The default action of certain signals is to cause a process to
terminate and produce a core dump file, a file containing an
image of the process's memory at the time of termination. This
image can be used in a debugger (e.g., gdb(1)) to inspect the
state of the program at the time that it terminated. A list of
the signals which cause a process to dump core can be found in
signal(7).
A process can set its soft RLIMIT_CORE resource limit to place an
upper limit on the size of the core dump file that will be
produced if it receives a "core dump" signal; see getrlimit(2)
for details.
There are various circumstances in which a core dump file is not
produced:
* The process does not have permission to write the core file.
(By default, the core file is called core or core.pid, where
pid is the ID of the process that dumped core, and is created
in the current working directory. See below for details on
naming.) Writing the core file fails if the directory in
which it is to be created is not writable, or if a file with
the same name exists and is not writable or is not a regular
file (e.g., it is a directory or a symbolic link).
* A (writable, regular) file with the same name as would be used
for the core dump already exists, but there is more than one
hard link to that file.
* The filesystem where the core dump file would be created is
full; or has run out of inodes; or is mounted read-only; or
the user has reached their quota for the filesystem.
* The directory in which the core dump file is to be created
does not exist.
* The RLIMIT_CORE (core file size) or RLIMIT_FSIZE (file size)
resource limits for the process are set to zero; see
getrlimit(2) and the documentation of the shell's ulimit
command (limit in csh(1)).
* The binary being executed by the process does not have read
permission enabled. (This is a security measure to ensure
that an executable whose contents are not readable does not
produce a—possibly readable—core dump containing an image of
the executable.)
* The process is executing a set-user-ID (set-group-ID) program
that is owned by a user (group) other than the real user
(group) ID of the process, or the process is executing a
program that has file capabilities (see capabilities(7)).
(However, see the description of the prctl(2) PR_SET_DUMPABLE
operation, and the description of the
/proc/sys/fs/suid_dumpable file in proc(5).)
* /proc/sys/kernel/core_pattern is empty and
/proc/sys/kernel/core_uses_pid contains the value 0. (These
files are described below.) Note that if
/proc/sys/kernel/core_pattern is empty and
/proc/sys/kernel/core_uses_pid contains the value 1, core dump
files will have names of the form .pid, and such files are
hidden unless one uses the ls(1) -a option.
* (Since Linux 3.7) The kernel was configured without the
CONFIG_COREDUMP option.
In addition, a core dump may exclude part of the address space of
the process if the madvise(2) MADV_DONTDUMP flag was employed.
On systems that employ systemd(1) as the init framework, core
dumps may instead be placed in a location determined by
systemd(1). See below for further details.
Naming of core dump files
By default, a core dump file is named core, but the
/proc/sys/kernel/core_pattern file (since Linux 2.6 and 2.4.21)
can be set to define a template that is used to name core dump
files. The template can contain % specifiers which are
substituted by the following values when a core file is created:
%% A single % character.
%c Core file size soft resource limit of crashing process
(since Linux 2.6.24).
%d Dump mode—same as value returned by prctl(2)
PR_GET_DUMPABLE (since Linux 3.7).
%e The process or thread's comm value, which typically is
the same as the executable filename (without path prefix,
and truncated to a maximum of 15 characters), but may
have been modified to be something different; see the
discussion of /proc/[pid]/comm and
/proc/[pid]/task/[tid]/comm in proc(5).
%E Pathname of executable, with slashes ('/') replaced by
exclamation marks ('!') (since Linux 3.0).
%g Numeric real GID of dumped process.
%h Hostname (same as nodename returned by uname(2)).
%i TID of thread that triggered core dump, as seen in the
PID namespace in which the thread resides (since Linux
3.18).
%I TID of thread that triggered core dump, as seen in the
initial PID namespace (since Linux 3.18).
%p PID of dumped process, as seen in the PID namespace in
which the process resides.
%P PID of dumped process, as seen in the initial PID
namespace (since Linux 3.12).
%s Number of signal causing dump.
%t Time of dump, expressed as seconds since the Epoch,
1970-01-01 00:00:00 +0000 (UTC).
%u Numeric real UID of dumped process.
A single % at the end of the template is dropped from the core
filename, as is the combination of a % followed by any character
other than those listed above. All other characters in the
template become a literal part of the core filename. The
template may include '/' characters, which are interpreted as
delimiters for directory names. The maximum size of the
resulting core filename is 128 bytes (64 bytes in kernels before
2.6.19). The default value in this file is "core". For backward
compatibility, if /proc/sys/kernel/core_pattern does not include
%p and /proc/sys/kernel/core_uses_pid (see below) is nonzero,
then .PID will be appended to the core filename.
Paths are interpreted according to the settings that are active
for the crashing process. That means the crashing process's
mount namespace (see mount_namespaces(7)), its current working
directory (found via getcwd(2)), and its root directory (see
chroot(2)).
Since version 2.4, Linux has also provided a more primitive
method of controlling the name of the core dump file. If the
/proc/sys/kernel/core_uses_pid file contains the value 0, then a
core dump file is simply named core. If this file contains a
nonzero value, then the core dump file includes the process ID in
a name of the form core.PID.
Since Linux 3.6, if /proc/sys/fs/suid_dumpable is set to 2
("suidsafe"), the pattern must be either an absolute pathname
(starting with a leading '/' character) or a pipe, as defined
below.
Piping core dumps to a program
Since kernel 2.6.19, Linux supports an alternate syntax for the
/proc/sys/kernel/core_pattern file. If the first character of
this file is a pipe symbol (|), then the remainder of the line is
interpreted as the command-line for a user-space program (or
script) that is to be executed.
Since kernel 5.3.0, the pipe template is split on spaces into an
argument list before the template parameters are expanded. In
earlier kernels, the template parameters are expanded first and
the resulting string is split on spaces into an argument list.
This means that in earlier kernels executable names added by the
%e and %E template parameters could get split into multiple
arguments. So the core dump handler needs to put the executable
names as the last argument and ensure it joins all parts of the
executable name using spaces. Executable names with multiple
spaces in them are not correctly represented in earlier kernels,
meaning that the core dump handler needs to use mechanisms to
find the executable name.
Instead of being written to a file, the core dump is given as
standard input to the program. Note the following points:
* The program must be specified using an absolute pathname (or a
pathname relative to the root directory, /), and must
immediately follow the '|' character.
* The command-line arguments can include any of the % specifiers
listed above. For example, to pass the PID of the process
that is being dumped, specify %p in an argument.
* The process created to run the program runs as user and group
root.
* Running as root does not confer any exceptional security
bypasses. Namely, LSMs (e.g., SELinux) are still active and
may prevent the handler from accessing details about the
crashed process via /proc/[pid].
* The program pathname is interpreted with respect to the
initial mount namespace as it is always executed there. It is
not affected by the settings (e.g., root directory, mount
namespace, current working directory) of the crashing process.
* The process runs in the initial namespaces (PID, mount, user,
and so on) and not in the namespaces of the crashing process.
One can utilize specifiers such as %P to find the right
/proc/[pid] directory and probe/enter the crashing process's
namespaces if needed.
* The process starts with its current working directory as the
root directory. If desired, it is possible change to the
working directory of the dumping process by employing the
value provided by the %P specifier to change to the location
of the dumping process via /proc/[pid]/cwd.
* Command-line arguments can be supplied to the program (since
Linux 2.6.24), delimited by white space (up to a total line
length of 128 bytes).
* The RLIMIT_CORE limit is not enforced for core dumps that are
piped to a program via this mechanism.
/proc/sys/kernel/core_pipe_limit
When collecting core dumps via a pipe to a user-space program, it
can be useful for the collecting program to gather data about the
crashing process from that process's /proc/[pid] directory. In
order to do this safely, the kernel must wait for the program
collecting the core dump to exit, so as not to remove the
crashing process's /proc/[pid] files prematurely. This in turn
creates the possibility that a misbehaving collecting program can
block the reaping of a crashed process by simply never exiting.
Since Linux 2.6.32, the /proc/sys/kernel/core_pipe_limit can be
used to defend against this possibility. The value in this file
defines how many concurrent crashing processes may be piped to
user-space programs in parallel. If this value is exceeded, then
those crashing processes above this value are noted in the kernel
log and their core dumps are skipped.
A value of 0 in this file is special. It indicates that
unlimited processes may be captured in parallel, but that no
waiting will take place (i.e., the collecting program is not
guaranteed access to /proc/<crashing-PID>). The default value
for this file is 0.
Controlling which mappings are written to the core dump
Since kernel 2.6.23, the Linux-specific
/proc/[pid]/coredump_filter file can be used to control which
memory segments are written to the core dump file in the event
that a core dump is performed for the process with the
corresponding process ID.
The value in the file is a bit mask of memory mapping types (see
mmap(2)). If a bit is set in the mask, then memory mappings of
the corresponding type are dumped; otherwise they are not dumped.
The bits in this file have the following meanings:
bit 0 Dump anonymous private mappings.
bit 1 Dump anonymous shared mappings.
bit 2 Dump file-backed private mappings.
bit 3 Dump file-backed shared mappings.
bit 4 (since Linux 2.6.24)
Dump ELF headers.
bit 5 (since Linux 2.6.28)
Dump private huge pages.
bit 6 (since Linux 2.6.28)
Dump shared huge pages.
bit 7 (since Linux 4.4)
Dump private DAX pages.
bit 8 (since Linux 4.4)
Dump shared DAX pages.
By default, the following bits are set: 0, 1, 4 (if the
CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS kernel configuration option
is enabled), and 5. This default can be modified at boot time
using the coredump_filter boot option.
The value of this file is displayed in hexadecimal. (The default
value is thus displayed as 33.)
Memory-mapped I/O pages such as frame buffer are never dumped,
and virtual DSO (vdso(7)) pages are always dumped, regardless of
the coredump_filter value.
A child process created via fork(2) inherits its parent's
coredump_filter value; the coredump_filter value is preserved
across an execve(2).
It can be useful to set coredump_filter in the parent shell
before running a program, for example:
$ echo 0x7 > /proc/self/coredump_filter
$ ./some_program
This file is provided only if the kernel was built with the
CONFIG_ELF_CORE configuration option.
Core dumps and systemd
On systems using the systemd(1) init framework, core dumps may be
placed in a location determined by systemd(1). To do this,
systemd(1) employs the core_pattern feature that allows piping
core dumps to a program. One can verify this by checking whether
core dumps are being piped to the systemd-coredump(8) program:
$ cat /proc/sys/kernel/core_pattern
|/usr/lib/systemd/systemd-coredump %P %u %g %s %t %c %e
In this case, core dumps will be placed in the location
configured for systemd-coredump(8), typically as lz4(1)
compressed files in the directory /var/lib/systemd/coredump/.
One can list the core dumps that have been recorded by
systemd-coredump(8) using coredumpctl(1):
$ coredumpctl list | tail -5
Wed 2017-10-11 22:25:30 CEST 2748 1000 1000 3 present /usr/bin/sleep
Thu 2017-10-12 06:29:10 CEST 2716 1000 1000 3 present /usr/bin/sleep
Thu 2017-10-12 06:30:50 CEST 2767 1000 1000 3 present /usr/bin/sleep
Thu 2017-10-12 06:37:40 CEST 2918 1000 1000 3 present /usr/bin/cat
Thu 2017-10-12 08:13:07 CEST 2955 1000 1000 3 present /usr/bin/cat
The information shown for each core dump includes the date and
time of the dump, the PID, UID, and GID of the dumping process,
the signal number that caused the core dump, and the pathname of
the executable that was being run by the dumped process. Various
options to coredumpctl(1) allow a specified coredump file to be
pulled from the systemd(1) location into a specified file. For
example, to extract the core dump for PID 2955 shown above to a
file named core in the current directory, one could use:
$ coredumpctl dump 2955 -o core
For more extensive details, see the coredumpctl(1) manual page.
To (persistently) disable the systemd(1) mechanism that archives
core dumps, restoring to something more like traditional Linux
behavior, one can set an override for the systemd(1) mechanism,
using something like:
# echo "kernel.core_pattern=core.%p" > \
/etc/sysctl.d/50-coredump.conf
# /lib/systemd/systemd-sysctl
It is also possible to temporarily (i.e., until the next reboot)
change the core_pattern setting using a command such as the
following (which causes the names of core dump files to include
the executable name as well as the number of the signal which
triggered the core dump):
# sysctl -w kernel.core_pattern="%e-%s.core"
