1
1
mirror of https://github.com/systemd/systemd-stable.git synced 2024-12-25 23:21:33 +03:00
systemd-stable/man/coredumpctl.xml
Zbigniew Jędrzejewski-Szmek 4f57f77267 man: make systemd-coredump and coredumpctl descriptions more accessible
Fixes #17910: we didn't clearly explain that coredumps may exist without
journal entries, and vice versa.

Also, make the examples more concrete, and use '$' instead of '#' to avoid
suggesting that running as root is required. The text is extended a bit in
various places. In the description of systemd-coredump, the details of executor
separation are split out to a separate subsection, since they are rather
detailed and not necessary to understand for normal use.
2021-02-28 11:29:21 +01:00

382 lines
14 KiB
XML

<?xml version='1.0'?> <!--*-nxml-*-->
<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN"
"http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
<!-- SPDX-License-Identifier: LGPL-2.1-or-later -->
<refentry id="coredumpctl" conditional='ENABLE_COREDUMP'
xmlns:xi="http://www.w3.org/2001/XInclude">
<refentryinfo>
<title>coredumpctl</title>
<productname>systemd</productname>
</refentryinfo>
<refmeta>
<refentrytitle>coredumpctl</refentrytitle>
<manvolnum>1</manvolnum>
</refmeta>
<refnamediv>
<refname>coredumpctl</refname>
<refpurpose>Retrieve and process saved core dumps and metadata</refpurpose>
</refnamediv>
<refsynopsisdiv>
<cmdsynopsis>
<command>coredumpctl</command>
<arg choice="opt" rep="repeat">OPTIONS</arg>
<arg choice="req">COMMAND</arg>
<arg choice="opt" rep="repeat">PID|COMM|EXE|MATCH</arg>
</cmdsynopsis>
</refsynopsisdiv>
<refsect1>
<title>Description</title>
<para><command>coredumpctl</command> is a tool that can be used to retrieve and process core
dumps and metadata which were saved by
<citerefentry><refentrytitle>systemd-coredump</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
</para>
</refsect1>
<refsect1>
<title>Commands</title>
<para>The following commands are understood:</para>
<variablelist>
<varlistentry>
<term><command>list</command></term>
<listitem><para>List core dumps captured in the journal
matching specified characteristics. If no command is
specified, this is the implied default.</para>
<para>The output is designed to be human readable and contains a table with the following
columns:</para>
<variablelist>
<varlistentry>
<term>TIME</term>
<listitem><para>The timestamp of the crash, as reported by the kernel.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>PID</term>
<listitem><para>The identifier of the process that crashed.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>UID</term>
<term>GID</term>
<listitem><para>The user and group identifiers of the process that crashed.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>SIGNAL</term>
<listitem><para>The signal that caused the process to crash, when applicable.
</para></listitem>
</varlistentry>
<varlistentry>
<term>COREFILE</term>
<listitem><para>Information whether the coredump was stored, and whether
it is still accessible: <literal>none</literal> means the core was
not stored, <literal>-</literal> means that it was not available (for
example because the process was not terminated by a signal),
<literal>present</literal> means that the core file is accessible by the
current user, <literal>journal</literal> means that the core was stored
in the <literal>journal</literal>, <literal>truncated</literal> is the
same as one of the previous two, but the core was too large and was not
stored in its entirety, <literal>error</literal> means that the core file
cannot be accessed, most likely because of insufficient permissions, and
<literal>missing</literal> means that the core was stored in a file, but
this file has since been removed.</para></listitem>
</varlistentry>
<varlistentry>
<term>EXE</term>
<listitem><para>The full path to the executable. For backtraces of scripts
this is the name of the interpreter.</para></listitem>
</varlistentry>
</variablelist>
<para>It's worth noting that different restrictions apply to
data saved in the journal and core dump files saved in
<filename>/var/lib/systemd/coredump</filename>, see overview in
<citerefentry><refentrytitle>systemd-coredump</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
Thus it may very well happen that a particular core dump is still listed
in the journal while its corresponding core dump file has already been
removed.</para></listitem>
</varlistentry>
<varlistentry>
<term><command>info</command></term>
<listitem><para>Show detailed information about the last core dump
or core dumps matching specified characteristics
captured in the journal.</para></listitem>
</varlistentry>
<varlistentry>
<term><command>dump</command></term>
<listitem><para>Extract the last core dump matching specified
characteristics. The core dump will be written on standard
output, unless an output file is specified with
<option>--output=</option>. </para></listitem>
</varlistentry>
<varlistentry>
<term><command>debug</command></term>
<listitem><para>Invoke a debugger on the last core dump
matching specified characteristics. By default,
<citerefentry project='man-pages'><refentrytitle>gdb</refentrytitle><manvolnum>1</manvolnum></citerefentry>
will be used. This may be changed using the <option>--debugger=</option>
option or the <varname>$SYSTEMD_DEBUGGER</varname> environment
variable. Use the <option>--debugger-arguments=</option> option to pass extra
command line arguments to the debugger.</para></listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1>
<title>Options</title>
<para>The following options are understood:</para>
<variablelist>
<xi:include href="standard-options.xml" xpointer="help" />
<xi:include href="standard-options.xml" xpointer="version" />
<xi:include href="standard-options.xml" xpointer="no-pager" />
<xi:include href="standard-options.xml" xpointer="no-legend" />
<xi:include href="standard-options.xml" xpointer="json" />
<varlistentry>
<term><option>-1</option></term>
<listitem><para>Show information of the most recent core dump only, instead of listing all known core
dumps. (Equivalent to <option>--reverse -n 1</option></para></listitem>
</varlistentry>
<varlistentry>
<term><option>-n</option> <replaceable>INT</replaceable></term>
<listitem><para>Show at most the specified number of entries. The specified parameter must be an
integer greater or equal to 1.</para></listitem>
</varlistentry>
<varlistentry>
<term><option>-S</option></term>
<term><option>--since</option></term>
<listitem><para>Only print entries which are since the specified date.</para></listitem>
</varlistentry>
<varlistentry>
<term><option>-U</option></term>
<term><option>--until</option></term>
<listitem><para>Only print entries which are until the specified date.</para></listitem>
</varlistentry>
<varlistentry>
<term><option>-r</option></term>
<term><option>--reverse</option></term>
<listitem><para>Reverse output so that the newest entries are displayed first.
</para></listitem>
</varlistentry>
<varlistentry>
<term><option>-F</option> <replaceable>FIELD</replaceable></term>
<term><option>--field=</option><replaceable>FIELD</replaceable></term>
<listitem><para>Print all possible data values the specified
field takes in matching core dump entries of the
journal.</para></listitem>
</varlistentry>
<varlistentry>
<term><option>-o</option> <replaceable>FILE</replaceable></term>
<term><option>--output=</option><replaceable>FILE</replaceable></term>
<listitem><para>Write the core to <option>FILE</option>.
</para></listitem>
</varlistentry>
<varlistentry>
<term><option>--debugger=</option><replaceable>DEBUGGER</replaceable></term>
<listitem><para>Use the given debugger for the <command>debug</command>
command. If not given and <varname>$SYSTEMD_DEBUGGER</varname> is unset, then
<citerefentry project='man-pages'><refentrytitle>gdb</refentrytitle><manvolnum>1</manvolnum></citerefentry>
will be used. </para></listitem>
</varlistentry>
<varlistentry>
<term><option>-A</option> <replaceable>ARGS</replaceable></term>
<term><option>--debugger-arguments=</option><replaceable>ARGS</replaceable></term>
<listitem><para>Pass the given <replaceable>ARGS</replaceable> as extra command line arguments
to the debugger. Quote as appropriate when <replaceable>ARGS</replaceable> contain whitespace.
(See Examples.)</para></listitem>
</varlistentry>
<varlistentry>
<term><option>--file=<replaceable>GLOB</replaceable></option></term>
<listitem><para>Takes a file glob as an argument. If
specified, coredumpctl will operate on the specified journal
files matching <replaceable>GLOB</replaceable> instead of the
default runtime and system journal paths. May be specified
multiple times, in which case files will be suitably
interleaved.</para></listitem>
</varlistentry>
<varlistentry>
<term><option>-D</option> <replaceable>DIR</replaceable></term>
<term><option>--directory=</option><replaceable>DIR</replaceable></term>
<listitem><para>Use the journal files in the specified <option>DIR</option>.
</para></listitem>
</varlistentry>
<varlistentry>
<term><option>-q</option></term>
<term><option>--quiet</option></term>
<listitem><para>Suppresses informational messages about lack
of access to journal files and possible in-flight coredumps.
</para></listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1>
<title>Matching</title>
<para>A match can be:</para>
<variablelist>
<varlistentry>
<term><replaceable>PID</replaceable></term>
<listitem><para>Process ID of the
process that dumped
core. An integer.</para></listitem>
</varlistentry>
<varlistentry>
<term><replaceable>COMM</replaceable></term>
<listitem><para>Name of the executable (matches
<option>COREDUMP_COMM=</option>). Must not contain slashes.
</para></listitem>
</varlistentry>
<varlistentry>
<term><replaceable>EXE</replaceable></term>
<listitem><para>Path to the executable (matches
<option>COREDUMP_EXE=</option>). Must contain at least one
slash. </para></listitem>
</varlistentry>
<varlistentry>
<term><replaceable>MATCH</replaceable></term>
<listitem><para>General journalctl match filter, must contain an equals
sign (<literal>=</literal>). See
<citerefentry><refentrytitle>journalctl</refentrytitle><manvolnum>1</manvolnum></citerefentry>.
</para></listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1>
<title>Exit status</title>
<para>On success, 0 is returned; otherwise, a non-zero failure
code is returned. Not finding any matching core dumps is treated as
failure.
</para>
</refsect1>
<refsect1>
<title>Environment</title>
<variablelist class='environment-variables'>
<varlistentry>
<term><varname>$SYSTEMD_DEBUGGER</varname></term>
<listitem><para>Use the given debugger for the <command>debug</command>
command. See the <option>--debugger=</option> option.</para></listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1>
<title>Examples</title>
<example>
<title>List all the core dumps of a program</title>
<programlisting>$ coredumpctl list /usr/lib64/firefox/firefox
TIME PID UID GID SIG COREFILE EXE SIZE
Tue … 8018 1000 1000 SIGSEGV missing /usr/lib64/firefox/firefox n/a
Wed … 251609 1000 1000 SIGTRAP missing /usr/lib64/firefox/firefox n/a
Fri … 552351 1000 1000 SIGSEGV present /usr/lib64/firefox/firefox 28.7M
</programlisting>
<para>The journal has three entries pertaining to <filename>/usr/lib64/firefox/firefox</filename>, and
only the last entry still has an available core file (in external storage on disk).</para>
<para>Note that <filename>coredumpctl</filename> needs access to the journal files to retrieve the
relevant entries from the journal. Thus, an unprivileged user will normally only see information about
crashing programs of this user.</para>
</example>
<example>
<title>Invoke <command>gdb</command> on the last core dump</title>
<programlisting>$ coredumpctl debug</programlisting>
</example>
<example>
<title>Use <command>gdb</command> to display full register info from the last core dump</title>
<programlisting>$ coredumpctl debug --debugger-arguments="-batch -ex 'info all-registers'"</programlisting>
</example>
<example>
<title>Show information about a process that dumped core,
matching by its PID 6654</title>
<programlisting>$ coredumpctl info 6654</programlisting>
</example>
<example>
<title>Extract the last core dump of /usr/bin/bar to a file named
<filename index="false">bar.coredump</filename></title>
<programlisting>$ coredumpctl -o bar.coredump dump /usr/bin/bar</programlisting>
</example>
</refsect1>
<refsect1>
<title>See Also</title>
<para>
<citerefentry><refentrytitle>systemd-coredump</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
<citerefentry><refentrytitle>coredump.conf</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
<citerefentry><refentrytitle>systemd-journald.service</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
<citerefentry project='man-pages'><refentrytitle>gdb</refentrytitle><manvolnum>1</manvolnum></citerefentry>
</para>
</refsect1>
</refentry>