1: [[!meta title="Xen HowTo"]]
2:
3: Introduction
4: ============
5:
6: [![[Xen
7: screenshot]](https://www.netbsd.org/gallery/in-Action/hubertf-xens.png)](https://www.netbsd.org/gallery/in-Action/hubertf-xen.png)
8:
9: Xen is a hypervisor (or virtual machine monitor) for x86 hardware
10: (i686-class or higher), which supports running multiple guest
11: operating systems on a single physical machine. Xen is a Type 1 or
12: bare-metal hypervisor; one uses the Xen kernel to control the CPU,
13: memory and console, a dom0 operating system which mediates access to
14: other hardware (e.g., disks, network, USB), and one or more domU
15: operating systems which operate in an unprivileged virtualized
16: environment. IO requests from the domU systems are forwarded by the
17: hypervisor (Xen) to the dom0 to be fulfilled.
18:
19: Xen supports two styles of guests. The original is Para-Virtualized
20: (PV) which means that the guest OS does not attempt to access hardware
21: directly, but instead makes hypercalls to the hypervisor. This is
22: analogous to a user-space program making system calls. (The dom0
23: operating system uses PV calls for some functions, such as updating
24: memory mapping page tables, but has direct hardware access for disk
25: and network.) PV guests must be specifically coded for Xen.
26:
27: The more recent style is HVM, which means that the guest does not have
28: code for Xen and need not be aware that it is running under Xen.
29: Attempts to access hardware registers are trapped and emulated. This
30: style is less efficient but can run unmodified guests.
31:
32: Generally any machine that runs NetBSD/amd64 will work with Xen and PV
33: guests. In theory i386 computers (without x86_64/amd64 support) can
34: be used for Xen <= 4.2, but we have no recent reports of this working
35: (this is a hint). For HVM guests, hardware support is needed, but it
36: is common on recent machines. For Intel CPUs, one needs the VT-x
37: extension, shown in "cpuctl identify 0" as VMX. For AMD CPUs, one
38: needs the AMD-V extensions, shown in "cpuctl identify 0" as SVM.
39: There are further features for IOMMU virtualization, Intel's VT-d and
40: AMD's AMD-Vi. TODO: Explain whether Xen on NetBSD makes use of these
41: features. TODO: Review by someone who really understands this.
42:
43: Note that a FreeBSD dom0 requires VT-x and VT-d (or equivalent); this
44: is because the FreeBSD dom0 does not run in PV mode.
45:
46: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
47: The dom0 can start one or more domUs. (Booting is explained in detail
48: in the dom0 section.)
49:
50: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
51: and that Xen kernels and tools are available in pkgsrc. This HOWTO
52: attempts to address both the case of running a NetBSD dom0 on hardware
53: and running domUs under it (NetBSD and other), and also running NetBSD
54: as a domU in a VPS.
55:
56: Xen 3.1 in pkgsrc used to support "PCI passthrough", which means that
57: specific PCI devices can be made available to a specific domU instead
58: of the dom0. This can be useful to let a domU run X11, or access some
59: network interface or other peripheral.
60:
61: NetBSD 6 and earlier supported Xen 2; support was removed from NetBSD
62: 7. Xen 2 has been removed from pkgsrc.
63:
64: Prerequisites
65: -------------
66:
67: Installing NetBSD/Xen is not extremely difficult, but it is more
68: complex than a normal installation of NetBSD.
69: In general, this HOWTO is occasionally overly restrictive about how
70: things must be done, guiding the reader to stay on the established
71: path when there are no known good reasons to stray.
72:
73: This HOWTO presumes a basic familiarity with the Xen system
74: architecture, with installing NetBSD on i386/amd64 hardware, and with
75: installing software from pkgsrc. See also the [Xen
76: website](http://www.xenproject.org/).
77:
78: Versions of Xen and NetBSD
79: ==========================
80:
81: Most of the installation concepts and instructions are independent
82: of Xen version and NetBSD version. This section gives advice on
83: which version to choose. Versions not in pkgsrc and older unsupported
84: versions of NetBSD are intentionally ignored.
85:
86: The term "amd64" is used to refer to both the NetBSD port and to the
87: hardware architecture on which it runs. (Such hardware is made by
88: both Intel and AMD, and in 2016 a normal PC has this CPU
89: architecture.)
90:
91: Xen versions
92: ------------
93:
94: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
95: xenkernel and xentools. We will refer only to the kernel versions,
96: but note that both packages must be installed together and must have
97: matching versions.
98:
99: Versions available in pkgsrc:
100:
101: [[!table data="""
102: Xen Version |Package Name |Xen CPU Support |EOL'ed By Upstream
103: 4.2 |xenkernel42 |32bit, 64bit |Yes
104: 4.5 |xenkernel45 |64bit |Yes
105: 4.6 |xenkernel46 |64bit |Partially
106: 4.8 |xenkernel48 |64bit |No
107: 4.11 |xenkernel411 |64bit |No
108: """]]
109:
110: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
111:
112: Note: Xen 4.2 was the last version to support 32bit CPUs.
113:
114: Xen command program
115: -------------------
116:
117: Early Xen used a program called xm to manipulate the system from the
118: dom0. Starting in 4.1, a replacement program with similar behavior
119: called xl is provided, but it does not work well in 4.1. In 4.2, both
120: xm and xl work fine. 4.4 is the last version that has xm.
121:
122: You must make a global choice to use xm or xl, because it affects not
123: only which command you use, but the command used by rc.d scripts
124: (specifically xendomains) and which daemons should be run. The
125: xentools packages provide xl for 4.2 and up.
126:
127: In 4.2, you can choose to use xm by simply changing the ctl_command
128: variable and setting xend=YES in rc.conf.
129:
130: With xl, virtual devices are configured in parallel, which can cause
131: problems if they are written assuming serial operation (e.g., updating
132: firewall rules without explicit locking). There is now locking for
133: the provided scripts, which works for normal casses (e.g, file-backed
134: xbd, where a vnd must be allocated). But, as of 201612, it has not
135: been adequately tested for a complex custom setup with a large number
136: of interfaces.
137:
138: NetBSD versions
139: ---------------
140:
141: The netbsd-7, netbsd-8, and -current branches are all reasonable
142: choices, with more or less the same considerations for non-Xen use.
143: NetBSD 8 is recommended as the stable version of the most recent
144: release for production use.
145:
146: For developing Xen, netbsd-current may be appropriate.
147:
148: As of NetBSD 6, a NetBSD domU will support multiple vcpus. There is
149: no SMP support for NetBSD as dom0. (The dom0 itself doesn't really
150: need SMP for dom0 functions; the lack of support is really a problem
151: when using a dom0 as a normal computer.)
152:
153: Note: NetBSD support is called XEN3. However, it does support Xen 4,
154: because the hypercall interface has remained identical.
155:
156: Architecture
157: ------------
158:
159: Xen itself can run on i386 (Xen < 4.2) or amd64 hardware (all Xen
160: versions). (Practically, almost any computer where one would want to
161: run Xen today supports amd64.)
162:
163: Xen, the dom0 system, and each domU system can be either i386 or
164: amd64. When building a xenkernel package, one obtains an i386 Xen
165: kernel on an i386 host, and an amd64 Xen kernel on an amd64 host. If
166: the Xen kernel is i386, then the dom0 kernel and all domU kernels must
167: be i386. With an amd64 Xen kernel, an amd64 dom0 kernel is known to
168: work, and an i386 dom0 kernel should in theory work. An amd64
169: Xen/dom0 is known to support both i386 and amd64 domUs.
170:
171: i386 dom0 and domU kernels must be PAE (except for an i386 Xen 3.1
172: kernel, where one can use non-PAE for dom0 and all domUs); PAE kernels
173: are included in the NetBSD default build. (Note that emacs (at least)
174: fails if run on i386 with PAE when built without, and vice versa,
175: presumably due to bugs in the undump code.)
176:
177: Because of the above, the standard approach is to use an amd64 Xen
178: kernel and NetBSD/amd64 for the dom0. For domUs, NetBSD/i386 (with
179: the PAE kernel) and NetBSD/amd64 are in widespread use, and there is
180: little to no Xen-specific reason to prefer one over the other.
181:
182: Note that to use an i386 dom0 with Xen 4.5 or higher, one must build
183: (or obtain from pre-built packages) an amd64 Xen kernel and install
184: that on the system. (One must also use a PAE i386 kernel, but this is
185: also required with an i386 Xen kernel.). Almost no one in the
186: NetBSD/Xen community does this, and the standard, well-tested,
187: approach is to use an amd64 dom0.
188:
189: A [posting on
190: xen-devel](https://lists.xen.org/archives/html/xen-devel/2012-07/msg00085.html)
191: explained that PV system call overhead was higher on amd64, and thus
192: there is some notion that i386 guests are faster. It goes on to
193: caution that the total situation is complex and not entirely
194: understood. On top of that caution, the post is about Linux, not
195: NetBSD. TODO: Include link to benchmarks, if someone posts them.
196:
197: Stability
198: ---------
199:
200: Mostly, NetBSD as a dom0 or domU is quite stable. However, just like every
201: other architecture, there are some open PRs indicating problems.
202:
203: Note also that there are issues with sparse vnd(4) instances, but
204: these are not about Xen -- they just are noticed with sparse vnd(4)
205: instances in support of virtual disks in a dom0.
206:
207: Recommendation
208: --------------
209:
210: Therefore, this HOWTO recommends running xenkernel46, xl, the NetBSD 7
211: stable branch, and therefore to use an amd64 kernel as the dom0.
212: Either the i386PAE or amd64 version of NetBSD may be used as domUs.
213:
214: A tentative replacement recommendation is xenkernel48, xl, and NetBSD
215: 8.
216:
217: Because bugs are fixed quite often, and because of Xen security
218: advisories, it is good to stay up to date with NetBSD (tracking a
219: stable branch), with the Xen kernel (tracking a Xen version via
220: pkgsrc), and with the Xen tools. Specifically, NetBSD (-7 and
221: -current) got an important fix affecting dom0/domU timesharing in
222: November, 2015, and xentools46 got a fix to enable Ubuntu guests to
223: boot in December, 2016.
224:
225: NetBSD as a dom0
226: ================
227:
228: NetBSD can be used as a dom0 and works very well. The following
229: sections address installation, updating NetBSD, and updating Xen.
230: Note that it doesn't make sense to talk about installing a dom0 OS
231: without also installing Xen itself. We first address installing
232: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
233: NetBSD install to a dom0 install by just changing the kernel and boot
234: configuration.
235:
236: For experimenting with Xen, a machine with as little as 1G of RAM and
237: 100G of disk can work. For running many domUs in productions, far
238: more will be needed; e.g. 4-8G and 1T of disk is reasonable for a
239: half-dozen domUs of 512M and 32G each. Basically, the RAM and disk
240: have to be bigger than the sum of the RAM/disk needs of the dom0 and
241: all the domUs.
242:
243: In 2018-05, trouble booting a dom0 was reported with 256M of RAM: with
244: 512M it worked reliably. This does not make sense, but if you see
245: "not ELF" after Xen boots, try increasing dom0 RAM.
246:
247: Styles of dom0 operation
248: ------------------------
249:
250: There are two basic ways to use Xen. The traditional method is for
251: the dom0 to do absolutely nothing other than providing support to some
252: number of domUs. Such a system was probably installed for the sole
253: purpose of hosting domUs, and sits in a server room on a UPS.
254:
255: The other way is to put Xen under a normal-usage computer, so that the
256: dom0 is what the computer would have been without Xen, perhaps a
257: desktop or laptop. Then, one can run domUs at will. Purists will
258: deride this as less secure than the previous approach, and for a
259: computer whose purpose is to run domUs, they are right. But Xen and a
260: dom0 (without domUs) is not meaningfully less secure than the same
261: things running without Xen. One can boot Xen or boot regular NetBSD
262: alternately with little problems, simply refraining from starting the
263: Xen daemons when not running Xen.
264:
265: Note that NetBSD as dom0 does not support multiple CPUs. This will
266: limit the performance of the Xen/dom0 workstation approach. In theory
267: the only issue is that the "backend drivers" are not yet MPSAFE:
268: https://mail-index.netbsd.org/netbsd-users/2014/08/29/msg015195.html
269:
270: Installation of NetBSD
271: ----------------------
272:
273: First,
274: [install NetBSD/amd64](/guide/inst/)
275: just as you would if you were not using Xen.
276: However, the partitioning approach is very important.
277:
278: If you want to use RAIDframe for the dom0, there are no special issues
279: for Xen. Typically one provides RAID storage for the dom0, and the
280: domU systems are unaware of RAID. The 2nd-stage loader bootxx_* skips
281: over a RAID1 header to find /boot from a file system within a RAID
282: partition; this is no different when booting Xen.
283:
284: There are 4 styles of providing backing storage for the virtual disks
285: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN.
286:
287: With raw partitions, one has a disklabel (or gpt) partition sized for
288: each virtual disk to be used by the domU. (If you are able to predict
289: how domU usage will evolve, please add an explanation to the HOWTO.
290: Seriously, needs tend to change over time.)
291:
292: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
293: for domU disks. This is almost as efficient as raw disk partitions
294: and more flexible. Hence raw disk partitions should typically not
295: be used.
296:
297: One can use files in the dom0 file system, typically created by dd'ing
298: /dev/zero to create a specific size. This is somewhat less efficient,
299: but very convenient, as one can cp the files for backup, or move them
300: between dom0 hosts.
301:
302: Finally, in theory one can place the files backing the domU disks in a
303: SAN. (This is an invitation for someone who has done this to add a
304: HOWTO page.)
305:
306: Installation of Xen
307: -------------------
308:
309: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
310: pkgsrc (or another matching pair). See [the pkgsrc
311: documentation](https://www.NetBSD.org/docs/pkgsrc/) for help with
312: pkgsrc. Ensure that your packages are recent; the HOWTO does not
313: contemplate old builds.
314:
315:
316: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm. More
317: recent versions have HVM support integrated in the main xentools
318: package. It is entirely reasonable to run only PV guests.
319:
320: Next you need to install the selected Xen kernel itself, which is
321: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz". Copy it to /.
322: For debugging, one may copy xen-debug.gz; this is conceptually similar
323: to DIAGNOSTIC and DEBUG in NetBSD. xen-debug.gz is basically only
324: useful with a serial console. Then, place a NetBSD XEN3_DOM0 kernel
325: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
326: of a NetBSD build. If using i386, use
327: releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz. (If using Xen
328: 3.1 and i386, you may use XEN3_DOM0 with the non-PAE Xen. But you
329: should not use Xen 3.1.) Both xen and the NetBSD kernel may be (and
330: typically are) left compressed.
331:
332: In a dom0, kernfs is mandatory for xend to communicate with the
333: kernel, so ensure that /kern is in fstab. (A standard NetBSD install
334: should already mount /kern.)
335:
336: Because you already installed NetBSD, you have a working boot setup
337: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
338: beginning of your root file system, have /boot, and likely also
339: /boot.cfg. (If not, fix before continuing!)
340:
341: Add a line to to /boot.cfg to boot Xen. See boot.cfg(5) for an
342: example. The basic line is
343:
344: menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
345:
346: which specifies that the dom0 should have 512M, leaving the rest to be
347: allocated for domUs. To use a serial console, use
348:
349: menu=Xen:load /netbsd-XEN3_DOM0.gz;multiboot /xen.gz dom0_mem=512M console=com1 com1=9600,8n1
350:
351: which will use the first serial port for Xen (which counts starting
352: from 1, unlike NetBSD which counts starting from 0), forcing
353: speed/parity. Because the NetBSD command line lacks a
354: "console=pc" argument, it will use the default "xencons" console device,
355: which directs the console I/O through Xen to the same console device Xen
356: itself uses (in this case, the serial port).
357:
358: In an attempt to add performance, one can also add
359:
360: dom0_max_vcpus=1 dom0_vcpus_pin
361:
362: to force only one vcpu to be provided (since NetBSD dom0 can't use
363: more) and to pin that vcpu to a physical CPU. TODO: benchmark this.
364:
365: Xen has [many boot
366: options](http://xenbits.xenproject.org/docs/4.5-testing/misc/xen-command-line.html),
367: and other than dom0 memory and max_vcpus, they are generally not
368: necessary.
369:
370: As with non-Xen systems, you should have a line to boot /netbsd (a
371: kernel that works without Xen). Consider a line to boot /netbsd.ok (a
372: fallback version of the non-Xen kernel, updated manually when you are
373: sure /netbsd is ok). Consider also a line to boot fallback versions
374: of Xen and the dom0 kernel, but note that non-Xen NetBSD can be used
375: to resolve Xen booting issues.
376:
377: Probably you want a default=N line to choose Xen in the absence of
378: intervention.
379:
380: Now, reboot so that you are running a DOM0 kernel under Xen, rather
381: than GENERIC without Xen.
382:
383: Using grub (historic)
384: ---------------------
385:
386: Before NetBSD's native bootloader could support Xen, the use of
387: grub was recommended. If necessary, see the
388: [old grub information](/ports/xen/howto-grub).
389:
390: The [HowTo on Installing into
391: RAID-1](https://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
392: explains how to set up booting a dom0 with Xen using grub with
393: NetBSD's RAIDframe. (This is obsolete with the use of NetBSD's native
394: boot. Now, just create a system with RAID-1, and alter /boot.cfg as
395: described above.)
396:
397: Configuring Xen
398: ---------------
399:
400: Xen logs will be in /var/log/xen.
401:
402: Now, you have a system that will boot Xen and the dom0 kernel, but not
403: do anything else special. Make sure that you have rebooted into Xen.
404: There will be no domUs, and none can be started because you still have
405: to configure the dom0 daemons.
406:
407: The daemons which should be run vary with Xen version and with whether
408: one is using xm or xl. The Xen 3.1, 3.3 and 4.1 packages use xm. Xen
409: 4.2 and up packages use xl. To use xm with 4.2, edit xendomains to
410: use xm instead.
411:
412: For 3.1 and 3.3, you should enable xend and xenbackendd:
413:
414: xend=YES
415: xenbackendd=YES
416:
417: For 4.1 and up, you should enable xencommons. Not enabling xencommons
418: will result in a hang; it is necessary to hit ^C on the console to let
419: the machine finish booting. If you are using xm (default in 4.1, or
420: if you changed xendomains in 4.2), you should also enable xend:
421:
422: xend=YES # only if using xm, and only installed <= 4.2
423: xencommons=YES
424:
425: TODO: Recommend for/against xen-watchdog.
426:
427: After you have configured the daemons and either started them (in the
428: order given) or rebooted, use xm or xl to inspect Xen's boot messages,
429: available resources, and running domains. An example with xl follows:
430:
431: # xl dmesg
432: [xen's boot info]
433: # xl info
434: [available memory, etc.]
435: # xl list
436: Name Id Mem(MB) CPU State Time(s) Console
437: Domain-0 0 64 0 r---- 58.1
438:
439: ### Issues with xencommons
440:
441: xencommons starts xenstored, which stores data on behalf of dom0 and
442: domUs. It does not currently work to stop and start xenstored.
443: Certainly all domUs should be shutdown first, following the sort order
444: of the rc.d scripts. However, the dom0 sets up state with xenstored,
445: and is not notified when xenstored exits, leading to not recreating
446: the state when the new xenstored starts. Until there's a mechanism to
447: make this work, one should not expect to be able to restart xenstored
448: (and thus xencommons). There is currently no reason to expect that
449: this will get fixed any time soon.
450:
451: ### No-longer needed advice about devices
452:
453: The installation of NetBSD should already have created devices for xen
454: (xencons, xenevt, xsd_kva), but if they are not present, create them:
455:
456: cd /dev && sh MAKEDEV xen
457:
458: anita (for testing NetBSD)
459: --------------------------
460:
461: With the setup so far (assuming 4.2/xl), one should be able to run
462: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
463: root, because anita must create a domU):
464:
465: anita --vmm=xl test file:///usr/obj/i386/
466:
467: Alternatively, one can use --vmm=xm to use xm-based domU creation
468: instead (and must, on Xen <= 4.1). TODO: confirm that anita xl really works.
469:
470: Xen-specific NetBSD issues
471: --------------------------
472:
473: There are (at least) two additional things different about NetBSD as a
474: dom0 kernel compared to hardware.
475:
476: One is that the module ABI is different because some of the #defines
477: change, so one must build modules for Xen. As of netbsd-7, the build
478: system does this automatically. TODO: check this. (Before building
479: Xen modules was added, it was awkward to use modules to the point
480: where it was considered that it did not work.)
481:
482: The other difference is that XEN3_DOM0 does not have exactly the same
483: options as GENERIC. While it is debatable whether or not this is a
484: bug, users should be aware of this and can simply add missing config
485: items if desired.
486:
487: Updating NetBSD in a dom0
488: -------------------------
489:
490: This is just like updating NetBSD on bare hardware, assuming the new
491: version supports the version of Xen you are running. Generally, one
492: replaces the kernel and reboots, and then overlays userland binaries
493: and adjusts /etc.
494:
495: Note that one must update both the non-Xen kernel typically used for
496: rescue purposes and the DOM0 kernel used with Xen.
497:
498: Converting from grub to /boot
499: -----------------------------
500:
501: These instructions were [TODO: will be] used to convert a system from
502: grub to /boot. The system was originally installed in February of
503: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
504: over time. Before these commands, it was running NetBSD 6 i386, Xen
505: 4.1 and grub, much like the message linked earlier in the grub
506: section.
507:
508: # Install MBR bootblocks on both disks.
509: fdisk -i /dev/rwd0d
510: fdisk -i /dev/rwd1d
511: # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
512: installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
513: installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
514: # Install secondary boot loader
515: cp -p /usr/mdec/boot /
516: # Create boot.cfg following earlier guidance:
517: menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=512M
518: menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=512M
519: menu=GENERIC:boot
520: menu=GENERIC single-user:boot -s
521: menu=GENERIC.ok:boot netbsd.ok
522: menu=GENERIC.ok single-user:boot netbsd.ok -s
523: menu=Drop to boot prompt:prompt
524: default=1
525: timeout=30
526:
527: TODO: actually do this and fix it if necessary.
528:
529: Upgrading Xen versions
530: ---------------------
531:
532: Minor version upgrades are trivial. Just rebuild/replace the
533: xenkernel version and copy the new xen.gz to / (where /boot.cfg
534: references it), and reboot.
535:
536: Major version upgrades are conceptually not difficult, but can run
537: into all the issues found when installing Xen. Assuming migration
538: from 4.1 to 4.2, remove the xenkernel41 and xentools41 packages and
539: install the xenkernel42 and xentools42 packages. Copy the 4.2 xen.gz
540: to /.
541:
542: Ensure that the contents of /etc/rc.d/xen* are correct. Specifically,
543: they must match the package you just installed and not be left over
544: from some previous installation.
545:
546: Enable the correct set of daemons; see the configuring section above.
547: (Upgrading from 3.x to 4.x without doing this will result in a hang.)
548:
549: Ensure that the domU config files are valid for the new version.
550: Specifically, for 4.x remove autorestart=True, and ensure that disks
551: are specified with numbers as the second argument, as the examples
552: above show, and not NetBSD device names.
553:
554: Hardware known to work
555: ----------------------
556:
557: Arguably, this section is misplaced, and there should be a page of
558: hardware that runs NetBSD/amd64 well, with the mostly-well-founded
559: assumption that NetBSD/xen runs fine on any modern hardware that
560: NetBSD/amd64 runs well on. Until then, we give motherboard/CPU (and
561: sometimes RAM) pairs/triples to aid those choosing a motherboard.
562: Note that Xen systems usually do not run X, so a listing here does not
563: imply that X works at all.
564:
565: Supermicro X9SRL-F, Xeon E5-1650 v2, 96 GiB ECC
566: Supermicro ??, Atom C2758 (8 core), 32 GiB ECC
567: ASUS M5A78L-M/USB3 AM3+ microATX, AMD Piledriver X8 4000MHz, 16 GiB ECC
568:
569: Older hardware:
570:
571: Intel D915GEV, Pentium4 CPU 3.40GHz, 4GB 533MHz Synchronous DDR2
572: INTEL DG33FB, "Intel(R) Core(TM)2 Duo CPU E6850 @ 3.00GHz"
573: INTEL DG33FB, "Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz"
574:
575: Running Xen under qemu
576: ----------------------
577:
578: The astute reader will note that this section is somewhat twisted.
579: However, it can be useful to run Xen under qemu either because the
580: version of NetBSD as a dom0 does not run on the hardware in use, or to
581: generate automated test cases involving Xen.
582:
583: In 2015-01, the following combination was reported to mostly work:
584:
585: host OS: NetBSD/amd64 6.1.4
586: qemu: 2.2.0 from pkgsrc
587: Xen kernel: xenkernel42-4.2.5nb1 from pkgsrc
588: dom0 kernel: NetBSD/amd64 6.1.5
589: Xen tools: xentools42-4.2.5 from pkgsrc
590:
591: See [PR 47720](https://gnats.netbsd.org/47720) for a problem with dom0
592: shutdown.
593:
594: Unprivileged domains (domU)
595: ===========================
596:
597: This section describes general concepts about domUs. It does not
598: address specific domU operating systems or how to install them. The
599: config files for domUs are typically in /usr/pkg/etc/xen, and are
600: typically named so that the file name, domU name and the domU's host
601: name match.
602:
603: The domU is provided with CPU and memory by Xen, configured by the
604: dom0. The domU is provided with disk and network by the dom0,
605: mediated by Xen, and configured in the dom0.
606:
607: Entropy in domUs can be an issue; physical disks and network are on
608: the dom0. NetBSD's /dev/random system works, but is often challenged.
609:
610: Config files
611: ------------
612:
613: There is no good order to present config files and the concepts
614: surrounding what is being configured. We first show an example config
615: file, and then in the various sections give details.
616:
617: See (at least in xentools41) /usr/pkg/share/examples/xen/xmexample*,
618: for a large number of well-commented examples, mostly for running
619: GNU/Linux.
620:
621: The following is an example minimal domain configuration file
622: "/usr/pkg/etc/xen/foo". It is (with only a name change) an actual
623: known working config file on Xen 4.1 (NetBSD 5 amd64 dom0 and NetBSD 5
624: i386 domU). The domU serves as a network file server.
625:
626: # -*- mode: python; -*-
627:
628: kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
629: memory = 1024
630: vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
631: disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
632: 'file:/n0/xen/foo-wd1,0x1,w' ]
633:
634: The domain will have the same name as the file. The kernel has the
635: host/domU name in it, so that on the dom0 one can update the various
636: domUs independently. The vif line causes an interface to be provided,
637: with a specific mac address (do not reuse MAC addresses!), in bridge
638: mode. Two disks are provided, and they are both writable; the bits
639: are stored in files and Xen attaches them to a vnd(4) device in the
640: dom0 on domain creation. The system treats xbd0 as the boot device
641: without needing explicit configuration.
642:
643: By default xm looks for domain config files in /usr/pkg/etc/xen. Note
644: that "xm create" takes the name of a config file, while other commands
645: take the name of a domain. To create the domain, connect to the
646: console, create the domain while attaching the console, shutdown the
647: domain, and see if it has finished stopping, do (or xl with Xen >=
648: 4.2):
649:
650: xm create foo
651: xm console foo
652: xm create -c foo
653: xm shutdown foo
654: xm list
655:
656: Typing ^] will exit the console session. Shutting down a domain is
657: equivalent to pushing the power button; a NetBSD domU will receive a
658: power-press event and do a clean shutdown. Shutting down the dom0
659: will trigger controlled shutdowns of all configured domUs.
660:
661: domU kernels
662: ------------
663:
664: On a physical computer, the BIOS reads sector 0, and a chain of boot
665: loaders finds and loads a kernel. Normally this comes from the root
666: file system. With Xen domUs, the process is totally different. The
667: normal path is for the domU kernel to be a file in the dom0's
668: file system. At the request of the dom0, Xen loads that kernel into a
669: new domU instance and starts execution. While domU kernels can be
670: anyplace, reasonable places to store domU kernels on the dom0 are in /
671: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
672: config files), or in /u0/xen (where the vdisks are).
673:
674: Note that loading the domU kernel from the dom0 implies that boot
675: blocks, /boot, /boot.cfg, and so on are all ignored in the domU.
676: See the VPS section near the end for discussion of alternate ways to
677: obtain domU kernels.
678:
679: CPU and memory
680: --------------
681:
682: A domain is provided with some number of vcpus, less than the number
683: of CPUs seen by the hypervisor. (For a dom0, this is controlled by
684: the boot argument "dom0_max_vcpus=1".) For a domU, it is controlled
685: from the config file by the "vcpus = N" directive.
686:
687: A domain is provided with memory; this is controlled in the config
688: file by "memory = N" (in megabytes). In the straightforward case, the
689: sum of the the memory allocated to the dom0 and all domUs must be less
690: than the available memory.
691:
692: Xen also provides a "balloon" driver, which can be used to let domains
693: use more memory temporarily. TODO: Explain better, and explain how
694: well it works with NetBSD.
695:
696: Virtual disks
697: -------------
698:
699: With the file/vnd style, typically one creates a directory,
700: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
701: domUs. Then, for each domU disk, one writes zeros to a file that then
702: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
703: for the first virtual disk for the domU called foo. Writing zeros to
704: the file serves two purposes. One is that preallocating the contents
705: improves performance. The other is that vnd on sparse files has
706: failed to work. TODO: give working/notworking NetBSD versions for
707: sparse vnd and gnats reference. Note that the use of file/vnd for Xen
708: is not really different than creating a file-backed virtual disk for
709: some other purpose, except that xentools handles the vnconfig
710: commands. To create an empty 4G virtual disk, simply do
711:
712: dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
713:
714: Do not use qemu-img-xen, because this will create sparse file. There
715: have been recent (2015) reports of sparse vnd(4) devices causing
716: lockups, but there is apparently no PR.
717:
718: With the lvm style, one creates logical devices. They are then used
719: similarly to vnds. TODO: Add an example with lvm.
720:
721: In domU config files, the disks are defined as a sequence of 3-tuples.
722: The first element is "method:/path/to/disk". Common methods are
723: "file:" for file-backed vnd. and "phy:" for something that is already
724: a (TODO: character or block) device.
725:
726: The second element is an artifact of how virtual disks are passed to
727: Linux, and a source of confusion with NetBSD Xen usage. Linux domUs
728: are given a device name to associate with the disk, and values like
729: "hda1" or "sda1" are common. In a NetBSD domU, the first disk appears
730: as xbd0, the second as xbd1, and so on. However, xm/xl demand a
731: second argument. The name given is converted to a major/minor by
732: calling stat(2) on the name in /dev and this is passed to the domU.
733: In the general case, the dom0 and domU can be different operating
734: systems, and it is an unwarranted assumption that they have consistent
735: numbering in /dev, or even that the dom0 OS has a /dev. With NetBSD
736: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
737: for the second works fine and avoids this issue. For a GNU/Linux
738: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
739: /dev/hda1.
740:
741: The third element is "w" for writable disks, and "r" for read-only
742: disks.
743:
744: Note that NetBSD by default creates only vnd[0123]. If you need more
745: than 4 total virtual disks at a time, run e.g. "./MAKEDEV vnd4" in the
746: dom0.
747:
748: Note that NetBSD by default creates only xbd[0123]. If you need more
749: virtual disks in a domU, run e.g. "./MAKEDEV xbd4" in the domU.
750:
751: Virtual Networking
752: ------------------
753:
754: Xen provides virtual Ethernets, each of which connects the dom0 and a
755: domU. For each virtual network, there is an interface "xvifN.M" in
756: the dom0, and in domU index N, a matching interface xennetM (NetBSD
757: name). The interfaces behave as if there is an Ethernet with two
758: adapters connected. From this primitive, one can construct various
759: configurations. We focus on two common and useful cases for which
760: there are existing scripts: bridging and NAT.
761:
762: With bridging (in the example above), the domU perceives itself to be
763: on the same network as the dom0. For server virtualization, this is
764: usually best. Bridging is accomplished by creating a bridge(4) device
765: and adding the dom0's physical interface and the various xvifN.0
766: interfaces to the bridge. One specifies "bridge=bridge0" in the domU
767: config file. The bridge must be set up already in the dom0; an
768: example /etc/ifconfig.bridge0 is:
769:
770: create
771: up
772: !brconfig bridge0 add wm0
773:
774: With NAT, the domU perceives itself to be behind a NAT running on the
775: dom0. This is often appropriate when running Xen on a workstation.
776: TODO: NAT appears to be configured by "vif = [ '' ]".
777:
778: The MAC address specified is the one used for the interface in the new
779: domain. The interface in dom0 will use this address XOR'd with
780: 00:00:00:01:00:00. Random MAC addresses are assigned if not given.
781:
782: Sizing domains
783: --------------
784:
785: Modern x86 hardware has vast amounts of resources. However, many
786: virtual servers can function just fine on far less. A system with
787: 512M of RAM and a 4G disk can be a reasonable choice. Note that it is
788: far easier to adjust virtual resources than physical ones. For
789: memory, it's just a config file edit and a reboot. For disk, one can
790: create a new file and vnconfig it (or lvm), and then dump/restore,
791: just like updating physical disks, but without having to be there and
792: without those pesky connectors.
793:
794: Starting domains automatically
795: ------------------------------
796:
797: To start domains foo at bar at boot and shut them down cleanly on dom0
798: shutdown, in rc.conf add:
799:
800: xendomains="foo bar"
801:
802: Note that earlier versions of the xentools41 xendomains rc.d script
803: used xl, when one should use xm with 4.1.
804:
805: Creating specific unprivileged domains (domU)
806: =============================================
807:
808: Creating domUs is almost entirely independent of operating system. We
809: have already presented the basics of config files. Note that you must
810: have already completed the dom0 setup so that "xl list" (or "xm list")
811: works.
812:
813: Creating an unprivileged NetBSD domain (domU)
814: ---------------------------------------------
815:
816: See the earlier config file, and adjust memory. Decide on how much
817: storage you will provide, and prepare it (file or lvm).
818:
819: While the kernel will be obtained from the dom0 file system, the same
820: file should be present in the domU as /netbsd so that tools like
821: savecore(8) can work. (This is helpful but not necessary.)
822:
823: The kernel must be specifically for Xen and for use as a domU. The
824: i386 and amd64 provide the following kernels:
825:
826: i386 XEN3_DOMU
827: i386 XEN3PAE_DOMU
828: amd64 XEN3_DOMU
829:
830: Unless using Xen 3.1 (and you shouldn't) with i386-mode Xen, you must
831: use the PAE version of the i386 kernel.
832:
833: This will boot NetBSD, but this is not that useful if the disk is
834: empty. One approach is to unpack sets onto the disk outside of xen
835: (by mounting it, just as you would prepare a physical disk for a
836: system you can't run the installer on).
837:
838: A second approach is to run an INSTALL kernel, which has a miniroot
839: and can load sets from the network. To do this, copy the INSTALL
840: kernel to / and change the kernel line in the config file to:
841:
842: kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
843:
844: Then, start the domain as "xl create -c configname".
845:
846: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
847: line should be used in the config file.
848:
849: disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
850:
851: After booting the domain, the option to install via CDROM may be
852: selected. The CDROM device should be changed to `xbd1d`.
853:
854: Once done installing, "halt -p" the new domain (don't reboot or halt,
855: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
856: config file), switch the config file back to the XEN3_DOMU kernel,
857: and start the new domain again. Now it should be able to use "root on
858: xbd0a" and you should have a, functional NetBSD domU.
859:
860: TODO: check if this is still accurate.
861: When the new domain is booting you'll see some warnings about *wscons*
862: and the pseudo-terminals. These can be fixed by editing the files
863: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
864: `/etc/ttys`, except *console*, like this:
865:
866: console "/usr/libexec/getty Pc" vt100 on secure
867: ttyE0 "/usr/libexec/getty Pc" vt220 off secure
868: ttyE1 "/usr/libexec/getty Pc" vt220 off secure
869: ttyE2 "/usr/libexec/getty Pc" vt220 off secure
870: ttyE3 "/usr/libexec/getty Pc" vt220 off secure
871:
872: Finally, all screens must be commented out from `/etc/wscons.conf`.
873:
874: It is also desirable to add
875:
876: powerd=YES
877:
878: in rc.conf. This way, the domain will be properly shut down if
879: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
880:
881: It is not strictly necessary to have a kernel (as /netbsd) in the domU
882: file system. However, various programs (e.g. netstat) will use that
883: kernel to look up symbols to read from kernel virtual memory. If
884: /netbsd is not the running kernel, those lookups will fail. (This is
885: not really a Xen-specific issue, but because the domU kernel is
886: obtained from the dom0, it is far more likely to be out of sync or
887: missing with Xen.)
888:
889: Creating an unprivileged Linux domain (domU)
890: --------------------------------------------
891:
892: Creating unprivileged Linux domains isn't much different from
893: unprivileged NetBSD domains, but there are some details to know.
894:
895: First, the second parameter passed to the disk declaration (the '0x1' in
896: the example below)
897:
898: disk = [ 'phy:/dev/wd0e,0x1,w' ]
899:
900: does matter to Linux. It wants a Linux device number here (e.g. 0x300
901: for hda). Linux builds device numbers as: (major \<\< 8 + minor).
902: So, hda1 which has major 3 and minor 1 on a Linux system will have
903: device number 0x301. Alternatively, devices names can be used (hda,
904: hdb, ...) as xentools has a table to map these names to devices
905: numbers. To export a partition to a Linux guest we can use:
906:
907: disk = [ 'phy:/dev/wd0e,0x300,w' ]
908: root = "/dev/hda1 ro"
909:
910: and it will appear as /dev/hda on the Linux system, and be used as root
911: partition.
912:
913: To install the Linux system on the partition to be exported to the
914: guest domain, the following method can be used: install
915: sysutils/e2fsprogs from pkgsrc. Use mke2fs to format the partition
916: that will be the root partition of your Linux domain, and mount it.
917: Then copy the files from a working Linux system, make adjustments in
918: `/etc` (fstab, network config). It should also be possible to extract
919: binary packages such as .rpm or .deb directly to the mounted partition
920: using the appropriate tool, possibly running under NetBSD's Linux
921: emulation. Once the file system has been populated, umount it. If
922: desirable, the file system can be converted to ext3 using tune2fs -j.
923: It should now be possible to boot the Linux guest domain, using one of
924: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
925:
926: To get the Linux console right, you need to add:
927:
928: extra = "xencons=tty1"
929:
930: to your configuration since not all Linux distributions auto-attach a
931: tty to the xen console.
932:
933: Creating an unprivileged Solaris domain (domU)
934: ----------------------------------------------
935:
936: See possibly outdated
937: [Solaris domU instructions](/ports/xen/howto-solaris/).
938:
939:
940: PCI passthrough: Using PCI devices in guest domains
941: ---------------------------------------------------
942:
943: The dom0 can give other domains access to selected PCI
944: devices. This can allow, for example, a non-privileged domain to have
945: access to a physical network interface or disk controller. However,
946: keep in mind that giving a domain access to a PCI device most likely
947: will give the domain read/write access to the whole physical memory,
948: as PCs don't have an IOMMU to restrict memory access to DMA-capable
949: device. Also, it's not possible to export ISA devices to non-dom0
950: domains, which means that the primary VGA adapter can't be exported.
951: A guest domain trying to access the VGA registers will panic.
952:
953: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
954: not been ported to later versions at this time.
955:
956: For a PCI device to be exported to a domU, is has to be attached to
957: the "pciback" driver in dom0. Devices passed to the dom0 via the
958: pciback.hide boot parameter will attach to "pciback" instead of the
959: usual driver. The list of devices is specified as "(bus:dev.func)",
960: where bus and dev are 2-digit hexadecimal numbers, and func a
961: single-digit number:
962:
963: pciback.hide=(00:0a.0)(00:06.0)
964:
965: pciback devices should show up in the dom0's boot messages, and the
966: devices should be listed in the `/kern/xen/pci` directory.
967:
968: PCI devices to be exported to a domU are listed in the "pci" array of
969: the domU's config file, with the format "0000:bus:dev.func".
970:
971: pci = [ '0000:00:06.0', '0000:00:0a.0' ]
972:
973: In the domU an "xpci" device will show up, to which one or more pci
974: buses will attach. Then the PCI drivers will attach to PCI buses as
975: usual. Note that the default NetBSD DOMU kernels do not have "xpci"
976: or any PCI drivers built in by default; you have to build your own
977: kernel to use PCI devices in a domU. Here's a kernel config example;
978: note that only the "xpci" lines are unusual.
979:
980: include "arch/i386/conf/XEN3_DOMU"
981:
982: # Add support for PCI buses to the XEN3_DOMU kernel
983: xpci* at xenbus ?
984: pci* at xpci ?
985:
986: # PCI USB controllers
987: uhci* at pci? dev ? function ? # Universal Host Controller (Intel)
988:
989: # USB bus support
990: usb* at uhci?
991:
992: # USB Hubs
993: uhub* at usb?
994: uhub* at uhub? port ? configuration ? interface ?
995:
996: # USB Mass Storage
997: umass* at uhub? port ? configuration ? interface ?
998: wd* at umass?
999: # SCSI controllers
1000: ahc* at pci? dev ? function ? # Adaptec [23]94x, aic78x0 SCSI
1001:
1002: # SCSI bus support (for both ahc and umass)
1003: scsibus* at scsi?
1004:
1005: # SCSI devices
1006: sd* at scsibus? target ? lun ? # SCSI disk drives
1007: cd* at scsibus? target ? lun ? # SCSI CD-ROM drives
1008:
1009:
1010: NetBSD as a domU in a VPS
1011: =========================
1012:
1013: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
1014: hardware. This section explains how to deal with Xen in a domU as a
1015: virtual private server where you do not control or have access to the
1016: dom0. This is not intended to be an exhaustive list of VPS providers;
1017: only a few are mentioned that specifically support NetBSD.
1018:
1019: VPS operators provide varying degrees of access and mechanisms for
1020: configuration. The big issue is usually how one controls which kernel
1021: is booted, because the kernel is nominally in the dom0 file system (to
1022: which VPS users do not normally have access). A second issue is how
1023: to install NetBSD.
1024: A VPS user may want to compile a kernel for security updates, to run
1025: npf, run IPsec, or any other reason why someone would want to change
1026: their kernel.
1027:
1028: One approach is to have an administrative interface to upload a kernel,
1029: or to select from a prepopulated list. Other approaches are pygrub
1030: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
1031: kernel from the domU file system. This is closer to a regular physical
1032: computer, where someone who controls a machine can replace the kernel.
1033:
1034: A second issue is multiple CPUs. With NetBSD 6, domUs support
1035: multiple vcpus, and it is typical for VPS providers to enable multiple
1036: CPUs for NetBSD domUs.
1037:
1038: pygrub
1039: -------
1040:
1041: pygrub runs in the dom0 and looks into the domU file system. This
1042: implies that the domU must have a kernel in a file system in a format
1043: known to pygrub. As of 2014, pygrub seems to be of mostly historical
1044: interest.
1045:
1046: pvgrub
1047: ------
1048:
1049: pvgrub is a version of grub that uses PV operations instead of BIOS
1050: calls. It is booted from the dom0 as the domU kernel, and then reads
1051: /grub/menu.lst and loads a kernel from the domU file system.
1052:
1053: [Panix](http://www.panix.com/) lets users use pvgrub. Panix reports
1054: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
1055: (and hence with defaults from "newfs -O 2"). See [Panix's pvgrub
1056: page](http://www.panix.com/v-colo/grub.html), which describes only
1057: Linux but should be updated to cover NetBSD :-).
1058:
1059: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
1060: their own kernel. See then [prgmr.com NetBSD
1061: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
1062: (which is in need of updating).
1063:
1064: It appears that [grub's FFS
1065: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
1066: does not support all aspects of modern FFS, but there are also reports
1067: that FFSv2 works fine. At prgmr, typically one has an ext2 or FAT
1068: partition for the kernel with the intent that grub can understand it,
1069: which leads to /netbsd not being the actual kernel. One must remember
1070: to update the special boot partition.
1071:
1072: Amazon
1073: ------
1074:
1075: See the [Amazon EC2 page](/amazon_ec2/).
1076:
1077: TODO items for improving NetBSD/xen
1078: ===================================
1079:
1080: * Make the NetBSD dom0 kernel work with SMP.
1081: * Test the Xen 4.5 packages adequately to be able to recommend them as
1082: the standard approach.
1083: * Get PCI passthrough working on Xen 4.5
1084: * Get pvgrub into pkgsrc, either via xentools or separately.
1085: * grub
1086: * Check/add support to pkgsrc grub2 for UFS2 and arbitrary
1087: fragsize/blocksize (UFS2 support may be present; the point is to
1088: make it so that with any UFS1/UFS2 file system setup that works
1089: with NetBSD grub will also work).
1090: See [pkg/40258](https://gnats.netbsd.org/40258).
1091: * Push patches upstream.
1092: * Get UFS2 patches into pvgrub.
1093: * Add support for PV ops to a version of /boot, and make it usable as
1094: a kernel in Xen, similar to pvgrub.
1095:
1096: Random pointers
1097: ===============
1098:
1099: This section contains links from elsewhere not yet integrated into the
1100: HOWTO, and other guides.
1101:
1102: * http://www.lumbercartel.ca/library/xen/
1103: * http://pbraun.nethence.com/doc/sysutils/xen_netbsd_dom0.html
1104: * https://gmplib.org/~tege/xen.html
CVSweb for NetBSD wikisrc <wikimaster@NetBSD.org> software: FreeBSD-CVSweb