proxmox-mirrors/mirror_lxc
1
0
Fork 0
mirror of https://git.proxmox.com/git/mirror_lxc synced 2025-08-17 21:43:57 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
stable-6.0
mirror_lxc/doc/lxc.container.conf.sgml.in
Stéphane Graber 8e67eb07c8
doc: Add SPDX headers and remove Author field 
Signed-off-by: Stéphane Graber <stgraber@stgraber.org>
2024-04-02 20:23:13 -04:00

3164 lines
122 KiB
Plaintext
Raw Permalink Blame History

<!-- SPDX-License-Identifier: LGPL-2.1+ -->
<!DOCTYPE refentry PUBLIC @docdtd@ [
<!ENTITY seealso SYSTEM "@builddir@/see_also.sgml">
]>
<refentry>
<docinfo><date>@LXC_GENERATE_DATE@</date></docinfo>
<refmeta>
<refentrytitle>lxc.container.conf</refentrytitle>
<manvolnum>5</manvolnum>
</refmeta>
<refnamediv>
<refname>lxc.container.conf</refname>
<refpurpose>
LXC container configuration file
</refpurpose>
</refnamediv>
<refsect1>
<title>Description</title>
<para>
LXC is the well-known and heavily tested low-level Linux container
runtime. It is in active development since 2008 and has proven itself in
critical production environments world-wide. Some of its core contributors
are the same people that helped to implement various well-known
containerization features inside the Linux kernel.
</para>
<para>
LXC's main focus is system containers. That is, containers which offer an
environment as close as possible as the one you'd get from a VM but
without the overhead that comes with running a separate kernel and
simulating all the hardware.
</para>
<para>
This is achieved through a combination of kernel security features such as
namespaces, mandatory access control and control groups.
</para>
<para>
LXC has support for unprivileged containers. Unprivileged containers are
containers that are run without any privilege. This requires support for
user namespaces in the kernel that the container is run on. LXC was the
first runtime to support unprivileged containers after user namespaces
were merged into the mainline kernel.
</para>
<para>
In essence, user namespaces isolate given sets of UIDs and GIDs. This is
achieved by establishing a mapping between a range of UIDs and GIDs on the
host to a different (unprivileged) range of UIDs and GIDs in the
container. The kernel will translate this mapping in such a way that
inside the container all UIDs and GIDs appear as you would expect from the
host whereas on the host these UIDs and GIDs are in fact unprivileged. For
example, a process running as UID and GID 0 inside the container might
appear as UID and GID 100000 on the host. The implementation and working
details can be gathered from the corresponding user namespace man page.
UID and GID mappings can be defined with the <option>lxc.idmap</option>
key.
</para>
<para>
Linux containers are defined with a simple configuration file. Each
option in the configuration file has the form <command>key =
value</command> fitting in one line. The "#" character means the line is a
comment. List options, like capabilities and cgroups options, can be used
with no value to clear any previously defined values of that option.
</para>
<para>
LXC namespaces configuration keys use single dots. This means complex
configuration keys such as <option>lxc.net.0</option> expose various
subkeys such as <option>lxc.net.0.type</option>,
<option>lxc.net.0.link</option>, <option>lxc.net.0.ipv6.address</option>, and
others for even more fine-grained configuration.
</para>
<refsect2>
<title>Configuration</title>
<para>
In order to ease administration of multiple related containers, it is
possible to have a container configuration file cause another file to be
loaded. For instance, network configuration can be defined in one common
file which is included by multiple containers. Then, if the containers
are moved to another host, only one file may need to be updated.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.include</option>
</term>
<listitem>
<para>
Specify the file to be included. The included file must be
in the same valid lxc configuration file format.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Architecture</title>
<para>
Allows one to set the architecture for the container. For example, set a
32bits architecture for a container running 32bits binaries on a 64bits
host. This fixes the container scripts which rely on the architecture to
do some work like downloading the packages.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.arch</option>
</term>
<listitem>
<para>
Specify the architecture for the container.
</para>
<para>
Some valid options are
<option>x86</option>,
<option>i686</option>,
<option>x86_64</option>,
<option>amd64</option>
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Hostname</title>
<para>
The utsname section defines the hostname to be set for the container.
That means the container can set its own hostname without changing the
one from the system. That makes the hostname private for the container.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.uts.name</option>
</term>
<listitem>
<para>
specify the hostname for the container
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Halt signal</title>
<para>
Allows one to specify signal name or number sent to the container's
init process to cleanly shutdown the container. Different init systems
could use different signals to perform clean shutdown sequence. This
option allows the signal to be specified in kill(1) fashion, e.g.
SIGPWR, SIGRTMIN+14, SIGRTMAX-10 or plain number. The default signal is
SIGPWR.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.signal.halt</option>
</term>
<listitem>
<para>
specify the signal used to halt the container
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Reboot signal</title>
<para>
Allows one to specify signal name or number to reboot the container.
This option allows signal to be specified in kill(1) fashion, e.g.
SIGTERM, SIGRTMIN+14, SIGRTMAX-10 or plain number. The default signal
is SIGINT.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.signal.reboot</option>
</term>
<listitem>
<para>
specify the signal used to reboot the container
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Stop signal</title>
<para>
Allows one to specify signal name or number to forcibly shutdown the
container. This option allows signal to be specified in kill(1) fashion,
e.g. SIGKILL, SIGRTMIN+14, SIGRTMAX-10 or plain number. The default
signal is SIGKILL.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.signal.stop</option>
</term>
<listitem>
<para>
specify the signal used to stop the container
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Init command</title>
<para>
Sets the command to use as the init system for the containers.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.execute.cmd</option>
</term>
<listitem>
<para>
Absolute path from container rootfs to the binary to run by default. This
mostly makes sense for <command>lxc-execute</command>.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.init.cmd</option>
</term>
<listitem>
<para>
Absolute path from container rootfs to the binary to use as init. This
mostly makes sense for <command>lxc-start</command>. Default is <command>/sbin/init</command>.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Init working directory</title>
<para>
Sets the absolute path inside the container as the working directory for the containers.
LXC will switch to this directory before executing init.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.init.cwd</option>
</term>
<listitem>
<para>
Absolute path inside the container to use as the working directory.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Init ID</title>
<para>
Sets the UID/GID to use for the init system, and subsequent commands.
Note that using a non-root UID when booting a system container will
likely not work due to missing privileges. Setting the UID/GID is mostly
useful when running application containers.
Defaults to: UID(0), GID(0)
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.init.uid</option>
</term>
<listitem>
<para>
UID to use for init.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.init.gid</option>
</term>
<listitem>
<para>
GID to use for init.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Core Scheduling</title>
<para>
Core scheduling defines if the container payload
is marked as being schedulable on the same core. Doing so will cause
the kernel scheduler to ensure that tasks that are not in the same
group never run simultaneously on a core. This can serve as an extra
security measure to prevent the container payload from using
cross hyper thread attacks.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.sched.core</option>
</term>
<listitem>
<para>
The only allowed values are 0 and 1. Set this to 1 to create a
core scheduling domain for the container or 0 to not create one.
If not set explicitly no core scheduling domain will be created
for the container.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Proc</title>
<para>
Configure proc filesystem for the container.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.proc.[proc file name]</option>
</term>
<listitem>
<para>
Specify the proc file name to be set. The file names available
are those listed under /proc/PID/.
Example:
</para>
<programlisting>
lxc.proc.oom_score_adj = 10
</programlisting>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Ephemeral</title>
<para>
Allows one to specify whether a container will be destroyed on shutdown.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.ephemeral</option>
</term>
<listitem>
<para>
The only allowed values are 0 and 1. Set this to 1 to destroy a
container on shutdown.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Network</title>
<para>
The network section defines how the network is virtualized in
the container. The network virtualization acts at layer
two. In order to use the network virtualization, parameters
must be specified to define the network interfaces of the
container. Several virtual interfaces can be assigned and used
in a container even if the system has only one physical
network interface.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.net</option>
</term>
<listitem>
<para>
may be used without a value to clear all previous network options.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].type</option>
</term>
<listitem>
<para>
specify what kind of network virtualization to be used
for the container.
Must be specified before any other option(s) on the net device.
Multiple networks can be specified by using an additional index
<option>i</option>
after all <option>lxc.net.*</option> keys. For example,
<option>lxc.net.0.type = veth</option> and
<option>lxc.net.1.type = veth</option> specify two different
networks of the same type. All keys sharing the same index
<option>i</option> will be treated as belonging to the same
network. For example, <option>lxc.net.0.link = br0</option>
will belong to <option>lxc.net.0.type</option>.
Currently, the different virtualization types can be:
</para>
<para>
<option>empty:</option> will create only the loopback
interface.
</para>
<para>
<option>veth:</option> a virtual ethernet pair
device is created with one side assigned to the container
and the other side on the host.
<option>lxc.net.[i].veth.mode</option> specifies the
mode the veth parent will use on the host.
The accepted modes are <option>bridge</option> and <option>router</option>.
The mode defaults to bridge if not specified.
In <option>bridge</option> mode the host side is attached to a bridge specified by
the <option>lxc.net.[i].link</option> option.
If the bridge link is not specified, then the veth pair device
will be created but not attached to any bridge.
Otherwise, the bridge has to be created on the system
before starting the container.
<command>lxc</command> won't handle any
configuration outside of the container.
In <option>router</option> mode static routes are created on the host for the
container's IP addresses pointing to the host side veth interface.
Additionally Proxy ARP and Proxy NDP entries are added on the host side veth interface
for the gateway IPs defined in the container to allow the container to reach the host.
By default, <command>lxc</command> chooses a name for the
network device belonging to the outside of the
container, but if you wish to handle
this name yourselves, you can tell <command>lxc</command>
to set a specific name with
the <option>lxc.net.[i].veth.pair</option> option (except for
unprivileged containers where this option is ignored for security
reasons).
Static routes can be added on the host pointing to the container using the
<option>lxc.net.[i].veth.ipv4.route</option> and
<option>lxc.net.[i].veth.ipv6.route</option> options.
Several lines specify several routes.
The route is in format x.y.z.t/m, eg. 192.168.1.0/24.
In <option>bridge</option> mode untagged VLAN membership can be set with the
<option>lxc.net.[i].veth.vlan.id</option> option. It accepts a special value of 'none' indicating
that the container port should be removed from the bridge's default untagged VLAN.
The <option>lxc.net.[i].veth.vlan.tagged.id</option> option can be specified multiple times to set
the container's bridge port membership to one or more tagged VLANs.
</para>
<para>
<option>vlan:</option> a vlan interface is linked with
the interface specified by
the <option>lxc.net.[i].link</option> and assigned to
the container. The vlan identifier is specified with the
option <option>lxc.net.[i].vlan.id</option>.
</para>
<para>
<option>macvlan:</option> a macvlan interface is linked
with the interface specified by
the <option>lxc.net.[i].link</option> and assigned to
the container.
<option>lxc.net.[i].macvlan.mode</option> specifies the
mode the macvlan will use to communicate between
different macvlan on the same upper device. The accepted
modes are <option>private</option>, <option>vepa</option>,
<option>bridge</option> and <option>passthru</option>.
In <option>private</option> mode, the device never
communicates with any other device on the same upper_dev (default).
In <option>vepa</option> mode, the new Virtual Ethernet Port
Aggregator (VEPA) mode, it assumes that the adjacent
bridge returns all frames where both source and
destination are local to the macvlan port, i.e. the
bridge is set up as a reflective relay. Broadcast
frames coming in from the upper_dev get flooded to all
macvlan interfaces in VEPA mode, local frames are not
delivered locally. In <option>bridge</option> mode, it
provides the behavior of a simple bridge between
different macvlan interfaces on the same port. Frames
from one interface to another one get delivered directly
and are not sent out externally. Broadcast frames get
flooded to all other bridge ports and to the external
interface, but when they come back from a reflective
relay, we don't deliver them again. Since we know all
the MAC addresses, the macvlan bridge mode does not
require learning or STP like the bridge module does. In
<option>passthru</option> mode, all frames received by
the physical interface are forwarded to the macvlan
interface. Only one macvlan interface in <option>passthru</option>
mode is possible for one physical interface.
</para>
<para>
<option>ipvlan:</option> an ipvlan interface is linked
with the interface specified by
the <option>lxc.net.[i].link</option> and assigned to
the container.
<option>lxc.net.[i].ipvlan.mode</option> specifies the
mode the ipvlan will use to communicate between
different ipvlan on the same upper device. The accepted
modes are <option>l3</option>, <option>l3s</option> and
<option>l2</option>. It defaults to <option>l3</option> mode.
In <option>l3</option> mode TX processing up to L3 happens on the stack instance
attached to the dependent device and packets are switched to the stack instance of the
parent device for the L2 processing and routing from that instance will be
used before packets are queued on the outbound device. In this mode the dependent devices
will not receive nor can send multicast / broadcast traffic.
In <option>l3s</option> mode TX processing is very similar to the L3 mode except that
iptables (conn-tracking) works in this mode and hence it is L3-symmetric (L3s).
This will have slightly less performance but that shouldn't matter since you are
choosing this mode over plain-L3 mode to make conn-tracking work.
In <option>l2</option> mode TX processing happens on the stack instance attached to
the dependent device and packets are switched and queued to the parent device to send devices
out. In this mode the dependent devices will RX/TX multicast and broadcast (if applicable) as well.
<option>lxc.net.[i].ipvlan.isolation</option> specifies the isolation mode.
The accepted isolation values are <option>bridge</option>,
<option>private</option> and <option>vepa</option>.
It defaults to <option>bridge</option>.
In <option>bridge</option> isolation mode dependent devices can cross-talk among themselves
apart from talking through the parent device.
In <option>private</option> isolation mode the port is set in private mode.
i.e. port won't allow cross communication between dependent devices.
In <option>vepa</option> isolation mode the port is set in VEPA mode.
i.e. port will offload switching functionality to the external entity as
described in 802.1Qbg.
</para>
<para>
<option>phys:</option> an already existing interface
specified by the <option>lxc.net.[i].link</option> is
assigned to the container.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].flags</option>
</term>
<listitem>
<para>
Specify an action to do for the network.
</para>
<para><option>up:</option> activates the interface.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].link</option>
</term>
<listitem>
<para>
Specify the interface to be used for real network traffic.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].l2proxy</option>
</term>
<listitem>
<para>
Controls whether layer 2 IP neighbour proxy entries will be added to the
lxc.net.[i].link interface for the IP addresses of the container.
Can be set to 0 or 1. Defaults to 0.
When used with IPv4 addresses, the following sysctl values need to be set:
net.ipv4.conf.[link].forwarding=1
When used with IPv6 addresses, the following sysctl values need to be set:
net.ipv6.conf.[link].proxy_ndp=1
net.ipv6.conf.[link].forwarding=1
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].mtu</option>
</term>
<listitem>
<para>
Specify the maximum transfer unit for this interface.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].name</option>
</term>
<listitem>
<para>
The interface name is dynamically allocated, but if another name
is needed because the configuration files being used by the
container use a generic name, eg. eth0, this option will rename
the interface in the container.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].hwaddr</option>
</term>
<listitem>
<para>
The interface mac address is dynamically allocated by default to
the virtual interface, but in some cases, this is needed to
resolve a mac address conflict or to always have the same
link-local ipv6 address. Any "x" in address will be replaced by
random value, this allows setting hwaddr templates.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].ipv4.address</option>
</term>
<listitem>
<para>
Specify the ipv4 address to assign to the virtualized interface.
Several lines specify several ipv4 addresses. The address is in
format x.y.z.t/m, eg. 192.168.1.123/24.
You can optionally specify the broadcast address after the IP address,
e.g. 192.168.1.123/24 255.255.255.255.
Otherwise it is automatically calculated from the IP address.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].ipv4.gateway</option>
</term>
<listitem>
<para>
Specify the ipv4 address to use as the gateway inside the
container. The address is in format x.y.z.t, eg. 192.168.1.123.
Can also have the special value <option>auto</option>,
which means to take the primary address from the bridge
interface (as specified by the
<option>lxc.net.[i].link</option> option) and use that as
the gateway. <option>auto</option> is only available when
using the <option>veth</option>,
<option>macvlan</option> and <option>ipvlan</option> network types.
Can also have the special value of <option>dev</option>,
which means to set the default gateway as a device route.
This is primarily for use with layer 3 network modes, such as IPVLAN.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].ipv6.address</option>
</term>
<listitem>
<para>
Specify the ipv6 address to assign to the virtualized
interface. Several lines specify several ipv6 addresses. The
address is in format x::y/m, eg.
2003:db8:1:0:214:1234:fe0b:3596/64
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].ipv6.gateway</option>
</term>
<listitem>
<para>
Specify the ipv6 address to use as the gateway inside the
container. The address is in format x::y, eg. 2003:db8:1:0::1
Can also have the special value <option>auto</option>,
which means to take the primary address from the bridge
interface (as specified by the
<option>lxc.net.[i].link</option> option) and use that as
the gateway. <option>auto</option> is only available when
using the <option>veth</option>,
<option>macvlan</option> and <option>ipvlan</option> network types.
Can also have the special value of <option>dev</option>,
which means to set the default gateway as a device route.
This is primarily for use with layer 3 network modes, such as IPVLAN.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].script.up</option>
</term>
<listitem>
<para>
Add a configuration option to specify a script to be
executed after creating and configuring the network used
from the host side.
</para>
<para>
In addition to the information available to all hooks. The
following information is provided to the script:
<itemizedlist>
<listitem>
<para>
LXC_HOOK_TYPE: the hook type. This is either 'up' or 'down'.
</para>
</listitem>
<listitem>
<para>
LXC_HOOK_SECTION: the section type 'net'.
</para>
</listitem>
<listitem>
<para>
LXC_NET_TYPE: the network type. This is one of the valid
network types listed here (e.g. 'vlan', 'macvlan', 'ipvlan', 'veth').
</para>
</listitem>
<listitem>
<para>
LXC_NET_PARENT: the parent device on the host. This is only
set for network types 'mavclan', 'veth', 'phys'.
</para>
</listitem>
<listitem>
<para>
LXC_NET_PEER: the name of the peer device on the host. This is
only set for 'veth' network types. Note that this information
is only available when <option>lxc.hook.version</option> is set
to 1.
</para>
</listitem>
</itemizedlist>
Whether this information is provided in the form of environment
variables or as arguments to the script depends on the value of
<option>lxc.hook.version</option>. If set to 1 then information is
provided in the form of environment variables. If set to 0
information is provided as arguments to the script.
</para>
<para>
Standard output from the script is logged at debug level.
Standard error is not logged, but can be captured by the
hook redirecting its standard error to standard output.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.net.[i].script.down</option>
</term>
<listitem>
<para>
Add a configuration option to specify a script to be
executed before destroying the network used from the
host side.
</para>
<para>
In addition to the information available to all hooks. The
following information is provided to the script:
<itemizedlist>
<listitem>
<para>
LXC_HOOK_TYPE: the hook type. This is either 'up' or 'down'.
</para>
</listitem>
<listitem>
<para>
LXC_HOOK_SECTION: the section type 'net'.
</para>
</listitem>
<listitem>
<para>
LXC_NET_TYPE: the network type. This is one of the valid
network types listed here (e.g. 'vlan', 'macvlan', 'ipvlan', 'veth').
</para>
</listitem>
<listitem>
<para>
LXC_NET_PARENT: the parent device on the host. This is only
set for network types 'mavclan', 'veth', 'phys'.
</para>
</listitem>
<listitem>
<para>
LXC_NET_PEER: the name of the peer device on the host. This is
only set for 'veth' network types. Note that this information
is only available when <option>lxc.hook.version</option> is set
to 1.
</para>
</listitem>
</itemizedlist>
Whether this information is provided in the form of environment
variables or as arguments to the script depends on the value of
<option>lxc.hook.version</option>. If set to 1 then information is
provided in the form of environment variables. If set to 0
information is provided as arguments to the script.
</para>
<para>
Standard output from the script is logged at debug level.
Standard error is not logged, but can be captured by the
hook redirecting its standard error to standard output.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>New pseudo tty instance (devpts)</title>
<para>
For stricter isolation the container can have its own private
instance of the pseudo tty.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.pty.max</option>
</term>
<listitem>
<para>
If set, the container will have a new pseudo tty
instance, making this private to it. The value specifies
the maximum number of pseudo ttys allowed for a pty
instance (this limitation is not implemented yet).
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Container system console</title>
<para>
If the container is configured with a root filesystem and the
inittab file is setup to use the console, you may want to specify
where the output of this console goes.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.console.buffer.size</option>
</term>
<listitem>
<para>
Setting this option instructs liblxc to allocate an in-memory
ringbuffer. The container's console output will be written to the
ringbuffer. Note that ringbuffer must be at least as big as a
standard page size. When passed a value smaller than a single page
size liblxc will allocate a ringbuffer of a single page size. A page
size is usually 4KB.
The keyword 'auto' will cause liblxc to allocate a ringbuffer of
128KB.
When manually specifying a size for the ringbuffer the value should
be a power of 2 when converted to bytes. Valid size prefixes are
'KB', 'MB', 'GB'. (Note that all conversions are based on multiples
of 1024. That means 'KB' == 'KiB', 'MB' == 'MiB', 'GB' == 'GiB'.
Additionally, the case of the suffix is ignored, i.e. 'kB', 'KB' and
'Kb' are treated equally.)
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.console.size</option>
</term>
<listitem>
<para>
Setting this option instructs liblxc to place a limit on the size of
the console log file specified in
<option>lxc.console.logfile</option>. Note that size of the log file
must be at least as big as a standard page size. When passed a value
smaller than a single page size liblxc will set the size of log file
to a single page size. A page size is usually 4KB.
The keyword 'auto' will cause liblxc to place a limit of 128KB on
the log file.
When manually specifying a size for the log file the value should
be a power of 2 when converted to bytes. Valid size prefixes are
'KB', 'MB', 'GB'. (Note that all conversions are based on multiples
of 1024. That means 'KB' == 'KiB', 'MB' == 'MiB', 'GB' == 'GiB'.
Additionally, the case of the suffix is ignored, i.e. 'kB', 'KB' and
'Kb' are treated equally.)
If users want to mirror the console ringbuffer on disk they should set
<option>lxc.console.size</option> equal to
<option>lxc.console.buffer.size</option>.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.console.logfile</option>
</term>
<listitem>
<para>
Specify a path to a file where the console output will be written.
Note that in contrast to the on-disk ringbuffer logfile this file
will keep growing potentially filling up the users disks if not
rotated and deleted. This problem can also be avoided by using the
in-memory ringbuffer options
<option>lxc.console.buffer.size</option> and
<option>lxc.console.buffer.logfile</option>.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.console.rotate</option>
</term>
<listitem>
<para>
Whether to rotate the console logfile specified in
<option>lxc.console.logfile</option>. Users can send an API
request to rotate the logfile. Note that the old logfile will have
the same name as the original with the suffix ".1" appended.
Users wishing to prevent the console log file from filling the
disk should rotate the logfile and delete it if unneeded. This
problem can also be avoided by using the in-memory ringbuffer
options <option>lxc.console.buffer.size</option> and
<option>lxc.console.buffer.logfile</option>.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.console.path</option>
</term>
<listitem>
<para>
Specify a path to a device to which the console will be
attached. The keyword 'none' will simply disable the
console. Note, when specifying 'none' and creating a device node
for the console in the container at /dev/console or bind-mounting
the hosts's /dev/console into the container at /dev/console the
container will have direct access to the hosts's /dev/console.
This is dangerous when the container has write access to the
device and should thus be used with caution.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Console through the ttys</title>
<para>
This option is useful if the container is configured with a root
filesystem and the inittab file is setup to launch a getty on the
ttys. The option specifies the number of ttys to be available for
the container. The number of gettys in the inittab file of the
container should not be greater than the number of ttys specified
in this option, otherwise the excess getty sessions will die and
respawn indefinitely giving annoying messages on the console or in
<filename>/var/log/messages</filename>.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.tty.max</option>
</term>
<listitem>
<para>
Specify the number of tty to make available to the
container.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Console devices location</title>
<para>
LXC consoles are provided through Unix98 PTYs created on the
host and bind-mounted over the expected devices in the container.
By default, they are bind-mounted over <filename>/dev/console</filename>
and <filename>/dev/ttyN</filename>. This can prevent package upgrades
in the guest. Therefore you can specify a directory location (under
<filename>/dev</filename> under which LXC will create the files and
bind-mount over them. These will then be symbolically linked to
<filename>/dev/console</filename> and <filename>/dev/ttyN</filename>.
A package upgrade can then succeed as it is able to remove and replace
the symbolic links.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.tty.dir</option>
</term>
<listitem>
<para>
Specify a directory under <filename>/dev</filename>
under which to create the container console devices. Note that LXC
will move any bind-mounts or device nodes for /dev/console into
this directory.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>/dev directory</title>
<para>
By default, lxc creates a few symbolic links (fd,stdin,stdout,stderr)
in the container's <filename>/dev</filename> directory but does not
automatically create device node entries. This allows the container's
<filename>/dev</filename> to be set up as needed in the container
rootfs. If lxc.autodev is set to 1, then after mounting the container's
rootfs LXC will mount a fresh tmpfs under <filename>/dev</filename>
(limited to 500K by default, unless defined in lxc.autodev.tmpfs.size)
and fill in a minimal set of initial devices.
This is generally required when starting a container containing
a "systemd" based "init" but may be optional at other times. Additional
devices in the containers /dev directory may be created through the
use of the <option>lxc.hook.autodev</option> hook.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.autodev</option>
</term>
<listitem>
<para>
Set this to 0 to stop LXC from mounting and populating a minimal
<filename>/dev</filename> when starting the container.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.autodev.tmpfs.size</option>
</term>
<listitem>
<para>
Set this to define the size of the /dev tmpfs.
The default value is 500000 (500K). If the parameter is used
but without value, the default value is used.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Mount points</title>
<para>
The mount points section specifies the different places to be
mounted. These mount points will be private to the container
and won't be visible by the processes running outside of the
container. This is useful to mount /etc, /var or /home for
examples.
</para>
<para>
NOTE - LXC will generally ensure that mount targets and relative
bind-mount sources are properly confined under the container
root, to avoid attacks involving over-mounting host directories
and files. (Symbolic links in absolute mount sources are ignored)
However, if the container configuration first mounts a directory which
is under the control of the container user, such as /home/joe, into
the container at some <filename>path</filename>, and then mounts
under <filename>path</filename>, then a TOCTTOU attack would be
possible where the container user modifies a symbolic link under
their home directory at just the right time.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.mount.fstab</option>
</term>
<listitem>
<para>
specify a file location in
the <filename>fstab</filename> format, containing the
mount information. The mount target location can and in
most cases should be a relative path, which will become
relative to the mounted container root. For instance,
</para>
<programlisting>
proc proc proc nodev,noexec,nosuid 0 0
</programlisting>
<para>
Will mount a proc filesystem under the container's /proc,
regardless of where the root filesystem comes from. This
is resilient to block device backed filesystems as well as
container cloning.
</para>
<para>
Note that when mounting a filesystem from an
image file or block device the third field (fs_vfstype)
cannot be auto as with
<citerefentry>
<refentrytitle>mount</refentrytitle>
<manvolnum>8</manvolnum>
</citerefentry>
but must be explicitly specified.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.mount.entry</option>
</term>
<listitem>
<para>
Specify a mount point corresponding to a line in the
fstab format.
Moreover lxc supports mount propagation, such as rshared or
rprivate, and adds three additional mount options.
<option>optional</option> don't fail if mount does not work.
<option>create=dir</option> or <option>create=file</option>
to create dir (or file) when the point will be mounted.
<option>relative</option> source path is taken to be relative to
the mounted container root. For instance,
</para>
<programlisting>
dev/null proc/kcore none bind,relative 0 0
</programlisting>
<para>
Will expand dev/null to ${<option>LXC_ROOTFS_MOUNT</option>}/dev/null,
and mount it to proc/kcore inside the container.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.mount.auto</option>
</term>
<listitem>
<para>
specify which standard kernel file systems should be
automatically mounted. This may dramatically simplify
the configuration. The file systems are:
</para>
<itemizedlist>
<listitem>
<para>
<option>proc:mixed</option> (or <option>proc</option>):
mount <filename>/proc</filename> as read-write, but
remount <filename>/proc/sys</filename> and
<filename>/proc/sysrq-trigger</filename> read-only
for security / container isolation purposes.
</para>
</listitem>
<listitem>
<para>
<option>proc:rw</option>: mount
<filename>/proc</filename> as read-write
</para>
</listitem>
<listitem>
<para>
<option>sys:mixed</option> (or <option>sys</option>):
mount <filename>/sys</filename> as read-only but with
/sys/devices/virtual/net writable.
</para>
</listitem>
<listitem>
<para>
<option>sys:ro</option>:
mount <filename>/sys</filename> as read-only
for security / container isolation purposes.
</para>
</listitem>
<listitem>
<para>
<option>sys:rw</option>: mount
<filename>/sys</filename> as read-write
</para>
</listitem>
<listitem>
<para>
<option>cgroup:mixed</option>:
Mount a tmpfs to <filename>/sys/fs/cgroup</filename>,
create directories for all hierarchies to which the container
is added, create subdirectories in those hierarchies with the
name of the cgroup, and bind-mount the container's own cgroup
into that directory. The container will be able to write to
its own cgroup directory, but not the parents, since they will
be remounted read-only.
</para>
</listitem>
<listitem>
<para>
<option>cgroup:mixed:force</option>:
The <option>force</option> option will cause LXC to perform
the cgroup mounts for the container under all circumstances.
Otherwise it is similar to <option>cgroup:mixed</option>.
This is mainly useful when the cgroup namespaces are enabled
where LXC will normally leave mounting cgroups to the init
binary of the container since it is perfectly safe to do so.
</para>
</listitem>
<listitem>
<para>
<option>cgroup:ro</option>:
similar to <option>cgroup:mixed</option>, but everything will
be mounted read-only.
</para>
</listitem>
<listitem>
<para>
<option>cgroup:ro:force</option>:
The <option>force</option> option will cause LXC to perform
the cgroup mounts for the container under all circumstances.
Otherwise it is similar to <option>cgroup:ro</option>.
This is mainly useful when the cgroup namespaces are enabled
where LXC will normally leave mounting cgroups to the init
binary of the container since it is perfectly safe to do so.
</para>
</listitem>
<listitem>
<para>
<option>cgroup:rw</option>: similar to
<option>cgroup:mixed</option>, but everything will be mounted
read-write. Note that the paths leading up to the container's
own cgroup will be writable, but will not be a cgroup
filesystem but just part of the tmpfs of
<filename>/sys/fs/cgroup</filename>
</para>
</listitem>
<listitem>
<para>
<option>cgroup:rw:force</option>:
The <option>force</option> option will cause LXC to perform
the cgroup mounts for the container under all circumstances.
Otherwise it is similar to <option>cgroup:rw</option>.
This is mainly useful when the cgroup namespaces are enabled
where LXC will normally leave mounting cgroups to the init
binary of the container since it is perfectly safe to do so.
</para>
</listitem>
<listitem>
<para>
<option>cgroup</option> (without specifier):
defaults to <option>cgroup:rw</option> if the
container retains the CAP_SYS_ADMIN capability,
<option>cgroup:mixed</option> otherwise.
</para>
</listitem>
<listitem>
<para>
<option>cgroup-full:mixed</option>:
mount a tmpfs to <filename>/sys/fs/cgroup</filename>,
create directories for all hierarchies to which
the container is added, bind-mount the hierarchies
from the host to the container and make everything
read-only except the container's own cgroup. Note
that compared to <option>cgroup</option>, where
all paths leading up to the container's own cgroup
are just simple directories in the underlying
tmpfs, here
<filename>/sys/fs/cgroup/$hierarchy</filename>
will contain the host's full cgroup hierarchy,
albeit read-only outside the container's own cgroup.
This may leak quite a bit of information into the
container.
</para>
</listitem>
<listitem>
<para>
<option>cgroup-full:mixed:force</option>:
The <option>force</option> option will cause LXC to perform
the cgroup mounts for the container under all circumstances.
Otherwise it is similar to <option>cgroup-full:mixed</option>.
This is mainly useful when the cgroup namespaces are enabled
where LXC will normally leave mounting cgroups to the init
binary of the container since it is perfectly safe to do so.
</para>
</listitem>
<listitem>
<para>
<option>cgroup-full:ro</option>: similar to
<option>cgroup-full:mixed</option>, but everything
will be mounted read-only.
</para>
</listitem>
<listitem>
<para>
<option>cgroup-full:ro:force</option>:
The <option>force</option> option will cause LXC to perform
the cgroup mounts for the container under all circumstances.
Otherwise it is similar to <option>cgroup-full:ro</option>.
This is mainly useful when the cgroup namespaces are enabled
where LXC will normally leave mounting cgroups to the init
binary of the container since it is perfectly safe to do so.
</para>
</listitem>
<listitem>
<para>
<option>cgroup-full:rw</option>: similar to
<option>cgroup-full:mixed</option>, but everything
will be mounted read-write. Note that in this case,
the container may escape its own cgroup. (Note also
that if the container has CAP_SYS_ADMIN support
and can mount the cgroup filesystem itself, it may
do so anyway.)
</para>
</listitem>
<listitem>
<para>
<option>cgroup-full:rw:force</option>:
The <option>force</option> option will cause LXC to perform
the cgroup mounts for the container under all circumstances.
Otherwise it is similar to <option>cgroup-full:rw</option>.
This is mainly useful when the cgroup namespaces are enabled
where LXC will normally leave mounting cgroups to the init
binary of the container since it is perfectly safe to do so.
</para>
</listitem>
<listitem>
<para>
<option>cgroup-full</option> (without specifier):
defaults to <option>cgroup-full:rw</option> if the
container retains the CAP_SYS_ADMIN capability,
<option>cgroup-full:mixed</option> otherwise.
</para>
</listitem>
</itemizedlist>
<para>
If cgroup namespaces are enabled, then any <option>cgroup</option>
auto-mounting request will be ignored, since the container can
mount the filesystems itself, and automounting can confuse the
container init.
</para>
<para>
Note that if automatic mounting of the cgroup filesystem
is enabled, the tmpfs under
<filename>/sys/fs/cgroup</filename> will always be
mounted read-write (but for the <option>:mixed</option>
and <option>:ro</option> cases, the individual
hierarchies,
<filename>/sys/fs/cgroup/$hierarchy</filename>, will be
read-only). This is in order to work around a quirk in
Ubuntu's
<citerefentry>
<refentrytitle>mountall</refentrytitle>
<manvolnum>8</manvolnum>
</citerefentry>
command that will cause containers to wait for user
input at boot if
<filename>/sys/fs/cgroup</filename> is mounted read-only
and the container can't remount it read-write due to a
lack of CAP_SYS_ADMIN.
</para>
<para>
Examples:
</para>
<programlisting>
lxc.mount.auto = proc sys cgroup
lxc.mount.auto = proc:rw sys:rw cgroup-full:rw
</programlisting>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Root file system</title>
<para>
The root file system of the container can be different than that
of the host system.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.rootfs.path</option>
</term>
<listitem>
<para>
specify the root file system for the container. It can
be an image file, a directory or a block device. If not
specified, the container shares its root file system
with the host.
</para>
<para>
For directory or simple block-device backed containers,
a pathname can be used. If the rootfs is backed by a nbd
device, then <filename>nbd:file:1</filename> specifies that
<filename>file</filename> should be attached to a nbd device,
and partition 1 should be mounted as the rootfs.
<filename>nbd:file</filename> specifies that the nbd device
itself should be mounted. <filename>overlayfs:/lower:/upper</filename>
specifies that the rootfs should be an overlay with <filename>/upper</filename>
being mounted read-write over a read-only mount of <filename>/lower</filename>.
For <filename>overlay</filename> multiple <filename>/lower</filename>
directories can be specified. <filename>loop:/file</filename> tells lxc to attach
<filename>/file</filename> to a loop device and mount the loop device.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.rootfs.mount</option>
</term>
<listitem>
<para>
where to recursively bind <option>lxc.rootfs.path</option>
before pivoting. This is to ensure success of the
<citerefentry>
<refentrytitle><command>pivot_root</command></refentrytitle>
<manvolnum>8</manvolnum>
</citerefentry>
syscall. Any directory suffices, the default should
generally work.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.rootfs.options</option>
</term>
<listitem>
<para>
Specify extra mount options to use when mounting the rootfs.
The format of the mount options corresponds to the
format used in fstab. In addition, LXC supports the custom
<option>idmap=</option> mount option. This option can be used
to tell LXC to create an idmapped mount for the container's
rootfs. This is useful when the user doesn't want to recursively
chown the rootfs of the container to match the idmapping of the
user namespace the container is going to use. Instead an
idmapped mount can be used to handle this.
The argument for
<option>idmap=</option>
can either be a path pointing to a user namespace file that
LXC will open and use to idmap the rootfs or the special value
"container" which will instruct LXC to use
the container's user namespace to idmap the rootfs.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.rootfs.managed</option>
</term>
<listitem>
<para>
Set this to 0 to indicate that LXC is not managing the
container storage, then LXC will not modify the
container storage. The default is 1.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Control groups ("cgroups")</title>
<para>
The control group section contains the configuration for the
different subsystem. <command>lxc</command> does not check the
correctness of the subsystem name. This has the disadvantage
of not detecting configuration errors until the container is
started, but has the advantage of permitting any future
subsystem.
</para>
<para>
The kernel implementation of cgroups has changed significantly over the
years. With Linux 4.5 support for a new cgroup filesystem was added
usually referred to as "cgroup2" or "unified hierarchy". Since then the
old cgroup filesystem is usually referred to as "cgroup1" or the
"legacy hierarchies". Please see the cgroups manual page for a detailed
explanation of the differences between the two versions.
</para>
<para>
LXC distinguishes settings for the legacy and the unified hierarchy by
using different configuration key prefixes. To alter settings for
controllers in a legacy hierarchy the key prefix
<option>lxc.cgroup.</option> must be used and in order to alter the
settings for a controller in the unified hierarchy the
<option>lxc.cgroup2.</option> key must be used. Note that LXC will
ignore <option>lxc.cgroup.</option> settings on systems that only use
the unified hierarchy. Conversely, it will ignore
<option>lxc.cgroup2.</option> options on systems that only use legacy
hierarchies.
</para>
<para>
At its core a cgroup hierarchy is a way to hierarchically organize
processes. Usually a cgroup hierarchy will have one or more
"controllers" enabled. A "controller" in a cgroup hierarchy is usually
responsible for distributing a specific type of system resource along
the hierarchy. Controllers include the "pids" controller, the "cpu"
controller, the "memory" controller and others. Some controllers
however do not fall into the category of distributing a system
resource, instead they are often referred to as "utility" controllers.
One utility controller is the device controller. Instead of
distributing a system resource it allows one to manage device access.
</para>
<para>
In the legacy hierarchy the device controller was implemented like most
other controllers as a set of files that could be written to. These
files where named "devices.allow" and "devices.deny". The legacy device
controller allowed the implementation of both "allowlists" and
"denylists".
</para>
<para>
An allowlist is a device program that by default blocks access to all
devices. In order to access specific devices "allow rules" for
particular devices or device classes must be specified. In contrast, a
denylist is a device program that by default allows access to all
devices. In order to restrict access to specific devices "deny rules"
for particular devices or device classes must be specified.
</para>
<para>
In the unified cgroup hierarchy the implementation of the device
controller has completely changed. Instead of files to read from and
write to a eBPF program of
<option>BPF_PROG_TYPE_CGROUP_DEVICE</option> can be attached to a
cgroup. Even though the kernel implementation has changed completely
LXC tries to allow for the same semantics to be followed in the legacy
device cgroup and the unified eBPF-based device controller. The
following paragraphs explain the semantics for the unified eBPF-based
device controller.
</para>
<para>
As mentioned the format for specifying device rules for the unified
eBPF-based device controller is the same as for the legacy cgroup
device controller; only the configuration key prefix has changed.
Specifically, device rules for the legacy cgroup device controller are
specified via <option>lxc.cgroup.devices.allow</option> and
<option>lxc.cgroup.devices.deny</option> whereas for the
cgroup2 eBPF-based device controller
<option>lxc.cgroup2.devices.allow</option> and
<option>lxc.cgroup2.devices.deny</option> must be used.
</para>
<para>
<itemizedlist>
<listitem>
<para>
A denylist device rule
<programlisting>
lxc.cgroup2.devices.deny = a
</programlisting>
will cause LXC to instruct the kernel to block access to all
devices by default. To grant access to devices allow device rules
must be added via the <option>lxc.cgroup2.devices.allow</option>
key. This is referred to as a "allowlist" device program.
</para>
</listitem>
<listitem>
<para>
An allowlist device rule
<programlisting>
lxc.cgroup2.devices.allow = a
</programlisting>
will cause LXC to instruct the kernel to allow access to all
devices by default. To deny access to devices deny device rules
must be added via <option>lxc.cgroup2.devices.deny</option> key.
This is referred to as a "denylist" device program.
</para>
</listitem>
<listitem>
<para>
Specifying any of the aforementioned two rules will cause all
previous rules to be cleared, i.e. the device list will be reset.
</para>
</listitem>
<listitem>
<para>
When an allowlist program is requested, i.e. access to all devices
is blocked by default, specific deny rules for individual devices
or device classes are ignored.
</para>
</listitem>
<listitem>
<para>
When a denylist program is requested, i.e. access to all devices
is allowed by default, specific allow rules for individual devices
or device classes are ignored.
</para>
</listitem>
</itemizedlist>
</para>
<para>
For example the set of rules:
<programlisting>
lxc.cgroup2.devices.deny = a
lxc.cgroup2.devices.allow = c *:* m
lxc.cgroup2.devices.allow = b *:* m
lxc.cgroup2.devices.allow = c 1:3 rwm
</programlisting>
implements an allowlist device program, i.e. the kernel will block
access to all devices not specifically allowed in this list. This
particular program states that all character and block devices may be
created but only /dev/null might be read or written.
</para>
<para>
If we instead switch to the following set of rules:
<programlisting>
lxc.cgroup2.devices.allow = a
lxc.cgroup2.devices.deny = c *:* m
lxc.cgroup2.devices.deny = b *:* m
lxc.cgroup2.devices.deny = c 1:3 rwm
</programlisting>
then LXC would instruct the kernel to implement a denylist, i.e. the
kernel will allow access to all devices not specifically denied in
this list. This particular program states that no character devices or
block devices might be created and that /dev/null is not allow allowed
to be read, written, or created.
</para>
<para>
Now consider the same program but followed by a "global rule"
which determines the type of device program (allowlist or
denylist) as explained above:
<programlisting>
lxc.cgroup2.devices.allow = a
lxc.cgroup2.devices.deny = c *:* m
lxc.cgroup2.devices.deny = b *:* m
lxc.cgroup2.devices.deny = c 1:3 rwm
lxc.cgroup2.devices.allow = a
</programlisting>
The last line will cause LXC to reset the device list without changing
the type of device program.
</para>
<para>
If we specify:
<programlisting>
lxc.cgroup2.devices.allow = a
lxc.cgroup2.devices.deny = c *:* m
lxc.cgroup2.devices.deny = b *:* m
lxc.cgroup2.devices.deny = c 1:3 rwm
lxc.cgroup2.devices.deny = a
</programlisting>
instead then the last line will cause LXC to reset the device list and
switch from an allowlist program to a denylist program.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.cgroup.[controller name].[controller file]</option>
</term>
<listitem>
<para>
Specify the control group value to be set on a legacy cgroup
hierarchy. The controller name is the literal name of the control
group. The permitted names and the syntax of their values is not
dictated by LXC, instead it depends on the features of the Linux
kernel running at the time the container is started, eg.
<option>lxc.cgroup.cpuset.cpus</option>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.cgroup2.[controller name].[controller file]</option>
</term>
<listitem>
<para>
Specify the control group value to be set on the unified cgroup
hierarchy. The controller name is the literal name of the control
group. The permitted names and the syntax of their values is not
dictated by LXC, instead it depends on the features of the Linux
kernel running at the time the container is started, eg.
<option>lxc.cgroup2.memory.high</option>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.cgroup.dir</option>
</term>
<listitem>
<para>
specify a directory or path in which the container's cgroup will
be created. For example, setting
<option>lxc.cgroup.dir = my-cgroup/first</option> for a container
named "c1" will create the container's cgroup as a sub-cgroup of
"my-cgroup". For example, if the user's current cgroup "my-user"
is located in the root cgroup of the cpuset controller in a
cgroup v1 hierarchy this would create the cgroup
"/sys/fs/cgroup/cpuset/my-user/my-cgroup/first/c1" for the
container. Any missing cgroups will be created by LXC. This
presupposes that the user has write access to its current cgroup.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.cgroup.dir.container</option>
</term>
<listitem>
<para>
This is similar to <option>lxc.cgroup.dir</option>, but must be
used together with <option>lxc.cgroup.dir.monitor</option> and
affects only the container's cgroup path. This option is mutually
exclusive with <option>lxc.cgroup.dir</option>.
Note that the final path the container attaches to may be
extended further by the
<option>lxc.cgroup.dir.container.inner</option> option.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.cgroup.dir.monitor</option>
</term>
<listitem>
<para>
This is the monitor process counterpart to
<option>lxc.cgroup.dir.container</option>.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.cgroup.dir.monitor.pivot</option>
</term>
<listitem>
<para>
On container termination the PID of the monitor process is attached to this cgroup.
This path should not be a subpath of any other configured cgroup dir to ensure
proper removal of other cgroup paths on container termination.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.cgroup.dir.container.inner</option>
</term>
<listitem>
<para>
Specify an additional subdirectory where the cgroup namespace
will be created. With this option, the cgroup limits will be
applied to the outer path specified in
<option>lxc.cgroup.dir.container</option>, which is not accessible
from within the container, making it possible to better enforce
limits for privileged containers in a way they cannot override
them.
This only works in conjunction with the
<option>lxc.cgroup.dir.container</option> and
<option>lxc.cgroup.dir.monitor</option> options and has otherwise
no effect.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.cgroup.relative</option>
</term>
<listitem>
<para>
Set this to 1 to instruct LXC to never escape to the
root cgroup. This makes it easy for users to adhere to
restrictions enforced by cgroup2 and
systemd. Specifically, this makes it possible to run LXC
containers as systemd services.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Capabilities</title>
<para>
The capabilities can be dropped in the container if this one
is run as root.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.cap.drop</option>
</term>
<listitem>
<para>
Specify the capability to be dropped in the container. A
single line defining several capabilities with a space
separation is allowed. The format is the lower case of
the capability definition without the "CAP_" prefix,
eg. CAP_SYS_MODULE should be specified as
sys_module. See
<citerefentry>
<refentrytitle><command>capabilities</command></refentrytitle>
<manvolnum>7</manvolnum>
</citerefentry>.
If used with no value, lxc will clear any drop capabilities
specified up to this point.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.cap.keep</option>
</term>
<listitem>
<para>
Specify the capability to be kept in the container. All other
capabilities will be dropped. When a special value of "none" is
encountered, lxc will clear any keep capabilities specified up
to this point. A value of "none" alone can be used to drop all
capabilities.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Namespaces</title>
<para>
A namespace can be cloned (<option>lxc.namespace.clone</option>),
kept (<option>lxc.namespace.keep</option>) or shared
(<option>lxc.namespace.share.[namespace identifier]</option>).
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.namespace.clone</option>
</term>
<listitem>
<para>
Specify namespaces which the container is supposed to be created
with. The namespaces to create are specified as a space separated
list. Each namespace must correspond to one of the standard
namespace identifiers as seen in the
<filename>/proc/PID/ns</filename> directory.
When <option>lxc.namespace.clone</option> is not explicitly set all
namespaces supported by the kernel and the current configuration
will be used.
</para>
<para>
To create a new mount, net and ipc namespace set
<option>lxc.namespace.clone=mount net ipc</option>.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.namespace.keep</option>
</term>
<listitem>
<para>
Specify namespaces which the container is supposed to inherit from
the process that created it. The namespaces to keep are specified as
a space separated list. Each namespace must correspond to one of the
standard namespace identifiers as seen in the
<filename>/proc/PID/ns</filename> directory.
The <option>lxc.namespace.keep</option> is a
denylist option, i.e. it is useful when enforcing that containers
must keep a specific set of namespaces.
</para>
<para>
To keep the network, user and ipc namespace set
<option>lxc.namespace.keep=user net ipc</option>.
</para>
<para>
Note that sharing pid namespaces will likely not work with most init
systems.
</para>
<para>
Note that if the container requests a new user namespace and the
container wants to inherit the network namespace it needs to inherit
the user namespace as well.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.namespace.share.[namespace identifier]</option>
</term>
<listitem>
<para>
Specify a namespace to inherit from another container or process.
The <option>[namespace identifier]</option> suffix needs to be
replaced with one of the namespaces that appear in the
<filename>/proc/PID/ns</filename> directory.
</para>
<para>
To inherit the namespace from another process set the
<option>lxc.namespace.share.[namespace identifier]</option> to the PID of
the process, e.g. <option>lxc.namespace.share.net=42</option>.
</para>
<para>
To inherit the namespace from another container set the
<option>lxc.namespace.share.[namespace identifier]</option> to the name of
the container, e.g. <option>lxc.namespace.share.pid=c3</option>.
</para>
<para>
To inherit the namespace from another container located in a
different path than the standard liblxc path set the
<option>lxc.namespace.share.[namespace identifier]</option> to the full
path to the container, e.g.
<option>lxc.namespace.share.user=/opt/c3</option>.
</para>
<para>
In order to inherit namespaces the caller needs to have sufficient
privilege over the process or container.
</para>
<para>
Note that sharing pid namespaces between system containers will
likely not work with most init systems.
</para>
<para>
Note that if two processes are in different user namespaces and one
process wants to inherit the other's network namespace it usually
needs to inherit the user namespace as well.
</para>
<para>
Note that without careful additional configuration of an LSM,
sharing user+pid namespaces with a task may allow that task to
escalate privileges to that of the task calling liblxc.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.time.offset.boot</option>
</term>
<listitem>
<para>
Specify a positive or negative offset for the boottime clock. The
format accepts hours (h), minutes (m), seconds (s),
milliseconds (ms), microseconds (us), and nanoseconds (ns).
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.time.offset.monotonic</option>
</term>
<listitem>
<para>
Specify a positive or negative offset for the monotonic clock. The
format accepts hours (h), minutes (m), seconds (s),
milliseconds (ms), microseconds (us), and nanoseconds (ns).
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Resource limits</title>
<para>
The soft and hard resource limits for the container can be changed.
Unprivileged containers can only lower them. Resources which are not
explicitly specified will be inherited.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.prlimit.[limit name]</option>
</term>
<listitem>
<para>
Specify the resource limit to be set. A limit is specified as two
colon separated values which are either numeric or the word
'unlimited'. A single value can be used as a shortcut to set both
soft and hard limit to the same value. The permitted names the
"RLIMIT_" resource names in lowercase without the "RLIMIT_"
prefix, eg. RLIMIT_NOFILE should be specified as "nofile". See
<citerefentry>
<refentrytitle><command>setrlimit</command></refentrytitle>
<manvolnum>2</manvolnum>
</citerefentry>.
If used with no value, lxc will clear the resource limit
specified up to this point. A resource with no explicitly
configured limitation will be inherited from the process starting
up the container.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Sysctl</title>
<para>
Configure kernel parameters for the container.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.sysctl.[kernel parameters name]</option>
</term>
<listitem>
<para>
Specify the kernel parameters to be set. The parameters available
are those listed under /proc/sys/.
Note that not all sysctls are namespaced. Changing Non-namespaced
sysctls will cause the system-wide setting to be modified.
<citerefentry>
<refentrytitle><command>sysctl</command></refentrytitle>
<manvolnum>8</manvolnum>
</citerefentry>.
If used with no value, lxc will clear the parameters specified up
to this point.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Apparmor profile</title>
<para>
If lxc was compiled and installed with apparmor support, and the host
system has apparmor enabled, then the apparmor profile under which the
container should be run can be specified in the container
configuration. The default is <command>lxc-container-default-cgns</command>
if the host kernel is cgroup namespace aware, or
<command>lxc-container-default</command> otherwise.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.apparmor.profile</option>
</term>
<listitem>
<para>
Specify the apparmor profile under which the container should
be run. To specify that the container should be unconfined,
use
</para>
<programlisting>lxc.apparmor.profile = unconfined</programlisting>
<para>
If the apparmor profile should remain unchanged (i.e. if you
are nesting containers and are already confined), then use
</para>
<programlisting>lxc.apparmor.profile = unchanged</programlisting>
<para>
If you instruct LXC to generate the apparmor profile,
then use
</para>
<programlisting>lxc.apparmor.profile = generated</programlisting>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.apparmor.allow_incomplete</option>
</term>
<listitem>
<para>
Apparmor profiles are pathname based. Therefore many file
restrictions require mount restrictions to be effective against
a determined attacker. However, these mount restrictions are not
yet implemented in the upstream kernel. Without the mount
restrictions, the apparmor profiles still protect against accidental
damager.
</para>
<para>
If this flag is 0 (default), then the container will not be
started if the kernel lacks the apparmor mount features, so that a
regression after a kernel upgrade will be detected. To start the
container under partial apparmor protection, set this flag to 1.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.apparmor.allow_nesting</option>
</term>
<listitem>
<para>
If set this to 1, causes the following changes. When
generated apparmor profiles are used, they will contain
the necessary changes to allow creating a nested
container. In addition to the usual mount points,
<filename>/dev/.lxc/proc</filename>
and <filename>/dev/.lxc/sys</filename> will contain
procfs and sysfs mount points without the lxcfs
overlays, which, if generated apparmor profiles are
being used, will not be read/writable directly.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.apparmor.raw</option>
</term>
<listitem>
<para>
A list of raw AppArmor profile lines to append to the
profile. Only valid when using generated profiles.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>SELinux context</title>
<para>
If lxc was compiled and installed with SELinux support, and the host
system has SELinux enabled, then the SELinux context under which the
container should be run can be specified in the container
configuration. The default is <command>unconfined_t</command>,
which means that lxc will not attempt to change contexts.
See @DATADIR@/lxc/selinux/lxc.te for an example policy and more
information.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.selinux.context</option>
</term>
<listitem>
<para>
Specify the SELinux context under which the container should
be run or <command>unconfined_t</command>. For example
</para>
<programlisting>lxc.selinux.context = system_u:system_r:lxc_t:s0:c22</programlisting>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.selinux.context.keyring</option>
</term>
<listitem>
<para>
Specify the SELinux context under which the container's keyring
should be created. By default this the same as lxc.selinux.context, or
the context lxc is executed under if lxc.selinux.context has not been set.
</para>
<programlisting>lxc.selinux.context.keyring = system_u:system_r:lxc_t:s0:c22</programlisting>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Kernel Keyring</title>
<para>
The Linux Keyring facility is primarily a way for various
kernel components to retain or cache security data, authentication
keys, encryption keys, and other data in the kernel. By default lxc
will create a new session keyring for the started application.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.keyring.session</option>
</term>
<listitem>
<para>
Disable the creation of new session keyring by lxc. The started
application will then inherit the current session keyring.
By default, or when passing the value 1, a new keyring will be created.
</para>
<programlisting>lxc.keyring.session = 0</programlisting>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Seccomp configuration</title>
<para>
A container can be started with a reduced set of available
system calls by loading a seccomp profile at startup. The
seccomp configuration file must begin with a version number
on the first line, a policy type on the second line, followed
by the configuration.
</para>
<para>
Versions 1 and 2 are currently supported. In version 1, the
policy is a simple allowlist. The second line therefore must
read "allowlist", with the rest of the file containing one (numeric)
syscall number per line. Each syscall number is allowlisted,
while every unlisted number is denylisted for use in the container
</para>
<para>
In version 2, the policy may be denylist or allowlist,
supports per-rule and per-policy default actions, and supports
per-architecture system call resolution from textual names.
</para>
<para>
An example denylist policy, in which all system calls are
allowed except for mknod, which will simply do nothing and
return 0 (success), looks like:
</para>
<programlisting>
2
denylist
mknod errno 0
ioctl notify
</programlisting>
<para>
Specifying "errno" as action will cause LXC to register a seccomp filter
that will cause a specific errno to be returned to the caller. The errno
value can be specified after the "errno" action word.
</para>
<para>
Specifying "notify" as action will cause LXC to register a seccomp
listener and retrieve a listener file descriptor from the kernel. When a
syscall is made that is registered as "notify" the kernel will generate a
poll event and send a message over the file descriptor. The caller can
read this message, inspect the syscalls including its arguments. Based on
this information the caller is expected to send back a message informing
the kernel which action to take. Until that message is sent the kernel
will block the calling process. The format of the messages to read and
sent is documented in seccomp itself.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.seccomp.profile</option>
</term>
<listitem>
<para>
Specify a file containing the seccomp configuration to
load before the container starts.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.seccomp.allow_nesting</option>
</term>
<listitem>
<para>
If this flag is set to 1, then seccomp filters will be stacked
regardless of whether a seccomp profile is already loaded.
This allows nested containers to load their own seccomp profile.
The default setting is 0.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.seccomp.notify.proxy</option>
</term>
<listitem>
<para>
Specify a unix socket to which LXC will connect and forward
seccomp events to. The path must be in the form
unix:/path/to/socket or unix:@socket. The former specifies a
path-bound unix domain socket while the latter specifies an
abstract unix domain socket.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.seccomp.notify.cookie</option>
</term>
<listitem>
<para>
An additional string sent along with proxied seccomp notification
requests.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>PR_SET_NO_NEW_PRIVS</title>
<para>
With PR_SET_NO_NEW_PRIVS active execve() promises not to grant
privileges to do anything that could not have been done without
the execve() call (for example, rendering the set-user-ID and
set-group-ID mode bits, and file capabilities non-functional).
Once set, this bit cannot be unset. The setting of this bit is
inherited by children created by fork() and clone(), and preserved
across execve().
Note that PR_SET_NO_NEW_PRIVS is applied after the container has
changed into its intended AppArmor profile or SElinux context.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.no_new_privs</option>
</term>
<listitem>
<para>
Specify whether the PR_SET_NO_NEW_PRIVS flag should be set for the
container. Set to 1 to activate.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>UID mappings</title>
<para>
A container can be started in a private user namespace with
user and group id mappings. For instance, you can map userid
0 in the container to userid 200000 on the host. The root
user in the container will be privileged in the container,
but unprivileged on the host. Normally a system container
will want a range of ids, so you would map, for instance,
user and group ids 0 through 20,000 in the container to the
ids 200,000 through 220,000.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.idmap</option>
</term>
<listitem>
<para>
Four values must be provided. First a character, either
'u', or 'g', to specify whether user or group ids are
being mapped. Next is the first userid as seen in the
user namespace of the container. Next is the userid as
seen on the host. Finally, a range indicating the number
of consecutive ids to map.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Container hooks</title>
<para>
Container hooks are programs or scripts which can be executed
at various times in a container's lifetime.
</para>
<para>
When a container hook is executed, additional information is passed
along. The <option>lxc.hook.version</option> argument can be used to
determine if the following arguments are passed as command line
arguments or through environment variables. The arguments are:
<itemizedlist>
<listitem><para> Container name. </para></listitem>
<listitem><para> Section (always 'lxc'). </para></listitem>
<listitem><para> The hook type (i.e. 'clone' or 'pre-mount'). </para></listitem>
<listitem><para> Additional arguments. In the
case of the clone hook, any extra arguments passed will appear as
further arguments to the hook. In the case of the stop hook, paths to
filedescriptors for each of the container's namespaces along with
their types are passed. </para></listitem>
</itemizedlist>
The following environment variables are set:
<itemizedlist>
<listitem><para> LXC_CGNS_AWARE: indicator whether the container is
cgroup namespace aware. </para></listitem>
<listitem><para> LXC_CONFIG_FILE: the path to the container
configuration file. </para></listitem>
<listitem><para> LXC_HOOK_TYPE: the hook type (e.g. 'clone', 'mount',
'pre-mount'). Note that the existence of this environment variable is
conditional on the value of <option>lxc.hook.version</option>. If it
is set to 1 then LXC_HOOK_TYPE will be set.
</para></listitem>
<listitem><para> LXC_HOOK_SECTION: the section type (e.g. 'lxc',
'net'). Note that the existence of this environment variable is
conditional on the value of <option>lxc.hook.version</option>. If it
is set to 1 then LXC_HOOK_SECTION will be set.
</para></listitem>
<listitem><para> LXC_HOOK_VERSION: the version of the hooks. This
value is identical to the value of the container's
<option>lxc.hook.version</option> config item. If it is set to 0 then
old-style hooks are used. If it is set to 1 then new-style hooks are
used. </para></listitem>
<listitem><para> LXC_LOG_LEVEL: the container's log level. </para></listitem>
<listitem><para> LXC_NAME: is the container's name. </para></listitem>
<listitem><para> LXC_[NAMESPACE IDENTIFIER]_NS: path under
/proc/PID/fd/ to a file descriptor referring to the container's
namespace. For each preserved namespace type there will be a separate
environment variable. These environment variables will only be set if
<option>lxc.hook.version</option> is set to 1. </para></listitem>
<listitem><para> LXC_ROOTFS_MOUNT: the path to the mounted root filesystem. </para></listitem>
<listitem><para> LXC_ROOTFS_PATH: this is the lxc.rootfs.path entry
for the container. Note this is likely not where the mounted rootfs is
to be found, use LXC_ROOTFS_MOUNT for that. </para></listitem>
<listitem><para> LXC_SRC_NAME: in the case of the clone hook, this is
the original container's name. </para></listitem>
</itemizedlist>
</para>
<para>
Standard output from the hooks is logged at debug level.
Standard error is not logged, but can be captured by the
hook redirecting its standard error to standard output.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.version</option>
</term>
<listitem>
<para>
To pass the arguments in new style via environment variables set to
1 otherwise set to 0 to pass them as arguments.
This setting affects all hooks arguments that were traditionally
passed as arguments to the script. Specifically, it affects the
container name, section (e.g. 'lxc', 'net') and hook type (e.g.
'clone', 'mount', 'pre-mount') arguments. If new-style hooks are
used then the arguments will be available as environment variables.
The container name will be set in LXC_NAME. (This is set
independently of the value used for this config item.) The section
will be set in LXC_HOOK_SECTION and the hook type will be set in
LXC_HOOK_TYPE.
It also affects how the paths to file descriptors referring to the
container's namespaces are passed. If set to 1 then for each
namespace a separate environment variable LXC_[NAMESPACE
IDENTIFIER]_NS will be set. If set to 0 then the paths will be
passed as arguments to the stop hook.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.pre-start</option>
</term>
<listitem>
<para>
A hook to be run in the host's namespace before the
container ttys, consoles, or mounts are up.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.pre-mount</option>
</term>
<listitem>
<para>
A hook to be run in the container's fs namespace but before
the rootfs has been set up. This allows for manipulation
of the rootfs, i.e. to mount an encrypted filesystem. Mounts
done in this hook will not be reflected on the host (apart from
mounts propagation), so they will be automatically cleaned up
when the container shuts down.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.mount</option>
</term>
<listitem>
<para>
A hook to be run in the container's namespace after
mounting has been done, but before the pivot_root.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.autodev</option>
</term>
<listitem>
<para>
A hook to be run in the container's namespace after
mounting has been done and after any mount hooks have
run, but before the pivot_root, if
<option>lxc.autodev</option> == 1.
The purpose of this hook is to assist in populating the
/dev directory of the container when using the autodev
option for systemd based containers. The container's /dev
directory is relative to the
${<option>LXC_ROOTFS_MOUNT</option>} environment
variable available when the hook is run.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.start-host</option>
</term>
<listitem>
<para>
A hook to be run in the host's namespace after the
container has been setup, and immediately before starting
the container init.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.start</option>
</term>
<listitem>
<para>
A hook to be run in the container's namespace immediately
before executing the container's init. This requires the
program to be available in the container.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.stop</option>
</term>
<listitem>
<para>
A hook to be run in the host's namespace with references
to the container's namespaces after the container has been shut
down. For each namespace an extra argument is passed to the hook
containing the namespace's type and a filename that can be used to
obtain a file descriptor to the corresponding namespace, separated
by a colon. The type is the name as it would appear in the
<filename>/proc/PID/ns</filename> directory.
For instance for the mount namespace the argument usually looks
like <filename>mnt:/proc/PID/fd/12</filename>.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.post-stop</option>
</term>
<listitem>
<para>
A hook to be run in the host's namespace after the
container has been shut down.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.clone</option>
</term>
<listitem>
<para>
A hook to be run when the container is cloned to a new one.
See <citerefentry><refentrytitle><command>lxc-clone</command></refentrytitle>
<manvolnum>1</manvolnum></citerefentry> for more information.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>lxc.hook.destroy</option>
</term>
<listitem>
<para>
A hook to be run when the container is destroyed.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Container hooks Environment Variables</title>
<para>
A number of environment variables are made available to the startup
hooks to provide configuration information and assist in the
functioning of the hooks. Not all variables are valid in all
contexts. In particular, all paths are relative to the host system
and, as such, not valid during the <option>lxc.hook.start</option> hook.
</para>
<variablelist>
<varlistentry>
<term>
<option>LXC_NAME</option>
</term>
<listitem>
<para>
The LXC name of the container. Useful for logging messages
in common log environments. [<option>-n</option>]
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>LXC_CONFIG_FILE</option>
</term>
<listitem>
<para>
Host relative path to the container configuration file. This
gives the container to reference the original, top level,
configuration file for the container in order to locate any
additional configuration information not otherwise made
available. [<option>-f</option>]
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>LXC_CONSOLE</option>
</term>
<listitem>
<para>
The path to the console output of the container if not NULL.
[<option>-c</option>] [<option>lxc.console.path</option>]
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>LXC_CONSOLE_LOGPATH</option>
</term>
<listitem>
<para>
The path to the console log output of the container if not NULL.
[<option>-L</option>]
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>LXC_ROOTFS_MOUNT</option>
</term>
<listitem>
<para>
The mount location to which the container is initially bound.
This will be the host relative path to the container rootfs
for the container instance being started and is where changes
should be made for that instance.
[<option>lxc.rootfs.mount</option>]
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>LXC_ROOTFS_PATH</option>
</term>
<listitem>
<para>
The host relative path to the container root which has been
mounted to the rootfs.mount location.
[<option>lxc.rootfs.path</option>]
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>LXC_SRC_NAME</option>
</term>
<listitem>
<para>
Only for the clone hook. Is set to the original container name.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>LXC_TARGET</option>
</term>
<listitem>
<para>
Only for the stop hook. Is set to "stop" for a container
shutdown or "reboot" for a container reboot.
</para>
</listitem>
</varlistentry>
</variablelist>
<variablelist>
<varlistentry>
<term>
<option>LXC_CGNS_AWARE</option>
</term>
<listitem>
<para>
If unset, then this version of lxc is not aware of cgroup
namespaces. If set, it will be set to 1, and lxc is aware
of cgroup namespaces. Note this does not guarantee that
cgroup namespaces are enabled in the kernel. This is used
by the lxcfs mount hook.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Logging</title>
<para>
Logging can be configured on a per-container basis. By default,
depending upon how the lxc package was compiled, container startup
is logged only at the ERROR level, and logged to a file named after
the container (with '.log' appended) either under the container path,
or under @LOGPATH@.
</para>
<para>
Both the default log level and the log file can be specified in the
container configuration file, overriding the default behavior. Note
that the configuration file entries can in turn be overridden by the
command line options to <command>lxc-start</command>.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.log.level</option>
</term>
<listitem>
<para>
The level at which to log. The log level is an integer in
the range of 0..8 inclusive, where a lower number means more
verbose debugging. In particular 0 = trace, 1 = debug, 2 =
info, 3 = notice, 4 = warn, 5 = error, 6 = critical, 7 =
alert, and 8 = fatal. If unspecified, the level defaults
to 5 (error), so that only errors and above are logged.
</para>
<para>
Note that when a script (such as either a hook script or a
network interface up or down script) is called, the script's
standard output is logged at level 1, debug.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.log.file</option>
</term>
<listitem>
<para>
The file to which logging info should be written.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.log.syslog</option>
</term>
<listitem>
<para>
Send logging info to syslog. It respects the log level defined in
<command>lxc.log.level</command>. The argument should be the syslog
facility to use, valid ones are: daemon, local0, local1, local2,
local3, local4, local5, local5, local6, local7.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Autostart</title>
<para>
The autostart options support marking which containers should be
auto-started and in what order. These options may be used by LXC tools
directly or by external tooling provided by the distributions.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.start.auto</option>
</term>
<listitem>
<para>
Whether the container should be auto-started.
Valid values are 0 (off) and 1 (on).
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.start.delay</option>
</term>
<listitem>
<para>
How long to wait (in seconds) after the container is
started before starting the next one.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.start.order</option>
</term>
<listitem>
<para>
An integer used to sort the containers when auto-starting
a series of containers at once. A lower value means an
earlier start.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.monitor.unshare</option>
</term>
<listitem>
<para>
If not zero the mount namespace will be unshared from the host
before initializing the container (before running any pre-start
hooks). This requires the CAP_SYS_ADMIN capability at startup.
Default is 0.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.monitor.signal.pdeath</option>
</term>
<listitem>
<para>
Set the signal to be sent to the container's init when the lxc
monitor exits. By default it is set to SIGKILL which will cause
all container processes to be killed when the lxc monitor process
dies.
To ensure that containers stay alive even if lxc monitor dies set
this to 0.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
<option>lxc.group</option>
</term>
<listitem>
<para>
A multi-value key (can be used multiple times) to put the
container in a container group. Those groups can then be
used (amongst other things) to start a series of related
containers.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
<refsect2>
<title>Autostart and System Boot</title>
<para>
Each container can be part of any number of groups or no group at all.
Two groups are special. One is the NULL group, i.e. the container does
not belong to any group. The other group is the "onboot" group.
</para>
<para>
When the system boots with the LXC service enabled, it will first
attempt to boot any containers with lxc.start.auto == 1 that is a member
of the "onboot" group. The startup will be in order of lxc.start.order.
If an lxc.start.delay has been specified, that delay will be honored
before attempting to start the next container to give the current
container time to begin initialization and reduce overloading the host
system. After starting the members of the "onboot" group, the LXC system
will proceed to boot containers with lxc.start.auto == 1 which are not
members of any group (the NULL group) and proceed as with the onboot
group.
</para>
</refsect2>
<refsect2>
<title>Container Environment</title>
<para>
If you want to pass environment variables into the container (that
is, environment variables which will be available to init and all of
its descendents), you can use <command>lxc.environment</command>
parameters to do so. Be careful that you do not pass in anything
sensitive; any process in the container which doesn't have its
environment scrubbed will have these variables available to it, and
environment variables are always available via
<command>/proc/PID/environ</command>.
</para>
<para>
This configuration parameter can be specified multiple times; once
for each environment variable you wish to configure.
</para>
<variablelist>
<varlistentry>
<term>
<option>lxc.environment</option>
</term>
<listitem>
<para>
Specify an environment variable to pass into the container.
Example:
</para>
<programlisting>
lxc.environment = APP_ENV=production
lxc.environment = SYSLOG_SERVER=192.0.2.42
</programlisting>
<para>
It is possible to inherit host environment variables by setting the
name of the variable without a "=" sign. For example:
</para>
<programlisting>
lxc.environment = PATH
</programlisting>
</listitem>
</varlistentry>
</variablelist>
</refsect2>
</refsect1>
<refsect1>
<title>Examples</title>
<para>
In addition to the few examples given below, you will find
some other examples of configuration file in @DOCDIR@/examples
</para>
<refsect2>
<title>Network</title>
<para>This configuration sets up a container to use a veth pair
device with one side plugged to a bridge br0 (which has been
configured before on the system by the administrator). The
virtual network device visible in the container is renamed to
eth0.</para>
<programlisting>
lxc.uts.name = myhostname
lxc.net.0.type = veth
lxc.net.0.flags = up
lxc.net.0.link = br0
lxc.net.0.name = eth0
lxc.net.0.hwaddr = 4a:49:43:49:79:bf
lxc.net.0.ipv4.address = 10.2.3.5/24 10.2.3.255
lxc.net.0.ipv6.address = 2003:db8:1:0:214:1234:fe0b:3597
</programlisting>
</refsect2>
<refsect2>
<title>UID/GID mapping</title>
<para>This configuration will map both user and group ids in the
range 0-9999 in the container to the ids 100000-109999 on the host.
</para>
<programlisting>
lxc.idmap = u 0 100000 10000
lxc.idmap = g 0 100000 10000
</programlisting>
</refsect2>
<refsect2>
<title>Control group</title>
<para>This configuration will setup several control groups for
the application, cpuset.cpus restricts usage of the defined cpu,
cpus.share prioritize the control group, devices.allow makes
usable the specified devices.</para>
<programlisting>
lxc.cgroup.cpuset.cpus = 0,1
lxc.cgroup.cpu.shares = 1234
lxc.cgroup.devices.deny = a
lxc.cgroup.devices.allow = c 1:3 rw
lxc.cgroup.devices.allow = b 8:0 rw
</programlisting>
</refsect2>
<refsect2>
<title>Complex configuration</title>
<para>This example show a complex configuration making a complex
network stack, using the control groups, setting a new hostname,
mounting some locations and a changing root file system.</para>
<programlisting>
lxc.uts.name = complex
lxc.net.0.type = veth
lxc.net.0.flags = up
lxc.net.0.link = br0
lxc.net.0.hwaddr = 4a:49:43:49:79:bf
lxc.net.0.ipv4.address = 10.2.3.5/24 10.2.3.255
lxc.net.0.ipv6.address = 2003:db8:1:0:214:1234:fe0b:3597
lxc.net.0.ipv6.address = 2003:db8:1:0:214:5432:feab:3588
lxc.net.1.type = macvlan
lxc.net.1.flags = up
lxc.net.1.link = eth0
lxc.net.1.hwaddr = 4a:49:43:49:79:bd
lxc.net.1.ipv4.address = 10.2.3.4/24
lxc.net.1.ipv4.address = 192.168.10.125/24
lxc.net.1.ipv6.address = 2003:db8:1:0:214:1234:fe0b:3596
lxc.net.2.type = phys
lxc.net.2.flags = up
lxc.net.2.link = random0
lxc.net.2.hwaddr = 4a:49:43:49:79:ff
lxc.net.2.ipv4.address = 10.2.3.6/24
lxc.net.2.ipv6.address = 2003:db8:1:0:214:1234:fe0b:3297
lxc.cgroup.cpuset.cpus = 0,1
lxc.cgroup.cpu.shares = 1234
lxc.cgroup.devices.deny = a
lxc.cgroup.devices.allow = c 1:3 rw
lxc.cgroup.devices.allow = b 8:0 rw
lxc.mount.fstab = /etc/fstab.complex
lxc.mount.entry = /lib /root/myrootfs/lib none ro,bind 0 0
lxc.rootfs.path = dir:/mnt/rootfs.complex
lxc.rootfs.options = idmap=container
lxc.cap.drop = sys_module mknod setuid net_raw
lxc.cap.drop = mac_override
</programlisting>
</refsect2>
</refsect1>
<refsect1>
<title>See Also</title>
<simpara>
<citerefentry>
<refentrytitle><command>chroot</command></refentrytitle>
<manvolnum>1</manvolnum>
</citerefentry>,
<citerefentry>
<refentrytitle><command>pivot_root</command></refentrytitle>
<manvolnum>8</manvolnum>
</citerefentry>,
<citerefentry>
<refentrytitle><filename>fstab</filename></refentrytitle>
<manvolnum>5</manvolnum>
</citerefentry>,
<citerefentry>
<refentrytitle><filename>capabilities</filename></refentrytitle>
<manvolnum>7</manvolnum>
</citerefentry>
</simpara>
</refsect1>
&seealso;
</refentry>
<!-- Keep this comment at the end of the file
Local variables:
mode: sgml
sgml-omittag:t
sgml-shorttag:t
sgml-minimize-attributes:nil
sgml-always-quote-attributes:t
sgml-indent-step:2
sgml-indent-data:t
sgml-parent-document:nil
sgml-default-dtd-file:nil
sgml-exposed-tags:nil
sgml-local-catalogs:nil
sgml-local-ecat-files:nil
End:
-->
Reference in New Issue View Git Blame Copy Permalink