding what is being configured.  We first show an example config  for a small number of well-commented examples, mostly for running
 file, and then in the various sections give details.  
   
 See /usr/pkg/share/examples/xen/xmexample*,  
 for a large number of well-commented examples, mostly for running  
 GNU/Linux.  GNU/Linux.
   
 The following is an example minimal domain configuration file. The domU  The following is an example minimal domain configuration file. The domU
Line 340  are stored in files and Xen attaches the Line 380  are stored in files and Xen attaches the
 dom0 on domain creation.  The system treats xbd0 as the boot device  dom0 on domain creation.  The system treats xbd0 as the boot device
 without needing explicit configuration.  without needing explicit configuration.
   
 By default, `xl` looks for domain config files in `/usr/pkg/etc/xen`.  Note  There is not type line; that implicitly defines a pv domU.
   
   By convention, domain config files are kept in `/usr/pkg/etc/xen`.  Note
 that "xl create" takes the name of a config file, while other commands  that "xl create" takes the name of a config file, while other commands
 take the name of a domain.  take the name of a domain.
   
 Examples of commands:  Examples of commands:
   
 [[!template id=programlisting text="""  [[!template id=programlisting text="""
 xl create foo  xl create /usr/pkg/etc/xen/foo
 xl console foo  xl console domU-id
 xl create -c foo  xl create -c /usr/pkg/etc/xen/foo
 xl shutdown foo  xl shutdown domU-id
 xl list  xl list
 """]]  """]]
   
Line 359  equivalent to pushing the power button;  Line 401  equivalent to pushing the power button; 
 power-press event and do a clean shutdown.  Shutting down the dom0  power-press event and do a clean shutdown.  Shutting down the dom0
 will trigger controlled shutdowns of all configured domUs.  will trigger controlled shutdowns of all configured domUs.
   
 domU kernels  ## CPU and memory
 ------------  
   
 On a physical computer, the BIOS reads sector 0, and a chain of boot  
 loaders finds and loads a kernel.  Normally this comes from the root  
 file system.  With Xen domUs, the process is totally different.  The  
 normal path is for the domU kernel to be a file in the dom0's  
 file system.  At the request of the dom0, Xen loads that kernel into a  
 new domU instance and starts execution. domU kernels can be anyplace.  
   
 Note that loading the domU kernel from the dom0 implies that boot  
 blocks, /boot, /boot.cfg, and so on are all ignored in the domU.  
 See the VPS section near the end for discussion of alternate ways to  
 obtain domU kernels.  
   
 CPU and memory  A domain is provided with some number of vcpus, up to the number
 --------------  
   
 A domain is provided with some number of vcpus, less than the number  
 of CPUs seen by the hypervisor. For a domU, it is controlled  of CPUs seen by the hypervisor. For a domU, it is controlled
 from the config file by the "vcpus = N" directive.  from the config file by the "vcpus = N" directive.
   
Line 389  than the available memory. Line 415  than the available memory.
 Xen also provides a "balloon" driver, which can be used to let domains  Xen also provides a "balloon" driver, which can be used to let domains
 use more memory temporarily.  use more memory temporarily.
   
 Virtual disks  ## Virtual disks
 -------------  
   In domU config files, the disks are defined as a sequence of 3-tuples:
   
    * The first element is "method:/path/to/disk". Common methods are
      "file:" for a file-backed vnd, and "phy:" for something that is already
      a device, such as an LVM logical volume.
   
    * The second element is an artifact of how virtual disks are passed to
      Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs
      are given a device name to associate with the disk, and values like
      "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears
      as xbd0, the second as xbd1, and so on.  However, xl demands a
      second argument.  The name given is converted to a major/minor by
      calling stat(2) on the name in /dev and this is passed to the domU.
      In the general case, the dom0 and domU can be different operating
      systems, and it is an unwarranted assumption that they have consistent
      numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD
      as both dom0 and domU, using values of 0x0 for the first disk and 0x1
      for the second works fine and avoids this issue.  For a GNU/Linux
      guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
      /dev/hda1.
   
 With the file/vnd style, typically one creates a directory,   * The third element is "w" for writable disks, and "r" for read-only
 e.g. /u0/xen, on a disk large enough to hold virtual disks for all     disks.
 domUs.  Then, for each domU disk, one writes zeros to a file that then  
 serves to hold the virtual disk's bits; a suggested name is foo-xbd0  
 for the first virtual disk for the domU called foo.  Writing zeros to  
 the file serves two purposes.  One is that preallocating the contents  
 improves performance.  The other is that vnd on sparse files has  
 failed to work.  TODO: give working/notworking NetBSD versions for  
 sparse vnd and gnats reference.  Note that the use of file/vnd for Xen  
 is not really different than creating a file-backed virtual disk for  
 some other purpose, except that xentools handles the vnconfig  
 commands.  To create an empty 4G virtual disk, simply do  
   
         dd if=/dev/zero of=foo-xbd0 bs=1m count=4096  
   
 Do not use qemu-img-xen, because this will create sparse file.  There  
 have been recent (2015) reports of sparse vnd(4) devices causing  
 lockups, but there is apparently no PR.  
   
 With the lvm style, one creates logical devices.  They are then used  
 similarly to vnds.  TODO: Add an example with lvm.  
   
 In domU config files, the disks are defined as a sequence of 3-tuples.  
 The first element is "method:/path/to/disk".  Common methods are  
 "file:" for file-backed vnd. and "phy:" for something that is already  
 a (TODO: character or block) device.  
   
 The second element is an artifact of how virtual disks are passed to  
 Linux, and a source of confusion with NetBSD Xen usage.  Linux domUs  
 are given a device name to associate with the disk, and values like  
 "hda1" or "sda1" are common.  In a NetBSD domU, the first disk appears  
 as xbd0, the second as xbd1, and so on.  However, xm/xl demand a  
 second argument.  The name given is converted to a major/minor by  
 calling stat(2) on the name in /dev and this is passed to the domU.  
 In the general case, the dom0 and domU can be different operating  
 systems, and it is an unwarranted assumption that they have consistent  
 numbering in /dev, or even that the dom0 OS has a /dev.  With NetBSD  
 as both dom0 and domU, using values of 0x0 for the first disk and 0x1  
 for the second works fine and avoids this issue.  For a GNU/Linux  
 guest, one can create /dev/hda1 in /dev, or to pass 0x301 for  
 /dev/hda1.  
   
 The third element is "w" for writable disks, and "r" for read-only  Example:
 disks.  [[!template id=filecontent name="/usr/pkg/etc/xen/foo" text="""
   disk = [ 'file:/n0/xen/foo-wd0,0x0,w' ]
   """]]
   
 Note that NetBSD by default creates only vnd[0123].  If you need more  Note that NetBSD by default creates only vnd[0123].  If you need more
 than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the  than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
 dom0.  dom0.
   
 Note that NetBSD by default creates only xbd[0123].  If you need more  ## Virtual Networking
 virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.  
   
 Virtual Networking  
 ------------------  
   
 Xen provides virtual Ethernets, each of which connects the dom0 and a  Xen provides virtual Ethernets, each of which connects the dom0 and a
 domU.  For each virtual network, there is an interface "xvifN.M" in  domU.  For each virtual network, there is an interface "xvifN.M" in
 the dom0, and in domU index N, a matching interface xennetM (NetBSD  the dom0, and a matching interface xennetM (NetBSD name) in domU index N.
 name).  The interfaces behave as if there is an Ethernet with two  The interfaces behave as if there is an Ethernet with two
 adapters connected.  From this primitive, one can construct various  adapters connected.  From this primitive, one can construct various
 configurations.  We focus on two common and useful cases for which  configurations.  We focus on two common and useful cases for which
 there are existing scripts: bridging and NAT.  there are existing scripts: bridging and NAT.
Line 477  The MAC address specified is the one use Line 482  The MAC address specified is the one use
 domain.  The interface in dom0 will use this address XOR'd with  domain.  The interface in dom0 will use this address XOR'd with
 00:00:00:01:00:00.  Random MAC addresses are assigned if not given.  00:00:00:01:00:00.  Random MAC addresses are assigned if not given.
   
 Sizing domains  ## Starting domains automatically
 --------------  
   
 Modern x86 hardware has vast amounts of resources.  However, many  
 virtual servers can function just fine on far less.  A system with  
 512M of RAM and a 4G disk can be a reasonable choice.  Note that it is  
 far easier to adjust virtual resources than physical ones.  For  
 memory, it's just a config file edit and a reboot.  For disk, one can  
 create a new file and vnconfig it (or lvm), and then dump/restore,  
 just like updating physical disks, but without having to be there and  
 without those pesky connectors.  
   
 Starting domains automatically  
 ------------------------------  
   
 To start domains foo at bar at boot and shut them down cleanly on dom0  
 shutdown, in rc.conf add:  
   
         xendomains="foo bar"  To start domains `domU-netbsd` and `domU-linux` at boot and shut them
   down cleanly on dom0 shutdown, add the following in rc.conf:
   
 Note that earlier versions of the xentools41 xendomains rc.d script  [[!template id=filecontent name="/etc/rc.conf" text="""
 used xl, when one should use xm with 4.1.  xendomains="domU-netbsd domU-linux"
   """]]
   
 Creating specific unprivileged domains (domU)  # domU setup for specific systems
 =============================================  
   
 Creating domUs is almost entirely independent of operating system.  We  Creating domUs is almost entirely independent of operating system.  We
 have already presented the basics of config files.  Note that you must  have already presented the basics of config files in the previous system.
 have already completed the dom0 setup so that "xl list" (or "xm list")  
 works.  Of course, this section presumes that you have a working dom0.
   
 Creating an unprivileged NetBSD domain (domU)  ## Creating a NetBSD PV domU
 ---------------------------------------------  
   
 See the earlier config file, and adjust memory.  Decide on how much  See the earlier config file, and adjust memory.  Decide on how much
 storage you will provide, and prepare it (file or lvm).  storage you will provide, and prepare it (file or LVM).
   
 While the kernel will be obtained from the dom0 file system, the same  While the kernel will be obtained from the dom0 file system, the same
 file should be present in the domU as /netbsd so that tools like  file should be present in the domU as /netbsd so that tools like
 savecore(8) can work.   (This is helpful but not necessary.)  savecore(8) can work.   (This is helpful but not necessary.)
   
 The kernel must be specifically for Xen and for use as a domU.  The  The kernel must be specifically built for Xen, to use PV interfacesas
 i386 and amd64 provide the following kernels:  a domU.  NetBSD release builds provide the following kernels:
   
         i386 XEN3PAE_DOMU          i386 XEN3PAE_DOMU
         amd64 XEN3_DOMU          amd64 XEN3_DOMU
   
 This will boot NetBSD, but this is not that useful if the disk is  This will boot NetBSD, but this is not that useful if the disk is
 empty.  One approach is to unpack sets onto the disk outside of xen  empty.  One approach is to unpack sets onto the disk outside of Xen
 (by mounting it, just as you would prepare a physical disk for a  (by mounting it, just as you would prepare a physical disk for a
 system you can't run the installer on).  system you can't run the installer on).
   
Line 535  kernel to / and change the kernel line i Line 524  kernel to / and change the kernel line i
   
         kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"          kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
   
 Then, start the domain as "xl create -c configname".  Then, start the domain as "xl create -c configfile".
   
 Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following  Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
 line should be used in the config file.  line should be used in the config file.
Line 545  line should be used in the config file. Line 534  line should be used in the config file.
 After booting the domain, the option to install via CDROM may be  After booting the domain, the option to install via CDROM may be
 selected.  The CDROM device should be changed to `xbd1d`.  selected.  The CDROM device should be changed to `xbd1d`.
   
 Once done installing, "halt -p" the new domain (don't reboot or halt,  Once done installing, "halt -p" the new domain (don't reboot or halt:
 it would reload the INSTALL_XEN3_DOMU kernel even if you changed the  it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
 config file), switch the config file back to the XEN3_DOMU kernel,  config file), switch the config file back to the XEN3_DOMU kernel,
 and start the new domain again. Now it should be able to use "root on  and start the new domain again. Now it should be able to use "root on
 xbd0a" and you should have a, functional NetBSD domU.  xbd0a" and you should have a functional NetBSD domU.
   
 TODO: check if this is still accurate.  TODO: check if this is still accurate.
 When the new domain is booting you'll see some warnings about *wscons*  When the new domain is booting you'll see some warnings about *wscons*
Line 565  and the pseudo-terminals. These can be f Line 554  and the pseudo-terminals. These can be f
   
 Finally, all screens must be commented out from `/etc/wscons.conf`.  Finally, all screens must be commented out from `/etc/wscons.conf`.
   
 It is also desirable to add  One should also run `powerd` in a domU, but this should not need
   configuring.  With powerd, the domain will run a controlled shutdown
         powerd=YES  if `xl shutdown -R` or `xl shutdown -H` is used on the dom0, via
   receiving a synthetic `power button pressed` signal.  In 9 and
 in rc.conf. This way, the domain will be properly shut down if  current, `powerd` is run by default under Xen kernels (or if ACPI is
 `xm shutdown -R` or `xm shutdown -H` is used on the dom0.  present), and it can be added to rc.conf if not.
   
 It is not strictly necessary to have a kernel (as /netbsd) in the domU  It is not strictly necessary to have a kernel (as /netbsd) in the domU
 file system.  However, various programs (e.g. netstat) will use that  file system.  However, various programs (e.g. netstat) will use that
Line 580  not really a Xen-specific issue, but bec Line 569  not really a Xen-specific issue, but bec
 obtained from the dom0, it is far more likely to be out of sync or  obtained from the dom0, it is far more likely to be out of sync or
 missing with Xen.)  missing with Xen.)
   
 Creating an unprivileged Linux domain (domU)  Note that NetBSD by default creates only xbd[0123].  If you need more
 --------------------------------------------  virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
   
   ## Creating a Linux domU
   
 Creating unprivileged Linux domains isn't much different from  Creating unprivileged Linux domains isn't much different from
 unprivileged NetBSD domains, but there are some details to know.  unprivileged NetBSD domains, but there are some details to know.
Line 624  To get the Linux console right, you need Line 615  To get the Linux console right, you need
 to your configuration since not all Linux distributions auto-attach a  to your configuration since not all Linux distributions auto-attach a
 tty to the xen console.  tty to the xen console.
   
 Creating an unprivileged Solaris domain (domU)  ## Creating a NetBSD HVM domU
 ----------------------------------------------  
   Use type='hmv', probably.  Use a GENERIC kernel within the disk image.
   
   ## Creating a NetBSD PVH domU
   
   Use type='pvh'.
   
   \todo Explain where the kernel comes from.
   
   ## Creating a Solaris domU
   
 See possibly outdated  See possibly outdated
 [Solaris domU instructions](/ports/xen/howto-solaris/).  [Solaris domU instructions](/ports/xen/howto-solaris/).
   
   ## PCI passthrough: Using PCI devices in guest domains
   
 PCI passthrough: Using PCI devices in guest domains  NB: PCI passthrough only works on some Xen versions and as of 2020 it
 ---------------------------------------------------  is not clear that it works on any version in pkgsrc.  Reports
   confirming or denying this notion should be sent to port-xen@.
   
 The dom0 can give other domains access to selected PCI  The dom0 can give other domains access to selected PCI
 devices. This can allow, for example, a non-privileged domain to have  devices. This can allow, for example, a non-privileged domain to have
Line 701  note that only the "xpci" lines are unus Line 703  note that only the "xpci" lines are unus
         cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives          cd*     at scsibus? target ? lun ?      # SCSI CD-ROM drives
   
   
 NetBSD as a domU in a VPS  # Miscellaneous Information
 =========================  
   ## Nesting under Linux KVM
   
   It is possible to run a Xen and a NetBSD dom0 under Linux KVM.  One
   can enable virtio in the dom0 for greater speed.
   
   ## Other nesting
   
   In theory, any full emulation should be able to run Xen and a NetBSD
   dom0.  The HOWTO does not currently have information about Xen XVM
   mode, nvmm, qemu, Virtualbox, etc.
   
   ## NetBSD 5 as domU
   
   [NetBSD 5 is known to panic.](http://mail-index.netbsd.org/port-xen/2018/04/17/msg009181.html)
   (However, NetBSD 5 systems should be updated to a supported version.)
   
   # NetBSD as a domU in a VPS
   
 The bulk of the HOWTO is about using NetBSD as a dom0 on your own  The bulk of the HOWTO is about using NetBSD as a dom0 on your own
 hardware.  This section explains how to deal with Xen in a domU as a  hardware.  This section explains how to deal with Xen in a domU as a
Line 729  A second issue is multiple CPUs.  With N Line 748  A second issue is multiple CPUs.  With N
 multiple vcpus, and it is typical for VPS providers to enable multiple  multiple vcpus, and it is typical for VPS providers to enable multiple
 CPUs for NetBSD domUs.  CPUs for NetBSD domUs.
   
 pygrub  ## Complexities due to Xen changes
 -------  
   
 pygrub runs in the dom0 and looks into the domU file system.  This  Xen has many security advisories and people running Xen systems make
 implies that the domU must have a kernel in a file system in a format  different choices.
 known to pygrub.  As of 2014, pygrub seems to be of mostly historical  
 interest.  ### stub domains
   
 pvgrub  Some (Linux only?) dom0 systems use something called "stub domains" to
 ------  isolate qemu from the dom0 system, as a security and reliabilty
   mechanism when running HVM domUs.  Somehow, NetBSD's GENERIC kernel
   ends up using PIO for disks rather than DMA.  Of course, all of this
   is emulated, but emulated PIO is unusably slow.  This problem is not
   currently understood.
   
   ### Grant tables
   
   There are multiple versions of using grant tables, and some security
   advisories have suggested disabling some versions.  Some versions of
   NetBSD apparently only use specific versions and this can lead to
   "NetBSD current doesn't run on hosting provider X" situations.
   
   \todo Explain better.
   
   ## Boot methods
   
   ### pvgrub
   
 pvgrub is a version of grub that uses PV operations instead of BIOS  pvgrub is a version of grub that uses PV operations instead of BIOS
 calls.  It is booted from the dom0 as the domU kernel, and then reads  calls.  It is booted from the dom0 as the domU kernel, and then reads
Line 763  partition for the kernel with the intent Line 798  partition for the kernel with the intent
 which leads to /netbsd not being the actual kernel.  One must remember  which leads to /netbsd not being the actual kernel.  One must remember
 to update the special boot partition.  to update the special boot partition.
   
 Amazon  ### pygrub
 ------  
   
 See the [Amazon EC2 page](/amazon_ec2/).  pygrub runs in the dom0 and looks into the domU file system.  This
   implies that the domU must have a kernel in a file system in a format
   known to pygrub.
   
   pygrub doesn't seem to work to load Linux images under NetBSD dom0,
   and is inherently less secure than pvgrub due to running inside dom0. For both these
   reasons, pygrub should not be used, and is only still present so that
   historical DomU images using it still work.
   
 Random pointers  As of 2014, pygrub seems to be of mostly historical
 ===============  interest. New DomUs should use pvgrub.
   
 This section contains links from elsewhere not yet integrated into the  ## Specific Providers
 HOWTO, and other guides.  
   
 * http://www.lumbercartel.ca/library/xen/  ### Amazon
 * http://pbraun.nethence.com/doc/sysutils/xen_netbsd_dom0.html  
 * https://gmplib.org/~tege/xen.html  See the [Amazon EC2 page](/amazon_ec2/).

Removed from v.1.157  
changed lines
  Added in v.1.195


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