]> git.proxmox.com Git - pve-docs.git/blame - qm.adoc
asciidoc-pve.in: add code to auto generate online help data
[pve-docs.git] / qm.adoc
CommitLineData
80c0adcb 1[[chapter_virtual_machines]]
f69cfd23 2ifdef::manvolnum[]
b2f242ab
DM
3qm(1)
4=====
38fd0958 5include::attributes.txt[]
5f09af76
DM
6:pve-toplevel:
7
f69cfd23
DM
8NAME
9----
10
11qm - Qemu/KVM Virtual Machine Manager
12
13
49a5e11c 14SYNOPSIS
f69cfd23
DM
15--------
16
17include::qm.1-synopsis.adoc[]
18
19DESCRIPTION
20-----------
21endif::manvolnum[]
f69cfd23
DM
22ifndef::manvolnum[]
23Qemu/KVM Virtual Machines
24=========================
38fd0958 25include::attributes.txt[]
f69cfd23 26endif::manvolnum[]
5f09af76
DM
27ifdef::wiki[]
28:pve-toplevel:
29endif::wiki[]
30
c4cba5d7
EK
31// deprecates
32// http://pve.proxmox.com/wiki/Container_and_Full_Virtualization
33// http://pve.proxmox.com/wiki/KVM
34// http://pve.proxmox.com/wiki/Qemu_Server
35
5eba0743 36Qemu (short form for Quick Emulator) is an open source hypervisor that emulates a
c4cba5d7
EK
37physical computer. From the perspective of the host system where Qemu is
38running, Qemu is a user program which has access to a number of local resources
39like partitions, files, network cards which are then passed to an
189d3661 40emulated computer which sees them as if they were real devices.
c4cba5d7
EK
41
42A guest operating system running in the emulated computer accesses these
43devices, and runs as it were running on real hardware. For instance you can pass
44an iso image as a parameter to Qemu, and the OS running in the emulated computer
189d3661 45will see a real CDROM inserted in a CD drive.
c4cba5d7 46
189d3661 47Qemu can emulates a great variety of hardware from ARM to Sparc, but {pve} is
c4cba5d7
EK
48only concerned with 32 and 64 bits PC clone emulation, since it represents the
49overwhelming majority of server hardware. The emulation of PC clones is also one
50of the fastest due to the availability of processor extensions which greatly
51speed up Qemu when the emulated architecture is the same as the host
9c63b5d9
EK
52architecture.
53
54NOTE: You may sometimes encounter the term _KVM_ (Kernel-based Virtual Machine).
55It means that Qemu is running with the support of the virtualization processor
56extensions, via the Linux kvm module. In the context of {pve} _Qemu_ and
57_KVM_ can be use interchangeably as Qemu in {pve} will always try to load the kvm
58module.
59
c4cba5d7
EK
60Qemu inside {pve} runs as a root process, since this is required to access block
61and PCI devices.
62
5eba0743 63
c4cba5d7
EK
64Emulated devices and paravirtualized devices
65--------------------------------------------
66
189d3661
DC
67The PC hardware emulated by Qemu includes a mainboard, network controllers,
68scsi, ide and sata controllers, serial ports (the complete list can be seen in
69the `kvm(1)` man page) all of them emulated in software. All these devices
70are the exact software equivalent of existing hardware devices, and if the OS
71running in the guest has the proper drivers it will use the devices as if it
c4cba5d7
EK
72were running on real hardware. This allows Qemu to runs _unmodified_ operating
73systems.
74
75This however has a performance cost, as running in software what was meant to
76run in hardware involves a lot of extra work for the host CPU. To mitigate this,
77Qemu can present to the guest operating system _paravirtualized devices_, where
78the guest OS recognizes it is running inside Qemu and cooperates with the
79hypervisor.
80
81Qemu relies on the virtio virtualization standard, and is thus able to presente
189d3661
DC
82paravirtualized virtio devices, which includes a paravirtualized generic disk
83controller, a paravirtualized network card, a paravirtualized serial port,
c4cba5d7
EK
84a paravirtualized SCSI controller, etc ...
85
189d3661
DC
86It is highly recommended to use the virtio devices whenever you can, as they
87provide a big performance improvement. Using the virtio generic disk controller
88versus an emulated IDE controller will double the sequential write throughput,
89as measured with `bonnie++(8)`. Using the virtio network interface can deliver
c4cba5d7 90up to three times the throughput of an emulated Intel E1000 network card, as
189d3661 91measured with `iperf(1)`. footnote:[See this benchmark on the KVM wiki
c4cba5d7
EK
92http://www.linux-kvm.org/page/Using_VirtIO_NIC]
93
5eba0743 94
80c0adcb 95[[qm_virtual_machines_settings]]
c4cba5d7
EK
96Virtual Machines settings
97-------------------------
80c0adcb 98
c4cba5d7
EK
99Generally speaking {pve} tries to choose sane defaults for virtual machines
100(VM). Make sure you understand the meaning of the settings you change, as it
101could incur a performance slowdown, or putting your data at risk.
102
5eba0743 103
80c0adcb 104[[qm_general_settings]]
c4cba5d7
EK
105General Settings
106~~~~~~~~~~~~~~~~
80c0adcb 107
c4cba5d7
EK
108General settings of a VM include
109
110* the *Node* : the physical server on which the VM will run
111* the *VM ID*: a unique number in this {pve} installation used to identify your VM
112* *Name*: a free form text string you can use to describe the VM
113* *Resource Pool*: a logical group of VMs
114
5eba0743 115
80c0adcb 116[[qm_os_settings]]
c4cba5d7
EK
117OS Settings
118~~~~~~~~~~~
80c0adcb 119
c4cba5d7
EK
120When creating a VM, setting the proper Operating System(OS) allows {pve} to
121optimize some low level parameters. For instance Windows OS expect the BIOS
122clock to use the local time, while Unix based OS expect the BIOS clock to have
123the UTC time.
124
5eba0743 125
80c0adcb 126[[qm_hard_disk]]
c4cba5d7
EK
127Hard Disk
128~~~~~~~~~
80c0adcb 129
2ec49380 130Qemu can emulate a number of storage controllers:
c4cba5d7
EK
131
132* the *IDE* controller, has a design which goes back to the 1984 PC/AT disk
133controller. Even if this controller has been superseded by more more designs,
134each and every OS you can think has support for it, making it a great choice
135if you want to run an OS released before 2003. You can connect up to 4 devices
136on this controller.
137
138* the *SATA* (Serial ATA) controller, dating from 2003, has a more modern
139design, allowing higher throughput and a greater number of devices to be
140connected. You can connect up to 6 devices on this controller.
141
b0b6802b
EK
142* the *SCSI* controller, designed in 1985, is commonly found on server grade
143hardware, and can connect up to 14 storage devices. {pve} emulates by default a
144LSI 53C895A controller. +
145A SCSI controller of type _Virtio_ is the recommended setting if you aim for
146performance and is automatically selected for newly created Linux VMs since
147{pve} 4.3. Linux distributions have support for this controller since 2012, and
c4cba5d7 148FreeBSD since 2014. For Windows OSes, you need to provide an extra iso
b0b6802b
EK
149containing the drivers during the installation.
150// https://pve.proxmox.com/wiki/Paravirtualized_Block_Drivers_for_Windows#During_windows_installation.
151
152* The *Virtio* controller, also called virtio-blk to distinguish from
153the Virtio SCSI controller, is an older type of paravirtualized controller
154which has been superseded in features by the Virtio SCSI Controller.
c4cba5d7
EK
155
156On each controller you attach a number of emulated hard disks, which are backed
157by a file or a block device residing in the configured storage. The choice of
158a storage type will determine the format of the hard disk image. Storages which
159present block devices (LVM, ZFS, Ceph) will require the *raw disk image format*,
160whereas files based storages (Ext4, NFS, GlusterFS) will let you to choose
161either the *raw disk image format* or the *QEMU image format*.
162
163 * the *QEMU image format* is a copy on write format which allows snapshots, and
164 thin provisioning of the disk image.
189d3661
DC
165 * the *raw disk image* is a bit-to-bit image of a hard disk, similar to what
166 you would get when executing the `dd` command on a block device in Linux. This
167 format do not support thin provisioning or snapshotting by itself, requiring
168 cooperation from the storage layer for these tasks. It is however 10% faster
169 than the *QEMU image format*. footnote:[See this benchmark for details
c4cba5d7 170 http://events.linuxfoundation.org/sites/events/files/slides/CloudOpen2013_Khoa_Huynh_v3.pdf]
189d3661 171 * the *VMware image format* only makes sense if you intend to import/export the
c4cba5d7
EK
172 disk image to other hypervisors.
173
174Setting the *Cache* mode of the hard drive will impact how the host system will
175notify the guest systems of block write completions. The *No cache* default
176means that the guest system will be notified that a write is complete when each
177block reaches the physical storage write queue, ignoring the host page cache.
178This provides a good balance between safety and speed.
179
180If you want the {pve} backup manager to skip a disk when doing a backup of a VM,
181you can set the *No backup* option on that disk.
182
183If your storage supports _thin provisioning_ (see the storage chapter in the
184{pve} guide), and your VM has a *SCSI* controller you can activate the *Discard*
185option on the hard disks connected to that controller. With *Discard* enabled,
186when the filesystem of a VM marks blocks as unused after removing files, the
187emulated SCSI controller will relay this information to the storage, which will
188then shrink the disk image accordingly.
189
af9c6de1 190.IO Thread
2b6e4b66
EK
191The option *IO Thread* can only be enabled when using a disk with the *VirtIO* controller,
192or with the *SCSI* controller, when the emulated controller type is *VirtIO SCSI*.
c564fc52
DC
193With this enabled, Qemu uses one thread per disk, instead of one thread for all,
194so it should increase performance when using multiple disks.
195Note that backups do not currently work with *IO Thread* enabled.
196
80c0adcb
DM
197
198[[qm_cpu]]
34e541c5
EK
199CPU
200~~~
80c0adcb 201
34e541c5
EK
202A *CPU socket* is a physical slot on a PC motherboard where you can plug a CPU.
203This CPU can then contain one or many *cores*, which are independent
204processing units. Whether you have a single CPU socket with 4 cores, or two CPU
205sockets with two cores is mostly irrelevant from a performance point of view.
206However some software is licensed depending on the number of sockets you have in
207your machine, in that case it makes sense to set the number of of sockets to
208what the license allows you, and increase the number of cores. +
209Increasing the number of virtual cpus (cores and sockets) will usually provide a
210performance improvement though that is heavily dependent on the use of the VM.
211Multithreaded applications will of course benefit from a large number of
212virtual cpus, as for each virtual cpu you add, Qemu will create a new thread of
213execution on the host system. If you're not sure about the workload of your VM,
214it is usually a safe bet to set the number of *Total cores* to 2.
215
216NOTE: It is perfectly safe to set the _overall_ number of total cores in all
217your VMs to be greater than the number of of cores you have on your server (ie.
2184 VMs with each 4 Total cores running in a 8 core machine is OK) In that case
219the host system will balance the Qemu execution threads between your server
220cores just like if you were running a standard multithreaded application.
221However {pve} will prevent you to allocate on a _single_ machine more vcpus than
222physically available, as this will only bring the performance down due to the
223cost of context switches.
224
225Qemu can emulate a number different of *CPU types* from 486 to the latest Xeon
226processors. Each new processor generation adds new features, like hardware
227assisted 3d rendering, random number generation, memory protection, etc ...
228Usually you should select for your VM a processor type which closely matches the
229CPU of the host system, as it means that the host CPU features (also called _CPU
230flags_ ) will be available in your VMs. If you want an exact match, you can set
231the CPU type to *host* in which case the VM will have exactly the same CPU flags
232as your host system. +
233This has a downside though. If you want to do a live migration of VMs between
234different hosts, your VM might end up on a new system with a different CPU type.
235If the CPU flags passed to the guest are missing, the qemu process will stop. To
236remedy this Qemu has also its own CPU type *kvm64*, that {pve} uses by defaults.
237kvm64 is a Pentium 4 look a like CPU type, which has a reduced CPU flags set,
238but is guaranteed to work everywhere. +
239 In short, if you care about live migration and moving VMs between nodes, leave
240the kvm64 default. If you don’t care about live migration, set the CPU type to
241host, as in theory this will give your guests maximum performance.
242
243You can also optionally emulate a *NUMA* architecture in your VMs. The basics of
244the NUMA architecture mean that instead of having a global memory pool available
245to all your cores, the memory is spread into local banks close to each socket.
246This can bring speed improvements as the memory bus is not a bottleneck
247anymore. If your system has a NUMA architecture footnote:[if the command
248`numactl --hardware | grep available` returns more than one node, then your host
249system has a NUMA architecture] we recommend to activate the option, as this
250will allow proper distribution of the VM resources on the host system. This
251option is also required in {pve} to allow hotplugging of cores and RAM to a VM.
252
253If the NUMA option is used, it is recommended to set the number of sockets to
254the number of sockets of the host system.
255
80c0adcb
DM
256
257[[qm_memory]]
34e541c5
EK
258Memory
259~~~~~~
80c0adcb 260
34e541c5
EK
261For each VM you have the option to set a fixed size memory or asking
262{pve} to dynamically allocate memory based on the current RAM usage of the
263host.
264
265When choosing a *fixed size memory* {pve} will simply allocate what you
266specify to your VM.
267
268// see autoballoon() in pvestatd.pm
269When choosing to *automatically allocate memory*, {pve} will make sure that the
270minimum amount you specified is always available to the VM, and if RAM usage on
271the host is below 80%, will dynamically add memory to the guest up to the
272maximum memory specified. +
273When the host is becoming short on RAM, the VM will then release some memory
274back to the host, swapping running processes if needed and starting the oom
275killer in last resort. The passing around of memory between host and guest is
276done via a special `balloon` kernel driver running inside the guest, which will
277grab or release memory pages from the host.
278footnote:[A good explanation of the inner workings of the balloon driver can be found here https://rwmj.wordpress.com/2010/07/17/virtio-balloon/]
279
c9f6e1a4
EK
280When multiple VMs use the autoallocate facility, it is possible to set a
281*Shares* coefficient which indicates the relative amount of the free host memory
282that each VM shoud take. Suppose for instance you have four VMs, three of them
283running a HTTP server and the last one is a database server. To cache more
284database blocks in the database server RAM, you would like to prioritize the
285database VM when spare RAM is available. For this you assign a Shares property
286of 3000 to the database VM, leaving the other VMs to the Shares default setting
287of 1000. The host server has 32GB of RAM, and is curring using 16GB, leaving 32
288* 80/100 - 16 = 9GB RAM to be allocated to the VMs. The database VM will get 9 *
2893000 / (3000 + 1000 + 1000 + 1000) = 4.5 GB extra RAM and each HTTP server will
290get 1/5 GB.
291
34e541c5
EK
292All Linux distributions released after 2010 have the balloon kernel driver
293included. For Windows OSes, the balloon driver needs to be added manually and can
294incur a slowdown of the guest, so we don't recommend using it on critical
295systems.
296// see https://forum.proxmox.com/threads/solved-hyper-threading-vs-no-hyper-threading-fixed-vs-variable-memory.20265/
297
298When allocating RAMs to your VMs, a good rule of thumb is always to leave 1GB
299of RAM available to the host.
300
80c0adcb
DM
301
302[[qm_network_device]]
1ff7835b
EK
303Network Device
304~~~~~~~~~~~~~~
80c0adcb 305
1ff7835b
EK
306Each VM can have many _Network interface controllers_ (NIC), of four different
307types:
308
309 * *Intel E1000* is the default, and emulates an Intel Gigabit network card.
310 * the *VirtIO* paravirtualized NIC should be used if you aim for maximum
311performance. Like all VirtIO devices, the guest OS should have the proper driver
312installed.
313 * the *Realtek 8139* emulates an older 100 MB/s network card, and should
314only be used when emulating older operating systems ( released before 2002 )
315 * the *vmxnet3* is another paravirtualized device, which should only be used
316when importing a VM from another hypervisor.
317
318{pve} will generate for each NIC a random *MAC address*, so that your VM is
319addressable on Ethernet networks.
320
af9c6de1
EK
321The NIC you added to the VM can follow one of two differents models:
322
323 * in the default *Bridged mode* each virtual NIC is backed on the host by a
324_tap device_, ( a software loopback device simulating an Ethernet NIC ). This
325tap device is added to a bridge, by default vmbr0 in {pve}. In this mode, VMs
326have direct access to the Ethernet LAN on which the host is located.
327 * in the alternative *NAT mode*, each virtual NIC will only communicate with
328the Qemu user networking stack, where a builting router and DHCP server can
329provide network access. This built-in DHCP will serve adresses in the private
33010.0.2.0/24 range. The NAT mode is much slower than the bridged mode, and
331should only be used for testing.
332
333You can also skip adding a network device when creating a VM by selecting *No
334network device*.
335
336.Multiqueue
1ff7835b 337If you are using the VirtIO driver, you can optionally activate the
af9c6de1 338*Multiqueue* option. This option allows the guest OS to process networking
1ff7835b
EK
339packets using multiple virtual CPUs, providing an increase in the total number
340of packets transfered.
341
342//http://blog.vmsplice.net/2011/09/qemu-internals-vhost-architecture.html
343When using the VirtIO driver with {pve}, each NIC network queue is passed to the
344host kernel, where the queue will be processed by a kernel thread spawn by the
345vhost driver. With this option activated, it is possible to pass _multiple_
346network queues to the host kernel for each NIC.
347
348//https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Virtualization_Tuning_and_Optimization_Guide/sect-Virtualization_Tuning_Optimization_Guide-Networking-Techniques.html#sect-Virtualization_Tuning_Optimization_Guide-Networking-Multi-queue_virtio-net
af9c6de1 349When using Multiqueue, it is recommended to set it to a value equal
1ff7835b
EK
350to the number of Total Cores of your guest. You also need to set in
351the VM the number of multi-purpose channels on each VirtIO NIC with the ethtool
352command:
353
354`ethtool -L eth0 combined X`
355
356where X is the number of the number of vcpus of the VM.
357
af9c6de1 358You should note that setting the Multiqueue parameter to a value greater
1ff7835b
EK
359than one will increase the CPU load on the host and guest systems as the
360traffic increases. We recommend to set this option only when the VM has to
361process a great number of incoming connections, such as when the VM is running
362as a router, reverse proxy or a busy HTTP server doing long polling.
363
80c0adcb 364
685cc8e0
DC
365USB Passthrough
366~~~~~~~~~~~~~~~
80c0adcb 367
685cc8e0
DC
368There are two different types of USB passthrough devices:
369
370* Host USB passtrough
371* SPICE USB passthrough
372
373Host USB passthrough works by giving a VM a USB device of the host.
374This can either be done via the vendor- and product-id, or
375via the host bus and port.
376
377The vendor/product-id looks like this: *0123:abcd*,
378where *0123* is the id of the vendor, and *abcd* is the id
379of the product, meaning two pieces of the same usb device
380have the same id.
381
382The bus/port looks like this: *1-2.3.4*, where *1* is the bus
383and *2.3.4* is the port path. This represents the physical
384ports of your host (depending of the internal order of the
385usb controllers).
386
387If a device is present in a VM configuration when the VM starts up,
388but the device is not present in the host, the VM can boot without problems.
389As soon as the device/port ist available in the host, it gets passed through.
390
391WARNING: Using this kind of USB passthrough, means that you cannot move
392a VM online to another host, since the hardware is only available
393on the host the VM is currently residing.
394
395The second type of passthrough is SPICE USB passthrough. This is useful
396if you use a SPICE client which supports it. If you add a SPICE USB port
397to your VM, you can passthrough a USB device from where your SPICE client is,
398directly to the VM (for example an input device or hardware dongle).
399
80c0adcb
DM
400
401[[qm_bios_and_uefi]]
076d60ae
DC
402BIOS and UEFI
403~~~~~~~~~~~~~
404
405In order to properly emulate a computer, QEMU needs to use a firmware.
406By default QEMU uses *SeaBIOS* for this, which is an open-source, x86 BIOS
407implementation. SeaBIOS is a good choice for most standard setups.
408
409There are, however, some scenarios in which a BIOS is not a good firmware
410to boot from, e.g. if you want to do VGA passthrough. footnote:[Alex Williamson has a very good blog entry about this.
411http://vfio.blogspot.co.at/2014/08/primary-graphics-assignment-without-vga.html]
412In such cases, you should rather use *OVMF*, which is an open-source UEFI implemenation. footnote:[See the OVMF Project http://www.tianocore.org/ovmf/]
413
414If you want to use OVMF, there are several things to consider:
415
416In order to save things like the *boot order*, there needs to be an EFI Disk.
417This disk will be included in backups and snapshots, and there can only be one.
418
419You can create such a disk with the following command:
420
421 qm set <vmid> -efidisk0 <storage>:1,format=<format>
422
423Where *<storage>* is the storage where you want to have the disk, and
424*<format>* is a format which the storage supports. Alternatively, you can
425create such a disk through the web interface with 'Add' -> 'EFI Disk' in the
426hardware section of a VM.
427
428When using OVMF with a virtual display (without VGA passthrough),
429you need to set the client resolution in the OVMF menu(which you can reach
430with a press of the ESC button during boot), or you have to choose
431SPICE as the display type.
432
433
8c1189b6 434Managing Virtual Machines with `qm`
dd042288 435------------------------------------
f69cfd23 436
dd042288 437qm is the tool to manage Qemu/Kvm virtual machines on {pve}. You can
f69cfd23
DM
438create and destroy virtual machines, and control execution
439(start/stop/suspend/resume). Besides that, you can use qm to set
440parameters in the associated config file. It is also possible to
441create and delete virtual disks.
442
dd042288
EK
443CLI Usage Examples
444~~~~~~~~~~~~~~~~~~
445
446Create a new VM with 4 GB IDE disk.
447
448 qm create 300 -ide0 4 -net0 e1000 -cdrom proxmox-mailgateway_2.1.iso
449
450Start the new VM
451
452 qm start 300
453
454Send a shutdown request, then wait until the VM is stopped.
455
456 qm shutdown 300 && qm wait 300
457
458Same as above, but only wait for 40 seconds.
459
460 qm shutdown 300 && qm wait 300 -timeout 40
461
f69cfd23
DM
462Configuration
463-------------
464
465All configuration files consists of lines in the form
466
467 PARAMETER: value
468
871e1fd6 469Configuration files are stored inside the Proxmox cluster file
8c1189b6 470system, and can be accessed at `/etc/pve/qemu-server/<VMID>.conf`.
f69cfd23 471
80c0adcb 472[[qm_options]]
a7f36905
DM
473Options
474~~~~~~~
475
476include::qm.conf.5-opts.adoc[]
477
f69cfd23
DM
478
479Locks
480-----
481
8c1189b6 482Online migrations and backups (`vzdump`) set a lock to prevent incompatible
871e1fd6
FG
483concurrent actions on the affected VMs. Sometimes you need to remove such a
484lock manually (e.g., after a power failure).
f69cfd23
DM
485
486 qm unlock <vmid>
487
f69cfd23
DM
488
489ifdef::manvolnum[]
490include::pve-copyright.adoc[]
491endif::manvolnum[]