libvirt/docs/formatbackup.html.in

192 lines
9.1 KiB
HTML
Raw Normal View History

backup: Document new XML for backups Prepare for new backup APIs by describing the XML that will represent a backup. The XML resembles snapshots and checkpoints in being able to select actions for a set of disks, but has other differences. It can support both push model (the hypervisor does the backup directly into the destination file) and pull model (the hypervisor exposes an access port for a third party to grab what is necessary). Add testsuite coverage for some minimal uses of the XML. The <disk> element within <domainbackup> tries to model the same elements as a <disk> under <domain>, but sharing the RNG grammar proved to be hairy. That is in part because while <domain> use <source> to describe a host resource in use by the guest, a backup job is using a host resource that is not visible to the guest: a push backup action is instead describing a <target> (which ultimately could be a remote network resource, but for simplicity the RNG just validates a local file for now), and a pull backup action is instead describing a temporary local file <scratch> (which probably should not be a remote resource). A future refactoring may thus introduce some way to parameterize RNG to accept <disk type='FOO'>...</disk> so that the name of the subelement can be <source> for domain, or <target> or <scratch> as needed for backups. Future patches may improve this area of code. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
2019-08-22 01:42:41 +00:00
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml">
<body>
<h1>Backup XML format</h1>
<ul id="toc"></ul>
<h2><a id="BackupAttributes">Backup XML</a></h2>
<p>
Creating a backup, whether full or incremental, is done
via <code>virDomainBackupBegin()</code>, which takes an XML
description of the actions to perform, as well as an optional
second XML document <a href="formatcheckpoint.html">describing a
checkpoint</a> to create at the same point in time. See
also <a href="kbase/domainstatecapture.html">a comparison</a> between
backup: Document new XML for backups Prepare for new backup APIs by describing the XML that will represent a backup. The XML resembles snapshots and checkpoints in being able to select actions for a set of disks, but has other differences. It can support both push model (the hypervisor does the backup directly into the destination file) and pull model (the hypervisor exposes an access port for a third party to grab what is necessary). Add testsuite coverage for some minimal uses of the XML. The <disk> element within <domainbackup> tries to model the same elements as a <disk> under <domain>, but sharing the RNG grammar proved to be hairy. That is in part because while <domain> use <source> to describe a host resource in use by the guest, a backup job is using a host resource that is not visible to the guest: a push backup action is instead describing a <target> (which ultimately could be a remote network resource, but for simplicity the RNG just validates a local file for now), and a pull backup action is instead describing a temporary local file <scratch> (which probably should not be a remote resource). A future refactoring may thus introduce some way to parameterize RNG to accept <disk type='FOO'>...</disk> so that the name of the subelement can be <source> for domain, or <target> or <scratch> as needed for backups. Future patches may improve this area of code. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
2019-08-22 01:42:41 +00:00
the various state capture APIs.
</p>
<p>
There are two general modes for backups: a push mode (where the
hypervisor writes out the data to the destination file, which
may be local or remote), and a pull mode (where the hypervisor
creates an NBD server that a third-party client can then read as
needed, and which requires the use of temporary storage,
typically local, until the backup is complete).
</p>
<p>
The instructions for beginning a backup job are provided as
attributes and elements of the
top-level <code>domainbackup</code> element. This element
includes an optional attribute <code>mode</code> which can be
either "push" or "pull" (default
push). <code>virDomainBackupGetXMLDesc()</code> can be used to
see the actual values selected for elements omitted during
creation (for example, learning which port the NBD server is
using in the pull model or what file names libvirt generated
when none were supplied). The following child elements and attributes
are supported:
</p>
<dl>
<dt><code>incremental</code></dt>
<dd>An optional element giving the name of an existing
checkpoint of the domain, which will be used to make this
backup an incremental one. In the push model, only changes
since the named checkpoint are written to the destination. In
the pull model, the NBD server uses the
NBD_OPT_SET_META_CONTEXT extension to advertise to the client
which portions of the export contain changes since the named
checkpoint. If omitted, a full backup is performed.
</dd>
<dt><code>server</code></dt>
<dd>Present only for a pull mode backup. Contains the same
attributes as
the <a href="formatdomain.html#elementsDisks"><code>protocol</code>
element of a disk</a> attached via NBD in the domain (such as
transport, socket, name, port, or tls), necessary to set up an
NBD server that exposes the content of each disk at the time
the backup is started.
</dd>
<dt><code>disks</code></dt>
<dd>An optional listing of instructions for disks participating
in the backup (if omitted, all disks participate and libvirt
attempts to generate filenames by appending the current
timestamp as a suffix). If the entire element was omitted on
input, then all disks participate in the backup, otherwise,
only the disks explicitly listed which do not also
use <code>backup='no'</code> will participate. On output, this
is the state of each of the domain's disk in relation to the
backup operation.
<dl>
<dt><code>disk</code></dt>
<dd>This sub-element describes the backup properties of a
specific disk, with the following attributes and child
elements:
<dl>
<dt><code>name</code></dt>
<dd>A mandatory attribute which must match
the <code>&lt;target dev='name'/&gt;</code>
of one of
the <a href="formatdomain.html#elementsDisks">disk
devices</a> specified for the domain at the time of
the checkpoint.</dd>
<dt><code>backup</code></dt>
<dd>Setting this attribute to <code>yes</code>(default) specifies
that the disk should take part in the backup and using
<code>no</code> excludes the disk from the backup.</dd>
<dt><code>exportname</code></dt>
<dd>Allows modification of the NBD export name for the given disk.
By default equal to disk target.
Valid only for pull mode backups.</dd>
<dt><code>exportbitmap</code></dt>
<dd>Allows modification of the name of the bitmap describing dirty
blocks for an incremental backup exported via NBD export name
for the given disk.
Valid only for pull mode backups.</dd>
backup: Document new XML for backups Prepare for new backup APIs by describing the XML that will represent a backup. The XML resembles snapshots and checkpoints in being able to select actions for a set of disks, but has other differences. It can support both push model (the hypervisor does the backup directly into the destination file) and pull model (the hypervisor exposes an access port for a third party to grab what is necessary). Add testsuite coverage for some minimal uses of the XML. The <disk> element within <domainbackup> tries to model the same elements as a <disk> under <domain>, but sharing the RNG grammar proved to be hairy. That is in part because while <domain> use <source> to describe a host resource in use by the guest, a backup job is using a host resource that is not visible to the guest: a push backup action is instead describing a <target> (which ultimately could be a remote network resource, but for simplicity the RNG just validates a local file for now), and a pull backup action is instead describing a temporary local file <scratch> (which probably should not be a remote resource). A future refactoring may thus introduce some way to parameterize RNG to accept <disk type='FOO'>...</disk> so that the name of the subelement can be <source> for domain, or <target> or <scratch> as needed for backups. Future patches may improve this area of code. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
2019-08-22 01:42:41 +00:00
<dt><code>type</code></dt>
<dd>A mandatory attribute to describe the type of the
disk, except when <code>backup='no'</code> is
used. Valid values include <code>file</code>, or
<code>block</code>.
backup: Document new XML for backups Prepare for new backup APIs by describing the XML that will represent a backup. The XML resembles snapshots and checkpoints in being able to select actions for a set of disks, but has other differences. It can support both push model (the hypervisor does the backup directly into the destination file) and pull model (the hypervisor exposes an access port for a third party to grab what is necessary). Add testsuite coverage for some minimal uses of the XML. The <disk> element within <domainbackup> tries to model the same elements as a <disk> under <domain>, but sharing the RNG grammar proved to be hairy. That is in part because while <domain> use <source> to describe a host resource in use by the guest, a backup job is using a host resource that is not visible to the guest: a push backup action is instead describing a <target> (which ultimately could be a remote network resource, but for simplicity the RNG just validates a local file for now), and a pull backup action is instead describing a temporary local file <scratch> (which probably should not be a remote resource). A future refactoring may thus introduce some way to parameterize RNG to accept <disk type='FOO'>...</disk> so that the name of the subelement can be <source> for domain, or <target> or <scratch> as needed for backups. Future patches may improve this area of code. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
2019-08-22 01:42:41 +00:00
Similar to a disk declaration for a domain, the choice of type
controls what additional sub-elements are needed to describe
the destination.</dd>
backup: Document new XML for backups Prepare for new backup APIs by describing the XML that will represent a backup. The XML resembles snapshots and checkpoints in being able to select actions for a set of disks, but has other differences. It can support both push model (the hypervisor does the backup directly into the destination file) and pull model (the hypervisor exposes an access port for a third party to grab what is necessary). Add testsuite coverage for some minimal uses of the XML. The <disk> element within <domainbackup> tries to model the same elements as a <disk> under <domain>, but sharing the RNG grammar proved to be hairy. That is in part because while <domain> use <source> to describe a host resource in use by the guest, a backup job is using a host resource that is not visible to the guest: a push backup action is instead describing a <target> (which ultimately could be a remote network resource, but for simplicity the RNG just validates a local file for now), and a pull backup action is instead describing a temporary local file <scratch> (which probably should not be a remote resource). A future refactoring may thus introduce some way to parameterize RNG to accept <disk type='FOO'>...</disk> so that the name of the subelement can be <source> for domain, or <target> or <scratch> as needed for backups. Future patches may improve this area of code. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
2019-08-22 01:42:41 +00:00
<dt><code>target</code></dt>
<dd>Valid only for push mode backups, this is the
primary sub-element that describes the file name of
the backup destination, similar to
the <code>source</code> sub-element of a domain
disk. An optional sub-element <code>driver</code> can
also be used, with an attribute <code>type</code> to
specify a destination format different from
qcow2. See documentation for <code>scratch</code> below for
additional configuration.</dd>
backup: Document new XML for backups Prepare for new backup APIs by describing the XML that will represent a backup. The XML resembles snapshots and checkpoints in being able to select actions for a set of disks, but has other differences. It can support both push model (the hypervisor does the backup directly into the destination file) and pull model (the hypervisor exposes an access port for a third party to grab what is necessary). Add testsuite coverage for some minimal uses of the XML. The <disk> element within <domainbackup> tries to model the same elements as a <disk> under <domain>, but sharing the RNG grammar proved to be hairy. That is in part because while <domain> use <source> to describe a host resource in use by the guest, a backup job is using a host resource that is not visible to the guest: a push backup action is instead describing a <target> (which ultimately could be a remote network resource, but for simplicity the RNG just validates a local file for now), and a pull backup action is instead describing a temporary local file <scratch> (which probably should not be a remote resource). A future refactoring may thus introduce some way to parameterize RNG to accept <disk type='FOO'>...</disk> so that the name of the subelement can be <source> for domain, or <target> or <scratch> as needed for backups. Future patches may improve this area of code. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
2019-08-22 01:42:41 +00:00
<dt><code>scratch</code></dt>
<dd>Valid only for pull mode backups, this is the
primary sub-element that describes the file name of
the local scratch file to be used in facilitating the
backup, and is similar to the <code>source</code>
sub-element of a domain disk. Currently only <code>file</code>
and <code>block</code> scratch storage is supported. The
<code>file</code> scratch file is created and deleted by
libvirt in the given location. A <code>block</code> scratch
device must exist prior to starting the backup and is formatted.
The block device must have enough space for the corresponding
disk data including format overhead.
If <code>VIR_DOMAIN_BACKUP_BEGIN_REUSE_EXTERNAL</code> flag is
used the file for a scratch of <code>file</code> type must
exist with the correct format and size to hold the copy and is
used without modification. The file is not deleted after the
backup but the contents of the file don't make sense outside
of the backup. The same applies for the block device which
must be formatted appropriately.
Similarly to the domain
<a href="formatdomain.html#elementsDisks"><code>disk</code></a>
definition <code>scratch</code> and <code>target</code> can
contain <code>seclabel</code> and/or <code>encryption</code>
subelements to configure the corresponding properties.
</dd>
backup: Document new XML for backups Prepare for new backup APIs by describing the XML that will represent a backup. The XML resembles snapshots and checkpoints in being able to select actions for a set of disks, but has other differences. It can support both push model (the hypervisor does the backup directly into the destination file) and pull model (the hypervisor exposes an access port for a third party to grab what is necessary). Add testsuite coverage for some minimal uses of the XML. The <disk> element within <domainbackup> tries to model the same elements as a <disk> under <domain>, but sharing the RNG grammar proved to be hairy. That is in part because while <domain> use <source> to describe a host resource in use by the guest, a backup job is using a host resource that is not visible to the guest: a push backup action is instead describing a <target> (which ultimately could be a remote network resource, but for simplicity the RNG just validates a local file for now), and a pull backup action is instead describing a temporary local file <scratch> (which probably should not be a remote resource). A future refactoring may thus introduce some way to parameterize RNG to accept <disk type='FOO'>...</disk> so that the name of the subelement can be <source> for domain, or <target> or <scratch> as needed for backups. Future patches may improve this area of code. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
2019-08-22 01:42:41 +00:00
</dl>
</dd>
</dl>
</dd>
</dl>
<h2><a id="example">Examples</a></h2>
<p>Use <code>virDomainBackupBegin()</code> to perform a full
backup using push mode. The example lets libvirt pick the
destination and format for 'vda', fully specifies that we want a
raw backup of 'vdb', and omits 'vdc' from the operation.
</p>
<pre>
&lt;domainbackup&gt;
&lt;disks&gt;
&lt;disk name='vda' backup='yes'/&gt;
&lt;disk name='vdb' type='file'&gt;
&lt;target file='/path/to/vdb.backup'/&gt;
&lt;driver type='raw'/&gt;
&lt;/disk&gt;
&lt;disk name='vdc' backup='no'/&gt;
&lt;/disks&gt;
&lt;/domainbackup&gt;
</pre>
<p>If the previous full backup also passed a parameter describing
<a href="formatcheckpoint.html">checkpoint XML</a> that resulted
in a checkpoint named <code>1525889631</code>, we can make
another call to <code>virDomainBackupBegin()</code> to perform
an incremental backup of just the data changed since that
checkpoint, this time using the following XML to start a pull
model export of the 'vda' and 'vdb' disks, where a third-party
NBD client connecting to '/path/to/server' completes the backup
(omitting 'vdc' from the explicit list has the same effect as
the backup='no' from the previous example):
</p>
<pre>
&lt;domainbackup mode="pull"&gt;
&lt;incremental&gt;1525889631&lt;/incremental&gt;
&lt;server transport="unix" socket="/path/to/server"/&gt;
&lt;disks&gt;
&lt;disk name='vda' backup='yes' type='file'&gt;
&lt;scratch file='/path/to/file1.scratch'/&gt;
&lt;/disk&gt;
&lt;/disks&gt;
&lt;/domainbackup&gt;
</pre>
</body>
</html>