1: Introduction
2: ============
3:
4: [![[Xen
5: screenshot]](http://www.netbsd.org/gallery/in-Action/hubertf-xens.png)](http://www.netbsd.org/gallery/in-Action/hubertf-xen.png)
6:
7: Xen is a hypervisor (or virtual machine monitor) for x86 hardware
8: (i686-class or higher), which supports running multiple guest
9: operating systems on a single physical machine. Xen is a Type 1 or
10: bare-metal hypervisor; one uses the Xen kernel to control the CPU,
11: memory and console, a dom0 operating system which mediates access to
12: other hardware (e.g., disks, network, USB), and one or more domU
13: operating systems which operate in an unprivileged virtualized
14: environment. IO requests from the domU systems are forwarded by the
15: hypervisor (Xen) to the dom0 to be fulfilled.
16:
17: Xen supports two styles of guests. The original is Para-Virtualized
18: (PV) which means that the guest OS does not attempt to access hardware
19: directly, but instead makes hypercalls to the hypervisor. This is
20: analogous to a user-space program making system calls. (The dom0
21: operating system uses PV calls for some functions, such as updating
22: memory mapping page tables, but has direct hardware access for disk
23: and network.) PV guests must be specifically coded for Xen.
24:
25: The more recent style is HVM, which means that the guest does not have
26: code for Xen and need not be aware that it is running under Xen.
27: Attempts to access hardware registers are trapped and emulated. This
28: style is less efficient but can run unmodified guests.
29:
30: Generally any amd64 machine will work with Xen and PV guests. In
31: theory i386 computers without amd64 support can be used for Xen <=
32: 4.2, but we have no recent reports of this working (this is a hint).
33: For HVM guests, the VT or VMX CPU feature (Intel) or SVM/HVM/VT
34: (amd64) is needed; "cpuctl identify 0" will show this. TODO: Clean up
35: and check the above features.
36:
37: At boot, the dom0 kernel is loaded as a module with Xen as the kernel.
38: The dom0 can start one or more domUs. (Booting is explained in detail
39: in the dom0 section.)
40:
41: NetBSD supports Xen in that it can serve as dom0, be used as a domU,
42: and that Xen kernels and tools are available in pkgsrc. This HOWTO
43: attempts to address both the case of running a NetBSD dom0 on hardware
44: and running domUs under it (NetBSD and other), and also running NetBSD
45: as a domU in a VPS.
46:
47: Some versions of Xen support "PCI passthrough", which means that
48: specific PCI devices can be made available to a specific domU instead
49: of the dom0. This can be useful to let a domU run X11, or access some
50: network interface or other peripheral.
51:
52: NetBSD 6 and earlier supported Xen 2; support was removed from NetBSD
53: 7. Xen 2 has been removed from pkgsrc.
54:
55: Prerequisites
56: -------------
57:
58: Installing NetBSD/Xen is not extremely difficult, but it is more
59: complex than a normal installation of NetBSD.
60: In general, this HOWTO is occasionally overly restrictive about how
61: things must be done, guiding the reader to stay on the established
62: path when there are no known good reasons to stray.
63:
64: This HOWTO presumes a basic familiarity with the Xen system
65: architecture. This HOWTO presumes familiarity with installing NetBSD
66: on i386/amd64 hardware and installing software from pkgsrc.
67: See also the [Xen website](http://www.xenproject.org/).
68:
69: Versions of Xen and NetBSD
70: ==========================
71:
72: Most of the installation concepts and instructions are independent
73: of Xen version and NetBSD version. This section gives advice on
74: which version to choose. Versions not in pkgsrc and older unsupported
75: versions of NetBSD are intentionally ignored.
76:
77: Xen
78: ---
79:
80: In NetBSD, Xen is provided in pkgsrc, via matching pairs of packages
81: xenkernel and xentools. We will refer only to the kernel versions,
82: but note that both packages must be installed together and must have
83: matching versions.
84:
85: xenkernel3 provides Xen 3.1. It is no longer maintained by Xen, and the last applied security patch was in
86: 2011. Thus, it should not be used. It supports PCI passthrough,
87: which is why people use it anyway. Xen 3.1 supports i386, both PAE and
88: non-PAE.
89:
90: xenkernel33 provides Xen 3.3. It is no longer maintained by Xen, and
91: the last applied security patch was in 2012. Thus, it should not be
92: used. Xen 3.3 supports i386, but only in PAE mode. There are no good
93: reasons to run this version.
94:
95: xenkernel41 provides Xen 4.1. It is no longer maintained by Xen, but
96: as of 2016-12 received backported security patches. Xen 4.1 supports
97: i386, but only in PAE mode. There are no good reasons to run this
98: version.
99:
100: xenkernel42 provides Xen 4.2. It is no longer maintained by Xen, but
101: as of 2016-12 received backported security patches. Xen 4.2 supports
102: i386, but only in PAE mode. The only reason to run this is if you
103: need to use xm instead of xl, or if you need to run an i386 dom0
104: (because your hardware is i386 only).
105:
106: xenkernel45 provides Xen 4.5. As of 2016-12, security patches were
107: released by Xen and applied to pkgsrc. Xen 4.5 requires an amd64
108: dom0, but domUs can be amd64 or i386 PAE. TODO: It is either a
109: conservative choice or somewhat old.
110:
111: xenkernel45 provides Xen 4.6. It is new to pkgsrc as of 2016-05. As
112: of 2016-12, security patches were released by Xen and applied to
113: pkgsrc. Xen 4.6 requires an amd64 dom0, but domUs can be amd64 or
114: i386 PAE. TODO: It is either a somewhat aggressive choice or the
115: standard choice
116:
117: Xen 4.7 (released 2016-06) and 4.8 (released 2016-12) are not yet in
118: pkgsrc.
119:
120: See also the [Xen Security Advisory page](http://xenbits.xen.org/xsa/).
121:
122: Note that NetBSD support is called XEN3. It works with Xen 3 and Xen
123: 4 because the hypercall interface has been stable.
124:
125: Xen command program
126: -------------------
127:
128: Early Xen used a program called xm to manipulate the system from the
129: dom0. Starting in 4.1, a replacement program with similar behavior
130: called xl is provided, but it does not work well in 4.1. In 4.2, both
131: xm and xl work fine. 4.4 is the last version that has xm. You must
132: choose one or the other, because it affects which daemons you run.
133: However, the rc.d scripts provided by xentools packages expect a
134: particular version, and you should use the version used by the
135: scripts.
136:
137: NetBSD
138: ------
139:
140: The netbsd-6, netbsd-7, and -current branches are all reasonable
141: choices, with more or less the same considerations for non-Xen use.
142: Therefore, netbsd-7 is recommended as the stable version of the most
143: recent release for production use. For those wanting to learn Xen or
144: without production stability concerns, netbsd-7 is still likely most
145: appropriate, but -current is also a reasonable choice. Xen runs fine
146: on netbsd-5, but the xentools packages are likely difficult to build.
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: Architecture
154: ------------
155:
156: Xen itself can run on i386 (Xen < 3.1) or amd64 machines (all Xen
157: versions). (Practically, almost any computer where one would want to
158: run Xen today supports amd64.)
159:
160: Xen, the dom0 kernel, and each domU kernel can be either i386 or
161: amd64. When building a xenkernel package, one obtains i386 on an i386
162: host, and amd64 on an amd64 host. If the Xen kernel is i386, then the
163: dom0 kernel and all domU kernels must be i386. With an amd64 Xen
164: kernel, an amd64 dom0 kernel is known to work, and an i386PAE dom0
165: kernel should in theory work. An amd64 Xen/dom0 is known to support
166: both i386PAE and amd64 domUs.
167:
168: i386 dom0 and domU kernels must be PAE (except for Xen 3.1); these are
169: built by default. (Note that emacs (at least) fails if run on i386
170: with PAE when built without, and vice versa, presumably due to bugs in
171: the undump code.)
172:
173: Because of the above, the standard approach is to use amd64 for the
174: dom0.
175:
176: Xen 4.2 is the last version to support i386 as a host. TODO: Clarify
177: if this is about the CPU, the Xen kernel, or the dom0 kernel having to
178: be amd64.
179:
180:
181: Stability
182: ---------
183:
184: Mostly, NetBSD as a dom0 or domU is quite stable.
185: However, there are some open PRs indicating problems.
186:
187: - [PR 48125](http://gnats.netbsd.org/48125)
188: - [PR 47720](http://gnats.netbsd.org/47720)
189:
190: Note also that there are issues with sparse vnd(4) instances, but
191: these are not about Xen -- they just are noticed with sparse vnd(4)
192: instances in support of virtual disks in a dom0.
193:
194: Recommendation
195: --------------
196:
197: Therefore, this HOWTO recommends running xenkernel45 or xenkernel46,
198: xl, the NetBSD 7 stable branch, and to use an amd64 kernel as the
199: dom0. Either the i386PAE or amd64 version of NetBSD may be used as
200: domUs.
201:
202: Build problems
203: --------------
204:
205: Ideally, all versions of Xen in pkgsrc would build on all versions of
206: NetBSD on both i386 and amd64. However, that isn't the case. Besides
207: aging code and aging compilers, qemu (included in xentools for HVM
208: support) is difficult to build. The following are known to work or FAIL:
209:
210: xenkernel3 netbsd-5 amd64
211: xentools3 netbsd-5 amd64
212: xentools3=hvm netbsd-5 amd64 ????
213: xenkernel33 netbsd-5 amd64
214: xentools33 netbsd-5 amd64
215: xenkernel41 netbsd-5 amd64
216: xentools41 netbsd-5 amd64
217: xenkernel42 netbsd-5 amd64
218: xentools42 netbsd-5 amd64
219:
220: xenkernel3 netbsd-6 i386 FAIL
221: xentools3 netbsd-6 i386
222: xentools3-hvm netbsd-6 i386 FAIL (dependencies fail)
223: xenkernel33 netbsd-6 i386
224: xentools33 netbsd-6 i386
225: xenkernel41 netbsd-6 i386
226: xentools41 netbsd-6 i386
227: xenkernel42 netbsd-6 i386
228: xentools42 netbsd-6 i386 *MIXED
229:
230: (all 3 and 33 seem to FAIL)
231: xenkernel41 netbsd-7 i386
232: xentools41 netbsd-7 i386
233: xenkernel42 netbsd-7 i386
234: xentools42 netbsd-7 i386 ??FAIL
235:
236: (*On netbsd-6 i386, there is a xentools42 in the 2014Q3 official builds,
237: but it does not build for gdt.)
238:
239: NetBSD as a dom0
240: ================
241:
242: NetBSD can be used as a dom0 and works very well. The following
243: sections address installation, updating NetBSD, and updating Xen.
244: Note that it doesn't make sense to talk about installing a dom0 OS
245: without also installing Xen itself. We first address installing
246: NetBSD, which is not yet a dom0, and then adding Xen, pivoting the
247: NetBSD install to a dom0 install by just changing the kernel and boot
248: configuration.
249:
250: For experimenting with Xen, a machine with as little as 1G of RAM and
251: 100G of disk can work. For running many domUs in productions, far
252: more will be needed.
253:
254: Styles of dom0 operation
255: ------------------------
256:
257: There are two basic ways to use Xen. The traditional method is for
258: the dom0 to do absolutely nothing other than providing support to some
259: number of domUs. Such a system was probably installed for the sole
260: purpose of hosting domUs, and sits in a server room on a UPS.
261:
262: The other way is to put Xen under a normal-usage computer, so that the
263: dom0 is what the computer would have been without Xen, perhaps a
264: desktop or laptop. Then, one can run domUs at will. Purists will
265: deride this as less secure than the previous approach, and for a
266: computer whose purpose is to run domUs, they are right. But Xen and a
267: dom0 (without domUs) is not meaningfully less secure than the same
268: things running without Xen. One can boot Xen or boot regular NetBSD
269: alternately with little problems, simply refraining from starting the
270: Xen daemons when not running Xen.
271:
272: Note that NetBSD as dom0 does not support multiple CPUs. This will
273: limit the performance of the Xen/dom0 workstation approach. In theory
274: the only issue is that the "backend drivers" are not yet MPSAFE:
275: http://mail-index.netbsd.org/netbsd-users/2014/08/29/msg015195.html
276:
277: Installation of NetBSD
278: ----------------------
279:
280: First,
281: [install NetBSD/amd64](/guide/inst/)
282: just as you would if you were not using Xen.
283: However, the partitioning approach is very important.
284:
285: If you want to use RAIDframe for the dom0, there are no special issues
286: for Xen. Typically one provides RAID storage for the dom0, and the
287: domU systems are unaware of RAID. The 2nd-stage loader bootxx_* skips
288: over a RAID1 header to find /boot from a file system within a RAID
289: partition; this is no different when booting Xen.
290:
291: There are 4 styles of providing backing storage for the virtual disks
292: used by domUs: raw partitions, LVM, file-backed vnd(4), and SAN.
293:
294: With raw partitions, one has a disklabel (or gpt) partition sized for
295: each virtual disk to be used by the domU. (If you are able to predict
296: how domU usage will evolve, please add an explanation to the HOWTO.
297: Seriously, needs tend to change over time.)
298:
299: One can use [lvm(8)](/guide/lvm/) to create logical devices to use
300: for domU disks. This is almost as efficient as raw disk partitions
301: and more flexible. Hence raw disk partitions should typically not
302: be used.
303:
304: One can use files in the dom0 file system, typically created by dd'ing
305: /dev/zero to create a specific size. This is somewhat less efficient,
306: but very convenient, as one can cp the files for backup, or move them
307: between dom0 hosts.
308:
309: Finally, in theory one can place the files backing the domU disks in a
310: SAN. (This is an invitation for someone who has done this to add a
311: HOWTO page.)
312:
313: Installation of Xen
314: -------------------
315:
316: In the dom0, install sysutils/xenkernel42 and sysutils/xentools42 from
317: pkgsrc (or another matching pair).
318: See [the pkgsrc
319: documentation](http://www.NetBSD.org/docs/pkgsrc/) for help with pkgsrc.
320:
321: For Xen 3.1, support for HVM guests is in sysutils/xentool3-hvm. More
322: recent versions have HVM support integrated in the main xentools
323: package. It is entirely reasonable to run only PV guests.
324:
325: Next you need to install the selected Xen kernel itself, which is
326: installed by pkgsrc as "/usr/pkg/xen*-kernel/xen.gz". Copy it to /.
327: For debugging, one may copy xen-debug.gz; this is conceptually similar
328: to DIAGNOSTIC and DEBUG in NetBSD. xen-debug.gz is basically only
329: useful with a serial console. Then, place a NetBSD XEN3_DOM0 kernel
330: in /, copied from releasedir/amd64/binary/kernel/netbsd-XEN3_DOM0.gz
331: of a NetBSD build. If using i386, use
332: releasedir/i386/binary/kernel/netbsd-XEN3PAE_DOM0.gz. (If using Xen
333: 3.1 and i386, you may use XEN3_DOM0 with the non-PAE Xen. But you
334: should not use Xen 3.1.) Both xen and the NetBSD kernel may be (and
335: typically are) left compressed.
336:
337: In a dom0 kernel, kernfs is mandatory for xend to communicate with the
338: kernel, so ensure that /kern is in fstab. TODO: Say this is default,
339: or file a PR and give a reference.
340:
341: Because you already installed NetBSD, you have a working boot setup
342: with an MBR bootblock, either bootxx_ffsv1 or bootxx_ffsv2 at the
343: beginning of your root file system, /boot present, and likely
344: /boot.cfg. (If not, fix before continuing!)
345:
346: Add a line to to /boot.cfg to boot Xen. See boot.cfg(5) for an
347: example. The basic line is
348:
349: menu=Xen:load /netbsd-XEN3_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
350:
351: which specifies that the dom0 should have 256M, leaving the rest to be
352: allocated for domUs. To use a serial console, use
353:
354: menu=Xen:load /netbsd-XEN3_DOM0.gz console=com0;multiboot /xen.gz dom0_mem=256M console=com1 com1=9600,8n1
355:
356: which will use the first serial port for Xen (which counts starting
357: from 1), forcing speed/parity, and also for NetBSD (which counts
358: starting at 0). 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) and fallback versions of the non-Xen
372: kernel, Xen, and the dom0 kernel.
373:
374: Now, reboot so that you are running a DOM0 kernel under Xen, rather
375: than GENERIC without Xen.
376:
377: Using grub (historic)
378: ---------------------
379:
380: Before NetBSD's native bootloader could support Xen, the use of
381: grub was recommended. If necessary, see the
382: [old grub information](/ports/xen/howto-grub/).
383:
384: The [HowTo on Installing into
385: RAID-1](http://mail-index.NetBSD.org/port-xen/2006/03/01/0010.html)
386: explains how to set up booting a dom0 with Xen using grub with
387: NetBSD's RAIDframe. (This is obsolete with the use of NetBSD's native
388: boot.)
389:
390: Configuring Xen
391: ---------------
392:
393: Xen logs will be in /var/log/xen.
394:
395: Now, you have a system that will boot Xen and the dom0 kernel, but not
396: do anything else special. Make sure that you have rebooted into Xen.
397: There will be no domUs, and none can be started because you still have
398: to configure the dom0 daemons.
399:
400: The daemons which should be run vary with Xen version and with whether
401: one is using xm or xl. The Xen 3.1 and 3.3 packages use xm. Xen 4.1
402: and higher packages use xl. While is is possible to use xm with some
403: 4.x versions (TODO: 4.1 and 4.2?), the pkgsrc-provided rc.d scripts do
404: not support this as of 2014-12-26, and thus the HOWTO does not support
405: it either. (Make sure your packages are reasonably recent.)
406:
407: For "xm" (3.1 and 3.3), you should enable xend and xenbackendd (but
408: note that you should be using 4.x):
409:
410: xend=YES
411: xenbackendd=YES
412:
413: For "xl" (4.x), you should enabled xend and xencommons (xenstored).
414: Trying to boot 4.x without xencommons=YES will result in a hang; it is
415: necessary to hit ^C on the console to let the machine finish booting.
416: TODO: explain why xend is installed by the package.
417:
418: xencommons=YES
419:
420: The installation of NetBSD should already have created devices for xen
421: (xencons, xenevt), but if they are not present, create them:
422:
423: cd /dev && sh MAKEDEV xen
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: anita (for testing NetBSD)
452: --------------------------
453:
454: With the setup so far (assuming 4.2/xl), one should be able to run
455: anita (see pkgsrc/misc/py-anita) to test NetBSD releases, by doing (as
456: root, because anita must create a domU):
457:
458: anita --vmm=xl test file:///usr/obj/i386/
459:
460: Alternatively, one can use --vmm=xm to use xm-based domU creation
461: instead (and must, on Xen <= 4.1). TODO: confirm that anita xl really works.
462:
463: Xen-specific NetBSD issues
464: --------------------------
465:
466: There are (at least) two additional things different about NetBSD as a
467: dom0 kernel compared to hardware.
468:
469: One is that the module ABI is different because some of the #defines
470: change, so one must build modules for Xen. As of netbsd-7, the build
471: system does this automatically. TODO: check this. (Before building
472: Xen modules was added, it was awkward to use modules to the point
473: where it was considered that it did not work.)
474:
475: The other difference is that XEN3_DOM0 does not have exactly the same
476: options as GENERIC. While it is debatable whether or not this is a
477: bug, users should be aware of this and can simply add missing config
478: items if desired.
479:
480: Updating NetBSD in a dom0
481: -------------------------
482:
483: This is just like updating NetBSD on bare hardware, assuming the new
484: version supports the version of Xen you are running. Generally, one
485: replaces the kernel and reboots, and then overlays userland binaries
486: and adjusts /etc.
487:
488: Note that one must update both the non-Xen kernel typically used for
489: rescue purposes and the DOM0 kernel used with Xen.
490:
491: Converting from grub to /boot
492: -----------------------------
493:
494: These instructions were [TODO: will be] used to convert a system from
495: grub to /boot. The system was originally installed in February of
496: 2006 with a RAID1 setup and grub to boot Xen 2, and has been updated
497: over time. Before these commands, it was running NetBSD 6 i386, Xen
498: 4.1 and grub, much like the message linked earlier in the grub
499: section.
500:
501: # Install MBR bootblocks on both disks.
502: fdisk -i /dev/rwd0d
503: fdisk -i /dev/rwd1d
504: # Install NetBSD primary boot loader (/ is FFSv1) into RAID1 components.
505: installboot -v /dev/rwd0d /usr/mdec/bootxx_ffsv1
506: installboot -v /dev/rwd1d /usr/mdec/bootxx_ffsv1
507: # Install secondary boot loader
508: cp -p /usr/mdec/boot /
509: # Create boot.cfg following earlier guidance:
510: menu=Xen:load /netbsd-XEN3PAE_DOM0.gz console=pc;multiboot /xen.gz dom0_mem=256M
511: menu=Xen.ok:load /netbsd-XEN3PAE_DOM0.ok.gz console=pc;multiboot /xen.ok.gz dom0_mem=256M
512: menu=GENERIC:boot
513: menu=GENERIC single-user:boot -s
514: menu=GENERIC.ok:boot netbsd.ok
515: menu=GENERIC.ok single-user:boot netbsd.ok -s
516: menu=Drop to boot prompt:prompt
517: default=1
518: timeout=30
519:
520: TODO: actually do this and fix it if necessary.
521:
522: Upgrading Xen versions
523: ---------------------
524:
525: Minor version upgrades are trivial. Just rebuild/replace the
526: xenkernel version and copy the new xen.gz to / (where /boot.cfg
527: references it), and reboot.
528:
529: Major version upgrades are conceptually not difficult, but can run
530: into all the issues found when installing Xen. Assuming migration
531: from 4.1 to 4.2, remove the xenkernel41 and xentools41 packages and
532: install the xenkernel42 and xentools42 packages. Copy the 4.2 xen.gz
533: to /.
534:
535: Ensure that the contents of /etc/rc.d/xen* are correct. Specifically,
536: they must match the package you just installed and not be left over
537: from some previous installation.
538:
539: Enable the correct set of daemons; see the configuring section above.
540: (Upgrading from 3.x to 4.x without doing this will result in a hang.)
541:
542: Ensure that the domU config files are valid for the new version.
543: Specifically, for 4.x remove autorestart=True, and ensure that disks
544: are specified with numbers as the second argument, as the examples
545: above show, and not NetBSD device names.
546:
547: Hardware known to work
548: ----------------------
549:
550: Arguably, this section is misplaced, and there should be a page of
551: hardware that runs NetBSD/amd64 well, with the mostly-well-founded
552: assumption that NetBSD/xen runs fine on any modern hardware that
553: NetBSD/amd64 runs well on. Until then, we give motherboard/CPU/RAM
554: triples to aid those choosing a motherboard. Note that Xen systems
555: usually do not run X, so a listing here does not imply that X works at
556: all.
557:
558: Supermicro X9SRL-F, Xeon E5-1650 v2, 96 GiB ECC
559: Supermicro ??, Atom C2758 (8 core), 32 GiB ECC
560: ASUS M5A78L-M/USB3 AM3+ microATX, AMD Piledriver X8 4000MHz, 16 GiB ECC
561:
562: Older hardware:
563:
564: Intel D915GEV, Pentium4 CPU 3.40GHz, 4GB 533MHz Synchronous DDR2
565:
566: Running Xen under qemu
567: ----------------------
568:
569: The astute reader will note that this section is somewhat twisted.
570: However, it can be useful to run Xen under qemu either because the
571: version of NetBSD as a dom0 does not run on the hardware in use, or to
572: generate automated test cases involving Xen.
573:
574: In 2015-01, the following combination was reported to mostly work:
575:
576: host OS: NetBSD/amd64 6.1.4
577: qemu: 2.2.0 from pkgsrc
578: Xen kernel: xenkernel42-4.2.5nb1 from pkgsrc
579: dom0 kernel: NetBSD/amd64 6.1.5
580: Xen tools: xentools42-4.2.5 from pkgsrc
581:
582: See [PR 47720](http://gnats.netbsd.org/47720) for a problem with dom0
583: shutdown.
584:
585: Unprivileged domains (domU)
586: ===========================
587:
588: This section describes general concepts about domUs. It does not
589: address specific domU operating systems or how to install them. The
590: config files for domUs are typically in /usr/pkg/etc/xen, and are
591: typically named so that the file name, domU name and the domU's host
592: name match.
593:
594: The domU is provided with CPU and memory by Xen, configured by the
595: dom0. The domU is provided with disk and network by the dom0,
596: mediated by Xen, and configured in the dom0.
597:
598: Entropy in domUs can be an issue; physical disks and network are on
599: the dom0. NetBSD's /dev/random system works, but is often challenged.
600:
601: Config files
602: ------------
603:
604: There is no good order to present config files and the concepts
605: surrounding what is being configured. We first show an example config
606: file, and then in the various sections give details.
607:
608: See (at least in xentools41) /usr/pkg/share/examples/xen/xmexample*,
609: for a large number of well-commented examples, mostly for running
610: GNU/Linux.
611:
612: The following is an example minimal domain configuration file
613: "/usr/pkg/etc/xen/foo". It is (with only a name change) an actual
614: known working config file on Xen 4.1 (NetBSD 5 amd64 dom0 and NetBSD 5
615: i386 domU). The domU serves as a network file server.
616:
617: # -*- mode: python; -*-
618:
619: kernel = "/netbsd-XEN3PAE_DOMU-i386-foo.gz"
620: memory = 1024
621: vif = [ 'mac=aa:00:00:d1:00:09,bridge=bridge0' ]
622: disk = [ 'file:/n0/xen/foo-wd0,0x0,w',
623: 'file:/n0/xen/foo-wd1,0x1,w' ]
624:
625: The domain will have the same name as the file. The kernel has the
626: host/domU name in it, so that on the dom0 one can update the various
627: domUs independently. The vif line causes an interface to be provided,
628: with a specific mac address (do not reuse MAC addresses!), in bridge
629: mode. Two disks are provided, and they are both writable; the bits
630: are stored in files and Xen attaches them to a vnd(4) device in the
631: dom0 on domain creation. The system treats xbd0 as the boot device
632: without needing explicit configuration.
633:
634: By default xm looks for domain config files in /usr/pkg/etc/xen. Note
635: that "xm create" takes the name of a config file, while other commands
636: take the name of a domain. To create the domain, connect to the
637: console, create the domain while attaching the console, shutdown the
638: domain, and see if it has finished stopping, do (or xl with Xen >=
639: 4.2):
640:
641: xm create foo
642: xm console foo
643: xm create -c foo
644: xm shutdown foo
645: xm list
646:
647: Typing ^] will exit the console session. Shutting down a domain is
648: equivalent to pushing the power button; a NetBSD domU will receive a
649: power-press event and do a clean shutdown. Shutting down the dom0
650: will trigger controlled shutdowns of all configured domUs.
651:
652: domU kernels
653: ------------
654:
655: On a physical computer, the BIOS reads sector 0, and a chain of boot
656: loaders finds and loads a kernel. Normally this comes from the root
657: file system. With Xen domUs, the process is totally different. The
658: normal path is for the domU kernel to be a file in the dom0's
659: file system. At the request of the dom0, Xen loads that kernel into a
660: new domU instance and starts execution. While domU kernels can be
661: anyplace, reasonable places to store domU kernels on the dom0 are in /
662: (so they are near the dom0 kernel), in /usr/pkg/etc/xen (near the
663: config files), or in /u0/xen (where the vdisks are).
664:
665: Note that loading the domU kernel from the dom0 implies that boot
666: blocks, /boot, /boot.cfg, and so on are all ignored in the domU.
667: See the VPS section near the end for discussion of alternate ways to
668: obtain domU kernels.
669:
670: CPU and memory
671: --------------
672:
673: A domain is provided with some number of vcpus, less than the number
674: of CPUs seen by the hypervisor. (For a dom0, this is controlled by
675: the boot argument "dom0_max_vcpus=1".) For a domU, it is controlled
676: from the config file by the "vcpus = N" directive.
677:
678: A domain is provided with memory; this is controlled in the config
679: file by "memory = N" (in megabytes). In the straightforward case, the
680: sum of the the memory allocated to the dom0 and all domUs must be less
681: than the available memory.
682:
683: Xen also provides a "balloon" driver, which can be used to let domains
684: use more memory temporarily. TODO: Explain better, and explain how
685: well it works with NetBSD.
686:
687: Virtual disks
688: -------------
689:
690: With the file/vnd style, typically one creates a directory,
691: e.g. /u0/xen, on a disk large enough to hold virtual disks for all
692: domUs. Then, for each domU disk, one writes zeros to a file that then
693: serves to hold the virtual disk's bits; a suggested name is foo-xbd0
694: for the first virtual disk for the domU called foo. Writing zeros to
695: the file serves two purposes. One is that preallocating the contents
696: improves performance. The other is that vnd on sparse files has
697: failed to work. TODO: give working/notworking NetBSD versions for
698: sparse vnd. Note that the use of file/vnd for Xen is not really
699: different than creating a file-backed virtual disk for some other
700: purpose, except that xentools handles the vnconfig commands. To
701: create an empty 4G virtual disk, simply do
702:
703: dd if=/dev/zero of=foo-xbd0 bs=1m count=4096
704:
705: Do not use qemu-img-xen, because this will create sparse file. There
706: have been recent (2015) reports of sparse vnd(4) devices causing
707: lockups, but there is apparently no PR.
708:
709: With the lvm style, one creates logical devices. They are then used
710: similarly to vnds. TODO: Add an example with lvm.
711:
712: In domU config files, the disks are defined as a sequence of 3-tuples.
713: The first element is "method:/path/to/disk". Common methods are
714: "file:" for file-backed vnd. and "phy:" for something that is already
715: a (TODO: character or block) device.
716:
717: The second element is an artifact of how virtual disks are passed to
718: Linux, and a source of confusion with NetBSD Xen usage. Linux domUs
719: are given a device name to associate with the disk, and values like
720: "hda1" or "sda1" are common. In a NetBSD domU, the first disk appears
721: as xbd0, the second as xbd1, and so on. However, xm/xl demand a
722: second argument. The name given is converted to a major/minor by
723: calling stat(2) on the name in /dev and this is passed to the domU.
724: In the general case, the dom0 and domU can be different operating
725: systems, and it is an unwarranted assumption that they have consistent
726: numbering in /dev, or even that the dom0 OS has a /dev. With NetBSD
727: as both dom0 and domU, using values of 0x0 for the first disk and 0x1
728: for the second works fine and avoids this issue. For a GNU/Linux
729: guest, one can create /dev/hda1 in /dev, or to pass 0x301 for
730: /dev/hda1.
731:
732: The third element is "w" for writable disks, and "r" for read-only
733: disks.
734:
735: Virtual Networking
736: ------------------
737:
738: Xen provides virtual Ethernets, each of which connects the dom0 and a
739: domU. For each virtual network, there is an interface "xvifN.M" in
740: the dom0, and in domU index N, a matching interface xennetM (NetBSD
741: name). The interfaces behave as if there is an Ethernet with two
742: adapters connected. From this primitive, one can construct various
743: configurations. We focus on two common and useful cases for which
744: there are existing scripts: bridging and NAT.
745:
746: With bridging (in the example above), the domU perceives itself to be
747: on the same network as the dom0. For server virtualization, this is
748: usually best. Bridging is accomplished by creating a bridge(4) device
749: and adding the dom0's physical interface and the various xvifN.0
750: interfaces to the bridge. One specifies "bridge=bridge0" in the domU
751: config file. The bridge must be set up already in the dom0; an
752: example /etc/ifconfig.bridge0 is:
753:
754: create
755: up
756: !brconfig bridge0 add wm0
757:
758: With NAT, the domU perceives itself to be behind a NAT running on the
759: dom0. This is often appropriate when running Xen on a workstation.
760: TODO: NAT appears to be configured by "vif = [ '' ]".
761:
762: The MAC address specified is the one used for the interface in the new
763: domain. The interface in dom0 will use this address XOR'd with
764: 00:00:00:01:00:00. Random MAC addresses are assigned if not given.
765:
766: Sizing domains
767: --------------
768:
769: Modern x86 hardware has vast amounts of resources. However, many
770: virtual servers can function just fine on far less. A system with
771: 256M of RAM and a 4G disk can be a reasonable choice. Note that it is
772: far easier to adjust virtual resources than physical ones. For
773: memory, it's just a config file edit and a reboot. For disk, one can
774: create a new file and vnconfig it (or lvm), and then dump/restore,
775: just like updating physical disks, but without having to be there and
776: without those pesky connectors.
777:
778: Starting domains automatically
779: ------------------------------
780:
781: To start domains foo at bar at boot and shut them down cleanly on dom0
782: shutdown, in rc.conf add:
783:
784: xendomains="foo bar"
785:
786: Note that earlier versions of the xentools41 xendomains rc.d script
787: used xl, when one should use xm with 4.1.
788:
789: Creating specific unprivileged domains (domU)
790: =============================================
791:
792: Creating domUs is almost entirely independent of operating system. We
793: have already presented the basics of config files. Note that you must
794: have already completed the dom0 setup so that "xl list" (or "xm list")
795: works.
796:
797: Creating an unprivileged NetBSD domain (domU)
798: ---------------------------------------------
799:
800: See the earlier config file, and adjust memory. Decide on how much
801: storage you will provide, and prepare it (file or lvm).
802:
803: While the kernel will be obtained from the dom0 file system, the same
804: file should be present in the domU as /netbsd so that tools like
805: savecore(8) can work. (This is helpful but not necessary.)
806:
807: The kernel must be specifically for Xen and for use as a domU. The
808: i386 and amd64 provide the following kernels:
809:
810: i386 XEN3_DOMU
811: i386 XEN3PAE_DOMU
812: amd64 XEN3_DOMU
813:
814: Unless using Xen 3.1 (and you shouldn't) with i386-mode Xen, you must
815: use the PAE version of the i386 kernel.
816:
817: This will boot NetBSD, but this is not that useful if the disk is
818: empty. One approach is to unpack sets onto the disk outside of xen
819: (by mounting it, just as you would prepare a physical disk for a
820: system you can't run the installer on).
821:
822: A second approach is to run an INSTALL kernel, which has a miniroot
823: and can load sets from the network. To do this, copy the INSTALL
824: kernel to / and change the kernel line in the config file to:
825:
826: kernel = "/home/bouyer/netbsd-INSTALL_XEN3_DOMU"
827:
828: Then, start the domain as "xl create -c configname".
829:
830: Alternatively, if you want to install NetBSD/Xen with a CDROM image, the following
831: line should be used in the config file.
832:
833: disk = [ 'phy:/dev/wd0e,0x1,w', 'phy:/dev/cd0a,0x2,r' ]
834:
835: After booting the domain, the option to install via CDROM may be
836: selected. The CDROM device should be changed to `xbd1d`.
837:
838: Once done installing, "halt -p" the new domain (don't reboot or halt,
839: it would reload the INSTALL_XEN3_DOMU kernel even if you changed the
840: config file), switch the config file back to the XEN3_DOMU kernel,
841: and start the new domain again. Now it should be able to use "root on
842: xbd0a" and you should have a, functional NetBSD domU.
843:
844: TODO: check if this is still accurate.
845: When the new domain is booting you'll see some warnings about *wscons*
846: and the pseudo-terminals. These can be fixed by editing the files
847: `/etc/ttys` and `/etc/wscons.conf`. You must disable all terminals in
848: `/etc/ttys`, except *console*, like this:
849:
850: console "/usr/libexec/getty Pc" vt100 on secure
851: ttyE0 "/usr/libexec/getty Pc" vt220 off secure
852: ttyE1 "/usr/libexec/getty Pc" vt220 off secure
853: ttyE2 "/usr/libexec/getty Pc" vt220 off secure
854: ttyE3 "/usr/libexec/getty Pc" vt220 off secure
855:
856: Finally, all screens must be commented out from `/etc/wscons.conf`.
857:
858: It is also desirable to add
859:
860: powerd=YES
861:
862: in rc.conf. This way, the domain will be properly shut down if
863: `xm shutdown -R` or `xm shutdown -H` is used on the dom0.
864:
865: It is not strictly necessary to have a kernel (as /netbsd) in the domU
866: file system. However, various programs (e.g. netstat) will use that
867: kernel to look up symbols to read from kernel virtual memory. If
868: /netbsd is not the running kernel, those lookups will fail. (This is
869: not really a Xen-specific issue, but because the domU kernel is
870: obtained from the dom0, it is far more likely to be out of sync or
871: missing with Xen.)
872:
873: Creating an unprivileged Linux domain (domU)
874: --------------------------------------------
875:
876: Creating unprivileged Linux domains isn't much different from
877: unprivileged NetBSD domains, but there are some details to know.
878:
879: First, the second parameter passed to the disk declaration (the '0x1' in
880: the example below)
881:
882: disk = [ 'phy:/dev/wd0e,0x1,w' ]
883:
884: does matter to Linux. It wants a Linux device number here (e.g. 0x300
885: for hda). Linux builds device numbers as: (major \<\< 8 + minor).
886: So, hda1 which has major 3 and minor 1 on a Linux system will have
887: device number 0x301. Alternatively, devices names can be used (hda,
888: hdb, ...) as xentools has a table to map these names to devices
889: numbers. To export a partition to a Linux guest we can use:
890:
891: disk = [ 'phy:/dev/wd0e,0x300,w' ]
892: root = "/dev/hda1 ro"
893:
894: and it will appear as /dev/hda on the Linux system, and be used as root
895: partition.
896:
897: To install the Linux system on the partition to be exported to the
898: guest domain, the following method can be used: install
899: sysutils/e2fsprogs from pkgsrc. Use mke2fs to format the partition
900: that will be the root partition of your Linux domain, and mount it.
901: Then copy the files from a working Linux system, make adjustments in
902: `/etc` (fstab, network config). It should also be possible to extract
903: binary packages such as .rpm or .deb directly to the mounted partition
904: using the appropriate tool, possibly running under NetBSD's Linux
905: emulation. Once the file system has been populated, umount it. If
906: desirable, the file system can be converted to ext3 using tune2fs -j.
907: It should now be possible to boot the Linux guest domain, using one of
908: the vmlinuz-\*-xenU kernels available in the Xen binary distribution.
909:
910: To get the Linux console right, you need to add:
911:
912: extra = "xencons=tty1"
913:
914: to your configuration since not all Linux distributions auto-attach a
915: tty to the xen console.
916:
917: Creating an unprivileged Solaris domain (domU)
918: ----------------------------------------------
919:
920: See possibly outdated
921: [Solaris domU instructions](/ports/xen/howto-solaris/).
922:
923:
924: PCI passthrough: Using PCI devices in guest domains
925: ---------------------------------------------------
926:
927: The dom0 can give other domains access to selected PCI
928: devices. This can allow, for example, a non-privileged domain to have
929: access to a physical network interface or disk controller. However,
930: keep in mind that giving a domain access to a PCI device most likely
931: will give the domain read/write access to the whole physical memory,
932: as PCs don't have an IOMMU to restrict memory access to DMA-capable
933: device. Also, it's not possible to export ISA devices to non-dom0
934: domains, which means that the primary VGA adapter can't be exported.
935: A guest domain trying to access the VGA registers will panic.
936:
937: If the dom0 is NetBSD, it has to be running Xen 3.1, as support has
938: not been ported to later versions at this time.
939:
940: For a PCI device to be exported to a domU, is has to be attached to
941: the "pciback" driver in dom0. Devices passed to the dom0 via the
942: pciback.hide boot parameter will attach to "pciback" instead of the
943: usual driver. The list of devices is specified as "(bus:dev.func)",
944: where bus and dev are 2-digit hexadecimal numbers, and func a
945: single-digit number:
946:
947: pciback.hide=(00:0a.0)(00:06.0)
948:
949: pciback devices should show up in the dom0's boot messages, and the
950: devices should be listed in the `/kern/xen/pci` directory.
951:
952: PCI devices to be exported to a domU are listed in the "pci" array of
953: the domU's config file, with the format "0000:bus:dev.func".
954:
955: pci = [ '0000:00:06.0', '0000:00:0a.0' ]
956:
957: In the domU an "xpci" device will show up, to which one or more pci
958: buses will attach. Then the PCI drivers will attach to PCI buses as
959: usual. Note that the default NetBSD DOMU kernels do not have "xpci"
960: or any PCI drivers built in by default; you have to build your own
961: kernel to use PCI devices in a domU. Here's a kernel config example;
962: note that only the "xpci" lines are unusual.
963:
964: include "arch/i386/conf/XEN3_DOMU"
965:
966: # Add support for PCI buses to the XEN3_DOMU kernel
967: xpci* at xenbus ?
968: pci* at xpci ?
969:
970: # PCI USB controllers
971: uhci* at pci? dev ? function ? # Universal Host Controller (Intel)
972:
973: # USB bus support
974: usb* at uhci?
975:
976: # USB Hubs
977: uhub* at usb?
978: uhub* at uhub? port ? configuration ? interface ?
979:
980: # USB Mass Storage
981: umass* at uhub? port ? configuration ? interface ?
982: wd* at umass?
983: # SCSI controllers
984: ahc* at pci? dev ? function ? # Adaptec [23]94x, aic78x0 SCSI
985:
986: # SCSI bus support (for both ahc and umass)
987: scsibus* at scsi?
988:
989: # SCSI devices
990: sd* at scsibus? target ? lun ? # SCSI disk drives
991: cd* at scsibus? target ? lun ? # SCSI CD-ROM drives
992:
993:
994: NetBSD as a domU in a VPS
995: =========================
996:
997: The bulk of the HOWTO is about using NetBSD as a dom0 on your own
998: hardware. This section explains how to deal with Xen in a domU as a
999: virtual private server where you do not control or have access to the
1000: dom0. This is not intended to be an exhaustive list of VPS providers;
1001: only a few are mentioned that specifically support NetBSD.
1002:
1003: VPS operators provide varying degrees of access and mechanisms for
1004: configuration. The big issue is usually how one controls which kernel
1005: is booted, because the kernel is nominally in the dom0 file system (to
1006: which VPS users do not normally have access). A second issue is how
1007: to install NetBSD.
1008: A VPS user may want to compile a kernel for security updates, to run
1009: npf, run IPsec, or any other reason why someone would want to change
1010: their kernel.
1011:
1012: One approach is to have an administrative interface to upload a kernel,
1013: or to select from a prepopulated list. Other approaches are pygrub
1014: (deprecated) and pvgrub, which are ways to have a bootloader obtain a
1015: kernel from the domU file system. This is closer to a regular physical
1016: computer, where someone who controls a machine can replace the kernel.
1017:
1018: A second issue is multiple CPUs. With NetBSD 6, domUs support
1019: multiple vcpus, and it is typical for VPS providers to enable multiple
1020: CPUs for NetBSD domUs.
1021:
1022: pygrub
1023: -------
1024:
1025: pygrub runs in the dom0 and looks into the domU file system. This
1026: implies that the domU must have a kernel in a file system in a format
1027: known to pygrub. As of 2014, pygrub seems to be of mostly historical
1028: interest.
1029:
1030: pvgrub
1031: ------
1032:
1033: pvgrub is a version of grub that uses PV operations instead of BIOS
1034: calls. It is booted from the dom0 as the domU kernel, and then reads
1035: /grub/menu.lst and loads a kernel from the domU file system.
1036:
1037: [Panix](http://www.panix.com/) lets users use pvgrub. Panix reports
1038: that pvgrub works with FFsv2 with 16K/2K and 32K/4K block/frag sizes
1039: (and hence with defaults from "newfs -O 2"). See [Panix's pvgrub
1040: page](http://www.panix.com/v-colo/grub.html), which describes only
1041: Linux but should be updated to cover NetBSD :-).
1042:
1043: [prgmr.com](http://prgmr.com/) also lets users with pvgrub to boot
1044: their own kernel. See then [prgmr.com NetBSD
1045: HOWTO](http://wiki.prgmr.com/mediawiki/index.php/NetBSD_as_a_DomU)
1046: (which is in need of updating).
1047:
1048: It appears that [grub's FFS
1049: code](http://xenbits.xensource.com/hg/xen-unstable.hg/file/bca284f67702/tools/libfsimage/ufs/fsys_ufs.c)
1050: does not support all aspects of modern FFS, but there are also reports
1051: that FFSv2 works fine. At prgmr, typically one has an ext2 or FAT
1052: partition for the kernel with the intent that grub can understand it,
1053: which leads to /netbsd not being the actual kernel. One must remember
1054: to update the special boot partition.
1055:
1056: Amazon
1057: ------
1058:
1059: See the [Amazon EC2 page](../amazon_ec2/).
1060:
1061: Using npf
1062: ---------
1063:
1064: In standard kernels, npf is a module, and thus cannot be loaded in a
1065: DOMU kernel.
1066:
1067: TODO: Explain how to compile npf into a custom kernel, answering (but
1068: note that the problem was caused by not booting the right kernel)
1069: [this email to
1070: netbsd-users](http://mail-index.netbsd.org/netbsd-users/2014/12/26/msg015576.html).
1071:
1072: TODO items for improving NetBSD/xen
1073: ===================================
1074:
1075: * Make the NetBSD dom0 kernel work with SMP.
1076: * Test the Xen 4.5 packages adequately to be able to recommend them as
1077: the standard approach.
1078: * Get PCI passthrough working on Xen 4.5
1079: * Get pvgrub into pkgsrc, either via xentools or separately.
1080: * grub
1081: * Check/add support to pkgsrc grub2 for UFS2 and arbitrary
1082: fragsize/blocksize (UFS2 support may be present; the point is to
1083: make it so that with any UFS1/UFS2 file system setup that works
1084: with NetBSD grub will also work).
1085: See [pkg/40258](http://gnats.netbsd.org/40258).
1086: * Push patches upstream.
1087: * Get UFS2 patches into pvgrub.
1088: * Add support for PV ops to a version of /boot, and make it usable as
1089: a kernel in Xen, similar to pvgrub.
1090: * Solve somehow the issue with modules for GENERIC not being loadable
1091: in a Xen dom0 or domU kernel.
1092:
1093: Random pointers
1094: ===============
1095:
1096: TODO: This section contains links from elsewhere not yet integrated
1097: into the HOWTO.
1098:
1099: * http://www.lumbercartel.ca/library/xen/
1100: * http://pbraun.nethence.com/doc/sysutils/xen_netbsd_dom0.html
CVSweb for NetBSD wikisrc <wikimaster@NetBSD.org> software: FreeBSD-CVSweb
![[BACK]](/icons/back.gif){kind=icon}
Return to howto.mdwn CVS log ![[TXT]](/icons/text.gif){kind=icon}
![[DIR]](/icons/dir.gif){kind=icon}
Up to [NetBSD Developer Wiki] / wikisrc / ports / xen