1
0
mirror of https://github.com/systemd/systemd.git synced 2024-11-01 17:51:22 +03:00
systemd/man/sd_id128_get_machine.xml

171 lines
8.4 KiB
XML
Raw Normal View History

<?xml version='1.0'?> <!--*-nxml-*-->
<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
"http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
<!--
SPDX-License-Identifier: LGPL-2.1+
-->
<refentry id="sd_id128_get_machine" xmlns:xi="http://www.w3.org/2001/XInclude">
2015-02-04 05:14:13 +03:00
<refentryinfo>
<title>sd_id128_get_machine</title>
<productname>systemd</productname>
</refentryinfo>
<refmeta>
<refentrytitle>sd_id128_get_machine</refentrytitle>
<manvolnum>3</manvolnum>
</refmeta>
<refnamediv>
<refname>sd_id128_get_machine</refname>
<refname>sd_id128_get_machine_app_specific</refname>
2015-02-04 05:14:13 +03:00
<refname>sd_id128_get_boot</refname>
<refname>sd_id128_get_boot_app_specific</refname>
core: add "invocation ID" concept to service manager This adds a new invocation ID concept to the service manager. The invocation ID identifies each runtime cycle of a unit uniquely. A new randomized 128bit ID is generated each time a unit moves from and inactive to an activating or active state. The primary usecase for this concept is to connect the runtime data PID 1 maintains about a service with the offline data the journal stores about it. Previously we'd use the unit name plus start/stop times, which however is highly racy since the journal will generally process log data after the service already ended. The "invocation ID" kinda matches the "boot ID" concept of the Linux kernel, except that it applies to an individual unit instead of the whole system. The invocation ID is passed to the activated processes as environment variable. It is additionally stored as extended attribute on the cgroup of the unit. The latter is used by journald to automatically retrieve it for each log logged message and attach it to the log entry. The environment variable is very easily accessible, even for unprivileged services. OTOH the extended attribute is only accessible to privileged processes (this is because cgroupfs only supports the "trusted." xattr namespace, not "user."). The environment variable may be altered by services, the extended attribute may not be, hence is the better choice for the journal. Note that reading the invocation ID off the extended attribute from journald is racy, similar to the way reading the unit name for a logging process is. This patch adds APIs to read the invocation ID to sd-id128: sd_id128_get_invocation() may be used in a similar fashion to sd_id128_get_boot(). PID1's own logging is updated to always include the invocation ID when it logs information about a unit. A new bus call GetUnitByInvocationID() is added that allows retrieving a bus path to a unit by its invocation ID. The bus path is built using the invocation ID, thus providing a path for referring to a unit that is valid only for the current runtime cycleof it. Outlook for the future: should the kernel eventually allow passing of cgroup information along AF_UNIX/SOCK_DGRAM messages via a unique cgroup id, then we can alter the invocation ID to be generated as hash from that rather than entirely randomly. This way we can derive the invocation race-freely from the messages.
2016-08-31 00:18:46 +03:00
<refname>sd_id128_get_invocation</refname>
2015-02-04 05:14:13 +03:00
<refpurpose>Retrieve 128-bit IDs</refpurpose>
</refnamediv>
<refsynopsisdiv>
<funcsynopsis>
<funcsynopsisinfo>#include &lt;systemd/sd-id128.h&gt;</funcsynopsisinfo>
<funcprototype>
<funcdef>int <function>sd_id128_get_machine</function></funcdef>
<paramdef>sd_id128_t *<parameter>ret</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>int <function>sd_id128_get_machine_app_specific</function></funcdef>
<paramdef>sd_id128_t <parameter>app_id</parameter></paramdef>
<paramdef>sd_id128_t *<parameter>ret</parameter></paramdef>
</funcprototype>
2015-02-04 05:14:13 +03:00
<funcprototype>
<funcdef>int <function>sd_id128_get_boot</function></funcdef>
<paramdef>sd_id128_t *<parameter>ret</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>int <function>sd_id128_get_boot_app_specific</function></funcdef>
<paramdef>sd_id128_t <parameter>app_id</parameter></paramdef>
<paramdef>sd_id128_t *<parameter>ret</parameter></paramdef>
</funcprototype>
core: add "invocation ID" concept to service manager This adds a new invocation ID concept to the service manager. The invocation ID identifies each runtime cycle of a unit uniquely. A new randomized 128bit ID is generated each time a unit moves from and inactive to an activating or active state. The primary usecase for this concept is to connect the runtime data PID 1 maintains about a service with the offline data the journal stores about it. Previously we'd use the unit name plus start/stop times, which however is highly racy since the journal will generally process log data after the service already ended. The "invocation ID" kinda matches the "boot ID" concept of the Linux kernel, except that it applies to an individual unit instead of the whole system. The invocation ID is passed to the activated processes as environment variable. It is additionally stored as extended attribute on the cgroup of the unit. The latter is used by journald to automatically retrieve it for each log logged message and attach it to the log entry. The environment variable is very easily accessible, even for unprivileged services. OTOH the extended attribute is only accessible to privileged processes (this is because cgroupfs only supports the "trusted." xattr namespace, not "user."). The environment variable may be altered by services, the extended attribute may not be, hence is the better choice for the journal. Note that reading the invocation ID off the extended attribute from journald is racy, similar to the way reading the unit name for a logging process is. This patch adds APIs to read the invocation ID to sd-id128: sd_id128_get_invocation() may be used in a similar fashion to sd_id128_get_boot(). PID1's own logging is updated to always include the invocation ID when it logs information about a unit. A new bus call GetUnitByInvocationID() is added that allows retrieving a bus path to a unit by its invocation ID. The bus path is built using the invocation ID, thus providing a path for referring to a unit that is valid only for the current runtime cycleof it. Outlook for the future: should the kernel eventually allow passing of cgroup information along AF_UNIX/SOCK_DGRAM messages via a unique cgroup id, then we can alter the invocation ID to be generated as hash from that rather than entirely randomly. This way we can derive the invocation race-freely from the messages.
2016-08-31 00:18:46 +03:00
<funcprototype>
<funcdef>int <function>sd_id128_get_invocation</function></funcdef>
<paramdef>sd_id128_t *<parameter>ret</parameter></paramdef>
</funcprototype>
2015-02-04 05:14:13 +03:00
</funcsynopsis>
</refsynopsisdiv>
<refsect1>
<title>Description</title>
<para><function>sd_id128_get_machine()</function> returns the machine ID of the executing host. This reads and
parses the <citerefentry><refentrytitle>machine-id</refentrytitle><manvolnum>5</manvolnum></citerefentry>
file. This function caches the machine ID internally to make retrieving the machine ID a cheap operation. This ID
may be used wherever a unique identifier for the local system is needed. However, it is recommended to use this ID
as-is only in trusted environments. In untrusted environments it is recommended to derive an application specific
ID from this machine ID, in an irreversable (cryptographically secure) way. To make this easy
<function>sd_id128_get_machine_app_specific()</function> is provided, see below.</para>
<para><function>sd_id128_get_machine_app_specific()</function> is similar to
<function>sd_id128_get_machine()</function>, but retrieves a machine ID that is specific to the application that is
identified by the indicated application ID. It is recommended to use this function instead of
<function>sd_id128_get_machine()</function> when passing an ID to untrusted environments, in order to make sure
that the original machine ID may not be determined externally. This way, the ID used by the application remains
stable on a given machine, but cannot be easily correlated with IDs used in other applications on the same
machine. The application-specific ID should be generated via a tool like <command>systemd-id128 new</command>,
and may be compiled into the application. This function will return the same application-specific ID for each
combination of machine ID and application ID. Internally, this function calculates HMAC-SHA256 of the application
ID, keyed by the machine ID.</para>
<para><function>sd_id128_get_boot()</function> returns the boot ID of the executing kernel. This reads and parses
the <filename>/proc/sys/kernel/random/boot_id</filename> file exposed by the kernel. It is randomly generated early
at boot and is unique for every running kernel instance. See <citerefentry
project='man-pages'><refentrytitle>random</refentrytitle><manvolnum>4</manvolnum></citerefentry> for more
information. This function also internally caches the returned ID to make this call a cheap operation. It is
recommended to use this ID as-is only in trusted environments. In untrusted environments it is recommended to
derive an application specific ID using <function>sd_id128_get_machine_app_specific()</function>, see below.</para>
<para><function>sd_id128_get_boot_app_specific()</function> is analogous to
<function>sd_id128_get_machine_app_specific()</function> but returns an ID that changes between boots. Some
machines may be used for a long time without rebooting, hence the boot ID may remain constant for a long time, and
has properties similar to the machine ID during that time.</para>
2015-02-04 05:14:13 +03:00
core: add "invocation ID" concept to service manager This adds a new invocation ID concept to the service manager. The invocation ID identifies each runtime cycle of a unit uniquely. A new randomized 128bit ID is generated each time a unit moves from and inactive to an activating or active state. The primary usecase for this concept is to connect the runtime data PID 1 maintains about a service with the offline data the journal stores about it. Previously we'd use the unit name plus start/stop times, which however is highly racy since the journal will generally process log data after the service already ended. The "invocation ID" kinda matches the "boot ID" concept of the Linux kernel, except that it applies to an individual unit instead of the whole system. The invocation ID is passed to the activated processes as environment variable. It is additionally stored as extended attribute on the cgroup of the unit. The latter is used by journald to automatically retrieve it for each log logged message and attach it to the log entry. The environment variable is very easily accessible, even for unprivileged services. OTOH the extended attribute is only accessible to privileged processes (this is because cgroupfs only supports the "trusted." xattr namespace, not "user."). The environment variable may be altered by services, the extended attribute may not be, hence is the better choice for the journal. Note that reading the invocation ID off the extended attribute from journald is racy, similar to the way reading the unit name for a logging process is. This patch adds APIs to read the invocation ID to sd-id128: sd_id128_get_invocation() may be used in a similar fashion to sd_id128_get_boot(). PID1's own logging is updated to always include the invocation ID when it logs information about a unit. A new bus call GetUnitByInvocationID() is added that allows retrieving a bus path to a unit by its invocation ID. The bus path is built using the invocation ID, thus providing a path for referring to a unit that is valid only for the current runtime cycleof it. Outlook for the future: should the kernel eventually allow passing of cgroup information along AF_UNIX/SOCK_DGRAM messages via a unique cgroup id, then we can alter the invocation ID to be generated as hash from that rather than entirely randomly. This way we can derive the invocation race-freely from the messages.
2016-08-31 00:18:46 +03:00
<para><function>sd_id128_get_invocation()</function> returns the invocation ID of the currently executed
service. In its current implementation, this reads and parses the <varname>$INVOCATION_ID</varname> environment
variable that the service manager sets when activating a service, see
<citerefentry><refentrytitle>systemd.exec</refentrytitle><manvolnum>5</manvolnum></citerefentry> for details. The
ID is cached internally. In future a different mechanism to determine the invocation ID may be added.</para>
<para>Note that <function>sd_id128_get_machine_app_specific()</function>, <function>sd_id128_get_boot()</function>,
<function>sd_id128_get_boot_app_specific()</function>, and <function>sd_id128_get_invocation()</function> always
return UUID v4 compatible IDs. <function>sd_id128_get_machine()</function> will also return a UUID v4-compatible
ID on new installations but might not on older. It is possible to convert the machine ID into a UUID v4-compatible
one. For more information, see
2015-02-04 05:14:13 +03:00
<citerefentry><refentrytitle>machine-id</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para>
<para>For more information about the <literal>sd_id128_t</literal>
type see
<citerefentry><refentrytitle>sd-id128</refentrytitle><manvolnum>3</manvolnum></citerefentry>.</para>
</refsect1>
<refsect1>
<title>Return Value</title>
<para>Those calls return 0 on success (in which case <parameter>ret</parameter> is filled in),
or a negative errno-style error code. In particular,
<function>sd_id128_get_machine()</function>,
<function>sd_id128_get_machine_app_specific()</function>, and
<function>sd_id128_get_boot_app_specific()</function> return <constant>-ENOENT</constant> if
<filename>/etc/machine-id</filename> is missing, and <constant>-ENOMEDIUM</constant> if
<filename>/etc/machine-id</filename> is empty or all zeros.</para>
2015-02-04 05:14:13 +03:00
</refsect1>
<xi:include href="libsystemd-pkgconfig.xml" />
2015-02-04 05:14:13 +03:00
<refsect1>
<title>Examples</title>
<example>
<title>Application-specific machine ID</title>
2018-08-21 17:25:21 +03:00
<para>First, generate the application ID:</para>
<programlisting>$ systemd-id128 -p new
As string:
c273277323db454ea63bb96e79b53e97
As UUID:
c2732773-23db-454e-a63b-b96e79b53e97
As man:sd-id128(3) macro:
#define MESSAGE_XYZ SD_ID128_MAKE(c2,73,27,73,23,db,45,4e,a6,3b,b9,6e,79,b5,3e,97)
...
</programlisting>
<para>Then use the new identifier in an example application:</para>
<programlisting><xi:include href="id128-app-specific.c" parse="text" /></programlisting>
</example>
</refsect1>
2015-02-04 05:14:13 +03:00
<refsect1>
<title>See Also</title>
<para>
<citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
<citerefentry><refentrytitle>systemd-id128</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
2015-02-04 05:14:13 +03:00
<citerefentry><refentrytitle>sd-id128</refentrytitle><manvolnum>3</manvolnum></citerefentry>,
<citerefentry><refentrytitle>machine-id</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
core: add "invocation ID" concept to service manager This adds a new invocation ID concept to the service manager. The invocation ID identifies each runtime cycle of a unit uniquely. A new randomized 128bit ID is generated each time a unit moves from and inactive to an activating or active state. The primary usecase for this concept is to connect the runtime data PID 1 maintains about a service with the offline data the journal stores about it. Previously we'd use the unit name plus start/stop times, which however is highly racy since the journal will generally process log data after the service already ended. The "invocation ID" kinda matches the "boot ID" concept of the Linux kernel, except that it applies to an individual unit instead of the whole system. The invocation ID is passed to the activated processes as environment variable. It is additionally stored as extended attribute on the cgroup of the unit. The latter is used by journald to automatically retrieve it for each log logged message and attach it to the log entry. The environment variable is very easily accessible, even for unprivileged services. OTOH the extended attribute is only accessible to privileged processes (this is because cgroupfs only supports the "trusted." xattr namespace, not "user."). The environment variable may be altered by services, the extended attribute may not be, hence is the better choice for the journal. Note that reading the invocation ID off the extended attribute from journald is racy, similar to the way reading the unit name for a logging process is. This patch adds APIs to read the invocation ID to sd-id128: sd_id128_get_invocation() may be used in a similar fashion to sd_id128_get_boot(). PID1's own logging is updated to always include the invocation ID when it logs information about a unit. A new bus call GetUnitByInvocationID() is added that allows retrieving a bus path to a unit by its invocation ID. The bus path is built using the invocation ID, thus providing a path for referring to a unit that is valid only for the current runtime cycleof it. Outlook for the future: should the kernel eventually allow passing of cgroup information along AF_UNIX/SOCK_DGRAM messages via a unique cgroup id, then we can alter the invocation ID to be generated as hash from that rather than entirely randomly. This way we can derive the invocation race-freely from the messages.
2016-08-31 00:18:46 +03:00
<citerefentry><refentrytitle>systemd.exec</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
<citerefentry><refentrytitle>sd_id128_randomize</refentrytitle><manvolnum>3</manvolnum></citerefentry>,
<citerefentry project='man-pages'><refentrytitle>random</refentrytitle><manvolnum>4</manvolnum></citerefentry>
2015-02-04 05:14:13 +03:00
</para>
</refsect1>
</refentry>