Skip to content

Commit

Permalink
[coredump] Refactor info and dump collection
Browse files Browse the repository at this point in the history
Refactor the plugin to adjust how `coredumpctl info` and coredump file
collection is handled.

Plugin will now collect the compressed coredump file for coredump
entries for which coredumpctl reports that the file is present, for the
first X number of coredump files where X is the new `dumps` plugin
option.

A second new plugin option, `executable`, can be used to specific a
regex string to match coredump entries against to determine if they
should be collected or not. The default behavior is to collect for all
entries.

A symlink for coredump file collections will be dropped in the plugin
directory to aide in associating dump files with coredumpctl entries, as
it may not always be obvious for end users based on filenames alone.

Signed-off-by: Jake Hunsaker <[email protected]>
  • Loading branch information
TurboTurtle committed Nov 4, 2024
1 parent 762e2da commit c8e9655
Showing 1 changed file with 85 additions and 10 deletions.
95 changes: 85 additions & 10 deletions sos/report/plugins/coredump.py
Original file line number Diff line number Diff line change
Expand Up @@ -8,36 +8,111 @@
#
# See the LICENSE file in the source distribution for further information.

import os
import re
from sos.report.plugins import Plugin, IndependentPlugin, PluginOpt


class Coredump(Plugin, IndependentPlugin):
"""
This plugin collects from the systemd-coredump service, which is leveraged
via the coredumpctl command.
short_desc = 'Retrieve coredump information'
Beyond basic configuration of the service, the plugin will also attempt to
collect information on available coredumps. For any coredumps reported by
coredumpctl for which the core is present, this plugin will collect that
coredump file. The number of dump files collects is controlled by the
`dumps` option, defaulting to 3. Note that core collection is performed
based on the most recent cores generated. The summary information provided
by `coredumpctl info` is also collected.
The coredump files will be at their canonical location in the sos archive,
usually under var/lib/systemd/coredump. Symlinks are dropped in the plugin
directory to assist in identifying which coredump aligns to which entry in
the provided `coredumpctl list` output collection.
Users may leverage the `executable` option to control which coredumps the
plugin will collect. This option takes a case-insensitive python regex
string. If provided, only coredump entries for which the EXE field in
`coredumpctl list` output matches will be collected.
The dump files collected are compressed, and users should be aware that
when inflated these files can be orders of magnitude larger than their
collected sizes.
"""

short_desc = 'systemd-coredump related information and dump files'

plugin_name = "coredump"
profiles = ('system', 'debug')
packages = ('systemd-udev', 'systemd-coredump')

option_list = [
PluginOpt("detailed", default=False,
desc="collect detailed information for every report")
PluginOpt("dumps", default=3, desc="number of dump files to collect"),
PluginOpt("executable", default='',
desc=("only collect info and dump output for executables "
"matching this regex"))
]

def setup(self):
self.add_copy_spec([
"/etc/systemd/coredump.conf",
"/etc/systemd/coredump.conf.d/",
"/run/systemd/coredump.conf.d/",
"/usr/lib/systemd/coredump.conf.d/"
"/usr/lib/systemd/coredump.conf.d/",
"/usr/lib/systemd/systemd-coredump"
])

self.add_cmd_output("coredumpctl dump")
coredump_list = self.collect_cmd_output("coredumpctl list --reverse")
if coredump_list['status'] != 0:
return

coredump_list = self.collect_cmd_output("coredumpctl list")
if self.get_option("detailed") and coredump_list['status'] == 0:
for line in coredump_list["output"].splitlines()[1:]:
self.add_cmd_output("coredumpctl info "
f"{line.split()[4]}")
cores_collected = 0
for line in coredump_list["output"].splitlines()[1:]:
cdump = line.split()
pid = cdump[4]
exe = cdump[-2]
if regex := self.get_option("executable"):
if not re.search(regex, exe, re.I):
continue
cinfo = self.collect_cmd_output(f"coredumpctl info {pid}")
if cinfo['status'] != 0:
continue
res = cinfo['output']
if cores_collected < self.get_option("dumps"):
core = re.search(r"(^\s*Storage:(.*)(\(present\)))", res, re.M)
try:
core_path = core.groups()[1].strip()
# a_c_s does not return any information for a skipped file,
# so stat the size here and if the core is larger than our
# limit, move on to the next
# TODO: do not hardcode this. Extend log-size to per-plugin
# TODO: option and link this to that value
if os.stat(core_path).st_size > 209715200:
self._log_info(
f"Skipping core dump file {core_path} due to size"
)
continue
self.add_copy_spec(core_path, tailit=False, sizelimit=200)
plugpath = self.path_join(
self.commons['cmddir'],
self.name(),
f"coredump_{pid}_{self._mangle_command(exe)}"
)
# use os.path instead of the plugin wrapper here so that
# we always get the real path of the archive on disk
arcpath = os.path.join(self.archive.get_archive_path(),
core_path)
linkpath = os.path.relpath(arcpath, plugpath)
linkpath = linkpath.replace('../', '', 1)
self.archive.add_link(linkpath, plugpath)
cores_collected += 1
except AttributeError:
# no match on the re.search()
pass
except Exception as err:
self._log_info(
f"Could not collect coredump for {pid} : {err}"
)

# vim: set et ts=4 sw=4 :

0 comments on commit c8e9655

Please sign in to comment.