1
0
mirror of https://gitlab.com/libvirt/libvirt.git synced 2024-12-22 17:34:18 +03:00
libvirt/docs/hooks.html.in
Daniel P. Berrangé 0ea50f0148 docs: fix misc spelling errors reported by codespell
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
2020-10-05 10:28:37 +01:00

407 lines
22 KiB
XML

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml">
<body>
<h1>Hooks for specific system management</h1>
<ul id="toc"></ul>
<h2><a id="intro">Custom event scripts</a></h2>
<p>Beginning with libvirt 0.8.0, specific events on a host system will
trigger custom scripts.</p>
<p>These custom <b>hook</b> scripts are executed when any of the following
actions occur:</p>
<ul>
<li>The libvirt daemon starts, stops, or reloads its
configuration
(<span class="since">since 0.8.0</span>)<br/><br/></li>
<li>A QEMU guest is started or stopped
(<span class="since">since 0.8.0</span>)<br/><br/></li>
<li>An LXC guest is started or stopped
(<span class="since">since 0.8.0</span>)<br/><br/></li>
<li>A libxl-handled Xen guest is started or stopped
(<span class="since">since 2.1.0</span>)<br/><br/></li>
<li>A network is started or stopped or an interface is
plugged/unplugged to/from the network
(<span class="since">since 1.2.2</span>)<br/><br/></li>
</ul>
<h2><a id="location">Script location</a></h2>
<p>The libvirt hook scripts are located in the directory
<code>$SYSCONFDIR/libvirt/hooks/</code>.</p>
<ul>
<li>In Linux distributions such as Fedora and RHEL, this is
<code>/etc/libvirt/hooks/</code>. Other Linux distributions may do
this differently.</li>
<li>If your installation of libvirt has instead been compiled from
source, it is likely to be
<code>/usr/local/etc/libvirt/hooks/</code>.</li>
<li><span class="since">Since 6.5.0</span>, you can also place several
hook scripts in the directories
<code>/etc/libvirt/hooks/&lt;driver&gt;.d/</code>.</li>
</ul>
<p>To use hook scripts, you will need to create this <code>hooks</code>
directory manually, place the desired hook scripts inside, then make
them executable.</p>
<br/>
<h2><a id="names">Script names</a></h2>
<p>At present, there are five hook scripts that can be called:</p>
<ul>
<li><code>/etc/libvirt/hooks/daemon</code><br/><br/>
Executed when the libvirt daemon is started, stopped, or reloads
its configuration<br/><br/></li>
<li><code>/etc/libvirt/hooks/qemu</code><br/><br/>
Executed when a QEMU guest is started, stopped, or migrated<br/><br/></li>
<li><code>/etc/libvirt/hooks/lxc</code><br /><br/>
Executed when an LXC guest is started or stopped</li>
<li><code>/etc/libvirt/hooks/libxl</code><br/><br/>
Executed when a libxl-handled Xen guest is started, stopped, or
migrated<br/><br/></li>
<li><code>/etc/libvirt/hooks/network</code><br/><br/>
Executed when a network is started or stopped or an
interface is plugged/unplugged to/from the network</li>
</ul>
<p><span class="since">Since 6.5.0</span>, you can also have
several scripts with any name in the directories
<code>/etc/libvirt/hooks/&lt;driver&gt;.d/</code>. They are
executed in alphabetical order after main script.</p>
<br/>
<h2><a id="structure">Script structure</a></h2>
<p>The hook scripts are executed using standard Linux process creation
functions. Therefore, they must begin with the declaration of the
command interpreter to use.</p>
<p>For example:</p>
<pre>#!/bin/bash</pre>
<p>or:</p>
<pre>#!/usr/bin/python</pre>
<p>Other command interpreters are equally valid, as is any executable
binary, so you are welcome to use your favourite languages.</p>
<br/>
<h2><a id="arguments">Script arguments</a></h2>
<p>The hook scripts are called with specific command line arguments,
depending upon the script, and the operation being performed.</p>
<p>The guest hook scripts, qemu and lxc, are also given the <b>full</b>
XML description for the domain on their stdin. This includes items
such the UUID of the domain and its storage information, and is
intended to provide all the libvirt information the script needs.</p>
<p>For all cases, stdin of the network hook script is provided with the
full XML description of the network status in the following form:</p>
<pre>&lt;hookData&gt;
&lt;network&gt;
&lt;name&gt;$network_name&lt;/name&gt;
&lt;uuid&gt;afca425a-2c3a-420c-b2fb-dd7b4950d722&lt;/uuid&gt;
...
&lt;/network&gt;
&lt;/hookData&gt;</pre>
<p>In the case of an network port being created / deleted, the network
XML will be followed with the full XML description of the port:</p>
<pre>&lt;hookData&gt;
&lt;network&gt;
&lt;name&gt;$network_name&lt;/name&gt;
&lt;uuid&gt;afca425a-2c3a-420c-b2fb-dd7b4950d722&lt;/uuid&gt;
...
&lt;/network&gt;
&lt;networkport&gt;
&lt;uuid&gt;5d744f21-ba4a-4d6e-bdb2-30a35ff3207d&lt;/uuid&gt;
...
&lt;plug type='direct' dev='ens3' mode='vepa'/&gt;
&lt;/networkport&gt;
&lt;/hookData&gt;</pre>
<p>Please note that this approach is different from other cases such as
<code>daemon</code>, <code>qemu</code> or <code>lxc</code> hook scripts,
because two XMLs may be passed here, while in the other cases only a single
XML is passed.</p>
<p>The command line arguments take this approach:</p>
<ol>
<li>The first argument is the name of the <b>object</b> involved in the
operation, or '-' if there is none.<br/><br/>
For example, the name of a guest being started.<br/><br/></li>
<li>The second argument is the name of the <b>operation</b> being
performed.<br/><br/>
For example, "start" if a guest is being started.<br/><br/></li>
<li>The third argument is a <b>sub-operation</b> indication, or '-' if there
is none.<br/><br/></li>
<li>The last argument is an <b>extra argument</b> string, or '-' if there is
none.</li>
</ol>
<h4><a id="arguments_specifics">Specifics</a></h4>
<p>This translates to the following specifics for each hook script:</p>
<h5><a id="daemon">/etc/libvirt/hooks/daemon</a></h5>
<ul>
<li>When the libvirt daemon is started, this script is called as:<br/>
<pre>/etc/libvirt/hooks/daemon - start - start</pre></li>
<li>When the libvirt daemon is shut down, this script is called as:<br/>
<pre>/etc/libvirt/hooks/daemon - shutdown - shutdown</pre></li>
<li>When the libvirt daemon receives the SIGHUP signal, it reloads its
configuration and triggers the hook script as:<br/>
<pre>/etc/libvirt/hooks/daemon - reload begin SIGHUP</pre></li>
</ul>
<p>Please note that when the libvirt daemon is restarted, the <i>daemon</i>
hook script is called once with the "shutdown" operation, and then once
with the "start" operation. There is no specific operation to indicate
a "restart" is occurring.</p>
<h5><a id="qemu">/etc/libvirt/hooks/qemu</a></h5>
<ul>
<li>Before a QEMU guest is started, the qemu hook script is
called in three locations; if any location fails, the guest
is not started. The first location, <span class="since">since
0.9.0</span>, is before libvirt performs any resource
labeling, and the hook can allocate resources not managed by
libvirt such as DRBD or missing bridges. This is called as:<br/>
<pre>/etc/libvirt/hooks/qemu guest_name prepare begin -</pre>
The second location, available <span class="since">Since
0.8.0</span>, occurs after libvirt has finished labeling
all resources, but has not yet started the guest, called as:<br/>
<pre>/etc/libvirt/hooks/qemu guest_name start begin -</pre>
The third location, <span class="since">0.9.13</span>,
occurs after the QEMU process has successfully started up:<br/>
<pre>/etc/libvirt/hooks/qemu guest_name started begin -</pre>
</li>
<li>When a QEMU guest is stopped, the qemu hook script is called
in two locations, to match the startup.
First, <span class="since">since 0.8.0</span>, the hook is
called before libvirt restores any labels:<br/>
<pre>/etc/libvirt/hooks/qemu guest_name stopped end -</pre>
Then, after libvirt has released all resources, the hook is
called again, <span class="since">since 0.9.0</span>, to allow
any additional resource cleanup:<br/>
<pre>/etc/libvirt/hooks/qemu guest_name release end -</pre></li>
<li><span class="since">Since 0.9.11</span>, the qemu hook script
is also called at the beginning of incoming migration. It is called
as: <pre>/etc/libvirt/hooks/qemu guest_name migrate begin -</pre>
with domain XML sent to standard input of the script. In this case,
the script acts as a filter and is supposed to modify the domain
XML and print it out on its standard output. Empty output is
identical to copying the input XML without changing it. In case the
script returns failure or the output XML is not valid, incoming
migration will be canceled. This hook may be used, e.g., to change
location of disk images for incoming domains.</li>
<li><span class="since">Since 1.2.9</span>, the qemu hook script is
also called when restoring a saved image either via the API or
automatically when restoring a managed save machine. It is called
as: <pre>/etc/libvirt/hooks/qemu guest_name restore begin -</pre>
with domain XML sent to standard input of the script. In this case,
the script acts as a filter and is supposed to modify the domain
XML and print it out on its standard output. Empty output is
identical to copying the input XML without changing it. In case the
script returns failure or the output XML is not valid, restore of the
image will be aborted. This hook may be used, e.g., to change
location of disk images for restored domains.</li>
<li><span class="since">Since 6.5.0</span>, you can also place several
hook scripts in the directory
<code>/etc/libvirt/hooks/qemu.d/</code>. They are executed in
alphabetical order after main script. In this case each script also
acts as filter and can modify the domain XML and print it out on
its standard output. This script output is passed to standard input
next script in order. Empty output from any script is also identical
to copying the input XML without changing it.
In case any script returns failure common process will be aborted,
but all scripts from the directory will are executed.</li>
<li><span class="since">Since 0.9.13</span>, the qemu hook script
is also called when the libvirtd daemon restarts and reconnects
to previously running QEMU processes. If the script fails, the
existing QEMU process will be killed off. It is called as:
<pre>/etc/libvirt/hooks/qemu guest_name reconnect begin -</pre>
</li>
<li><span class="since">Since 0.9.13</span>, the qemu hook script
is also called when the QEMU driver is told to attach to an
externally launched QEMU process. It is called as:
<pre>/etc/libvirt/hooks/qemu guest_name attach begin -</pre>
</li>
</ul>
<h5><a id="lxc">/etc/libvirt/hooks/lxc</a></h5>
<ul>
<li>Before a LXC guest is started, the lxc hook script is
called in three locations; if any location fails, the guest
is not started. The first location, <span class="since">since
0.9.13</span>, is before libvirt performs any resource
labeling, and the hook can allocate resources not managed by
libvirt such as DRBD or missing bridges. This is called as:<br/>
<pre>/etc/libvirt/hooks/lxc guest_name prepare begin -</pre>
The second location, available <span class="since">Since
0.8.0</span>, occurs after libvirt has finished labeling
all resources, but has not yet started the guest, called as:<br/>
<pre>/etc/libvirt/hooks/lxc guest_name start begin -</pre>
The third location, <span class="since">0.9.13</span>,
occurs after the LXC process has successfully started up:<br/>
<pre>/etc/libvirt/hooks/lxc guest_name started begin -</pre>
</li>
<li>When a LXC guest is stopped, the lxc hook script is called
in two locations, to match the startup.
First, <span class="since">since 0.8.0</span>, the hook is
called before libvirt restores any labels:<br/>
<pre>/etc/libvirt/hooks/lxc guest_name stopped end -</pre>
Then, after libvirt has released all resources, the hook is
called again, <span class="since">since 0.9.0</span>, to allow
any additional resource cleanup:<br/>
<pre>/etc/libvirt/hooks/lxc guest_name release end -</pre></li>
<li><span class="since">Since 0.9.13</span>, the lxc hook script
is also called when the libvirtd daemon restarts and reconnects
to previously running LXC processes. If the script fails, the
existing LXC process will be killed off. It is called as:
<pre>/etc/libvirt/hooks/lxc guest_name reconnect begin -</pre>
</li>
</ul>
<h5><a id="libxl">/etc/libvirt/hooks/libxl</a></h5>
<ul>
<li>Before a Xen guest is started using libxl driver, the libxl hook
script is called in three locations; if any location fails, the guest
is not started. The first location, <span class="since">since
2.1.0</span>, is before libvirt performs any resource
labeling, and the hook can allocate resources not managed by
libvirt. This is called as:<br/>
<pre>/etc/libvirt/hooks/libxl guest_name prepare begin -</pre>
The second location, available <span class="since">Since
2.1.0</span>, occurs after libvirt has finished labeling
all resources, but has not yet started the guest, called as:<br/>
<pre>/etc/libvirt/hooks/libxl guest_name start begin -</pre>
The third location, <span class="since">2.1.0</span>,
occurs after the domain has successfully started up:<br/>
<pre>/etc/libvirt/hooks/libxl guest_name started begin -</pre>
</li>
<li>When a libxl-handled Xen guest is stopped, the libxl hook script
is called in two locations, to match the startup.
First, <span class="since">since 2.1.0</span>, the hook is
called before libvirt restores any labels:<br/>
<pre>/etc/libvirt/hooks/libxl guest_name stopped end -</pre>
Then, after libvirt has released all resources, the hook is
called again, <span class="since">since 2.1.0</span>, to allow
any additional resource cleanup:<br/>
<pre>/etc/libvirt/hooks/libxl guest_name release end -</pre></li>
<li><span class="since">Since 2.1.0</span>, the libxl hook script
is also called at the beginning of incoming migration. It is called
as: <pre>/etc/libvirt/hooks/libxl guest_name migrate begin -</pre>
with domain XML sent to standard input of the script. In this case,
the script acts as a filter and is supposed to modify the domain
XML and print it out on its standard output. Empty output is
identical to copying the input XML without changing it. In case the
script returns failure or the output XML is not valid, incoming
migration will be canceled. This hook may be used, e.g., to change
location of disk images for incoming domains.</li>
<li><span class="since">Since 6.5.0</span>, you can also place several
hook scripts in the directory
<code>/etc/libvirt/hooks/libxl.d/</code>. They are executed in
alphabetical order after main script. In this case each script also
acts as filter and can modify the domain XML and print it out on
its standard output. This script output is passed to standard input
next script in order. Empty output from any script is also identical
to copying the input XML without changing it.
In case any script returns failure common process will be aborted,
but all scripts from the directory will are executed.</li>
<li><span class="since">Since 2.1.0</span>, the libxl hook script
is also called when the libvirtd daemon restarts and reconnects
to previously running Xen domains. If the script fails, the
existing Xen domains will be killed off. It is called as:
<pre>/etc/libvirt/hooks/libxl guest_name reconnect begin -</pre>
</li>
</ul>
<h5><a id="network">/etc/libvirt/hooks/network</a></h5>
<ul>
<li><span class="since">Since 1.2.2</span>, before a network is started,
this script is called as:<br/>
<pre>/etc/libvirt/hooks/network network_name start begin -</pre></li>
<li>After the network is started, up &amp; running, the script is
called as:<br/>
<pre>/etc/libvirt/hooks/network network_name started begin -</pre></li>
<li>When a network is shut down, this script is called as:<br/>
<pre>/etc/libvirt/hooks/network network_name stopped end -</pre></li>
<li>Later, when network is started and there's an interface from a
domain to be plugged into the network, the hook script is called as:<br/>
<pre>/etc/libvirt/hooks/network network_name port-created begin -</pre>
Please note, that in this case, the script is passed both network and
port XMLs on its stdin.</li>
<li>When network is updated, the hook script is called as:<br/>
<pre>/etc/libvirt/hooks/network network_name updated begin -</pre></li>
<li>When the domain from previous case is shutting down, the interface
is unplugged. This leads to another script invocation:<br/>
<pre>/etc/libvirt/hooks/network network_name port-deleted begin -</pre>
And again, as in previous case, both network and port XMLs are passed
onto script's stdin.</li>
</ul>
<br/>
<h2><a id="execution">Script execution</a></h2>
<ul>
<li>The "start" operation for the guest and network hook scripts,
executes <b>prior</b> to the object (guest or network) being created.
This allows the object start operation to be aborted if the script
returns indicating failure.<br/><br/></li>
<li>The "stopped" operation for the guest and network hook scripts,
executes <b>after</b> the object (guest or network) has stopped. If
the hook script indicates failure in its return, the shut down of the
object cannot be aborted because it has already been performed.
<br/><br/></li>
<li>Hook scripts execute in a synchronous fashion. Libvirt waits
for them to return before continuing the given operation.<br/><br/>
This is most noticeable with the guest or network start operation,
as a lengthy operation in the hook script can mean an extended wait
for the guest or network to be available to end users.<br/><br/></li>
<li>For a hook script to be utilised, it must have its execute bit set
(e.g. chmod o+rx <i>qemu</i>), and must be present when the libvirt
daemon is started.<br/><br/></li>
<li>If a hook script is added to a host after the libvirt daemon is
already running, it won't be used until the libvirt daemon
next starts.</li>
</ul>
<br/>
<h2><a id="qemu_migration">QEMU guest migration</a></h2>
<p>Migration of a QEMU guest involves running hook scripts on both the
source and destination hosts:</p>
<ol>
<li>At the beginning of the migration, the <i>qemu</i> hook script on
the <b>destination</b> host is executed with the "migrate"
operation.</li>
<li>Before QEMU process is spawned, the two operations ("prepare" and
"start") called for domain start are executed on
<b>destination</b> host.</li>
<li>If both of these hook script executions exit successfully (exit
status 0), the migration continues. Any other exit code indicates
failure, and the migration is aborted.</li>
<li>The QEMU guest is then migrated to the destination host.</li>
<li>Unless an error occurs during the migration process, the <i>qemu</i>
hook script on the <b>source</b> host is then executed with the
"stopped" and "release" operations to indicate it is no longer
running on this host. Regardless of the return codes, the
migration is not aborted as it has already been performed.</li>
</ol>
<br/>
<h2><a id="recursive">Calling libvirt functions from within a hook script</a></h2>
<p><b>DO NOT DO THIS!</b></p>
<p>A hook script must not call back into libvirt, as the libvirt daemon
is already waiting for the script to exit.</p>
<p>A deadlock is likely to occur.</p>
<br/>
<h2><a id="return_codes">Return codes and logging</a></h2>
<p>If a hook script returns with an exit code of 0, the libvirt daemon
regards this as successful and performs no logging of it.</p>
<p>However, if a hook script returns with a non zero exit code, the libvirt
daemon regards this as a failure, logs its return code, and
additionally logs anything on stderr the hook script returns.</p>
<p>For example, a hook script might use this code to indicate failure,
and send a text string to stderr:</p>
<pre>echo "Could not find required XYZZY" &gt;&amp;2
exit 1</pre>
<p>The resulting entry in the libvirt log will appear as:</p>
<pre>20:02:40.297: error : virHookCall:285 : Hook script execution failed: internal error Child process (LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
HOME=/root USER=root LOGNAME=root /etc/libvirt/hooks/qemu qemu prepare begin -) unexpected exit status 1: Could not find required XYZZY</pre>
</body>
</html>