1SYSTEMD-SYSTEM.CONF(5) systemd-system.conf SYSTEMD-SYSTEM.CONF(5)
2
6 systemd-system.conf, system.conf.d, systemd-user.conf, user.conf.d -
7 System and session service manager configuration files
8
10 /etc/systemd/system.conf, /etc/systemd/system.conf.d/*.conf,
11 /run/systemd/system.conf.d/*.conf,
12 /usr/lib/systemd/system.conf.d/*.conf
13 /etc/systemd/user.conf, /etc/systemd/user.conf.d/*.conf,
15 /run/systemd/user.conf.d/*.conf, /usr/lib/systemd/user.conf.d/*.conf
16
18 When run as a system instance, systemd interprets the configuration
19 file system.conf and the files in system.conf.d directories; when run
20 as a user instance, systemd interprets the configuration file user.conf
21 and the files in user.conf.d directories. These configuration files
22 contain a few settings controlling basic manager operations. See
23 systemd.syntax(7) for a general description of the syntax.
24
26 The default configuration is defined during compilation, so a
27 configuration file is only needed when it is necessary to deviate from
28 those defaults. By default, the configuration file in /etc/systemd/
29 contains commented out entries showing the defaults as a guide to the
30 administrator. This file can be edited to create local overrides.
31 When packages need to customize the configuration, they can install
33 configuration snippets in /usr/lib/systemd/*.conf.d/ or
34 /usr/local/lib/systemd/*.conf.d/. The main configuration file is read
35 before any of the configuration directories, and has the lowest
36 precedence; entries in a file in any configuration directory override
37 entries in the single configuration file. Files in the *.conf.d/
38 configuration subdirectories are sorted by their filename in
39 lexicographic order, regardless of in which of the subdirectories they
40 reside. When multiple files specify the same option, for options which
41 accept just a single value, the entry in the file with the
42 lexicographically latest name takes precedence. For options which
43 accept a list of values, entries are collected as they occur in files
44 sorted lexicographically.
45 Files in /etc/ are reserved for the local administrator, who may use
47 this logic to override the configuration files installed by vendor
48 packages. It is recommended to prefix all filenames in those
49 subdirectories with a two-digit number and a dash, to simplify the
50 ordering of the files.
51 To disable a configuration file supplied by the vendor, the recommended
53 way is to place a symlink to /dev/null in the configuration directory
54 in /etc/, with the same filename as the vendor configuration file.
55
57 All options are configured in the [Manager] section:
58 LogColor=, LogLevel=, LogLocation=, LogTarget=, LogTime=, DumpCore=yes,
60 CrashChangeVT=no, CrashShell=no, CrashReboot=no, ShowStatus=yes,
61 DefaultStandardOutput=journal, DefaultStandardError=inherit
62 Configures various parameters of basic manager operation. These
63 options may be overridden by the respective process and kernel
64 command line arguments. See systemd(1) for details.
65 CtrlAltDelBurstAction=
67 Defines what action will be performed if user presses
68 Ctrl-Alt-Delete more than 7 times in 2s. Can be set to
69 "reboot-force", "poweroff-force", "reboot-immediate",
70 "poweroff-immediate" or disabled with "none". Defaults to
71 "reboot-force".
72 CPUAffinity=
74 Configures the CPU affinity for the service manager as well as the
75 default CPU affinity for all forked off processes. Takes a list of
76 CPU indices or ranges separated by either whitespace or commas. CPU
77 ranges are specified by the lower and upper CPU indices separated
78 by a dash. This option may be specified more than once, in which
79 case the specified CPU affinity masks are merged. If the empty
80 string is assigned, the mask is reset, all assignments prior to
81 this will have no effect. Individual services may override the CPU
82 affinity for their processes with the CPUAffinity= setting in unit
83 files, see systemd.exec(5).
84 NUMAPolicy=
86 Configures the NUMA memory policy for the service manager and the
87 default NUMA memory policy for all forked off processes. Individual
88 services may override the default policy with the NUMAPolicy=
89 setting in unit files, see systemd.exec(5).
90 NUMAMask=
92 Configures the NUMA node mask that will be associated with the
93 selected NUMA policy. Note that default and local NUMA policies
94 don't require explicit NUMA node mask and value of the option can
95 be empty. Similarly to NUMAPolicy=, value can be overridden by
96 individual services in unit files, see systemd.exec(5).
97 RuntimeWatchdogSec=, RebootWatchdogSec=, KExecWatchdogSec=
99 Configure the hardware watchdog at runtime and at reboot. Takes a
100 timeout value in seconds (or in other time units if suffixed with
101 "ms", "min", "h", "d", "w"). If RuntimeWatchdogSec= is set to a
102 non-zero value, the watchdog hardware (/dev/watchdog or the path
103 specified with WatchdogDevice= or the kernel option
104 systemd.watchdog-device=) will be programmed to automatically
105 reboot the system if it is not contacted within the specified
106 timeout interval. The system manager will ensure to contact it at
107 least once in half the specified timeout interval. This feature
108 requires a hardware watchdog device to be present, as it is
109 commonly the case in embedded and server systems. Not all hardware
110 watchdogs allow configuration of all possible reboot timeout
111 values, in which case the closest available timeout is picked.
112 RebootWatchdogSec= may be used to configure the hardware watchdog
113 when the system is asked to reboot. It works as a safety net to
114 ensure that the reboot takes place even if a clean reboot attempt
115 times out. Note that the RebootWatchdogSec= timeout applies only to
116 the second phase of the reboot, i.e. after all regular services are
117 already terminated, and after the system and service manager
118 process (PID 1) got replaced by the systemd-shutdown binary, see
119 system bootup(7) for details. During the first phase of the
120 shutdown operation the system and service manager remains running
121 and hence RuntimeWatchdogSec= is still honoured. In order to define
122 a timeout on this first phase of system shutdown, configure
123 JobTimeoutSec= and JobTimeoutAction= in the [Unit] section of the
124 shutdown.target unit. By default RuntimeWatchdogSec= defaults to 0
125 (off), and RebootWatchdogSec= to 10min. KExecWatchdogSec= may be
126 used to additionally enable the watchdog when kexec is being
127 executed rather than when rebooting. Note that if the kernel does
128 not reset the watchdog on kexec (depending on the specific hardware
129 and/or driver), in this case the watchdog might not get disabled
130 after kexec succeeds and thus the system might get rebooted, unless
131 RuntimeWatchdogSec= is also enabled at the same time. For this
132 reason it is recommended to enable KExecWatchdogSec= only if
133 RuntimeWatchdogSec= is also enabled. These settings have no effect
134 if a hardware watchdog is not available.
135 WatchdogDevice=
137 Configure the hardware watchdog device that the runtime and
138 shutdown watchdog timers will open and use. Defaults to
139 /dev/watchdog. This setting has no effect if a hardware watchdog is
140 not available.
141 CapabilityBoundingSet=
143 Controls which capabilities to include in the capability bounding
144 set for PID 1 and its children. See capabilities(7) for details.
145 Takes a whitespace-separated list of capability names as read by
146 cap_from_name(3). Capabilities listed will be included in the
147 bounding set, all others are removed. If the list of capabilities
148 is prefixed with ~, all but the listed capabilities will be
149 included, the effect of the assignment inverted. Note that this
150 option also affects the respective capabilities in the effective,
151 permitted and inheritable capability sets. The capability bounding
152 set may also be individually configured for units using the
153 CapabilityBoundingSet= directive for units, but note that
154 capabilities dropped for PID 1 cannot be regained in individual
155 units, they are lost for good.
156 NoNewPrivileges=
158 Takes a boolean argument. If true, ensures that PID 1 and all its
159 children can never gain new privileges through execve(2) (e.g. via
160 setuid or setgid bits, or filesystem capabilities). Defaults to
161 false. General purpose distributions commonly rely on executables
162 with setuid or setgid bits and will thus not function properly with
163 this option enabled. Individual units cannot disable this option.
164 Also see No New Privileges Flag[1].
165 SystemCallArchitectures=
167 Takes a space-separated list of architecture identifiers. Selects
168 from which architectures system calls may be invoked on this
169 system. This may be used as an effective way to disable invocation
170 of non-native binaries system-wide, for example to prohibit
171 execution of 32-bit x86 binaries on 64-bit x86-64 systems. This
172 option operates system-wide, and acts similar to the
173 SystemCallArchitectures= setting of unit files, see systemd.exec(5)
174 for details. This setting defaults to the empty list, in which case
175 no filtering of system calls based on architecture is applied.
176 Known architecture identifiers are "x86", "x86-64", "x32", "arm"
177 and the special identifier "native". The latter implicitly maps to
178 the native architecture of the system (or more specifically, the
179 architecture the system manager was compiled for). Set this setting
180 to "native" to prohibit execution of any non-native binaries. When
181 a binary executes a system call of an architecture that is not
182 listed in this setting, it will be immediately terminated with the
183 SIGSYS signal.
184 TimerSlackNSec=
186 Sets the timer slack in nanoseconds for PID 1, which is inherited
187 by all executed processes, unless overridden individually, for
188 example with the TimerSlackNSec= setting in service units (for
189 details see systemd.exec(5)). The timer slack controls the accuracy
190 of wake-ups triggered by system timers. See prctl(2) for more
191 information. Note that in contrast to most other time span
192 definitions this parameter takes an integer value in nano-seconds
193 if no unit is specified. The usual time units are understood too.
194 StatusUnitFormat=
196 Takes either name or description as the value. If name, the system
197 manager will use unit names in status messages, instead of the
198 longer and more informative descriptions set with Description=, see
199 systemd.unit(5).
200 DefaultTimerAccuracySec=
202 Sets the default accuracy of timer units. This controls the global
203 default for the AccuracySec= setting of timer units, see
204 systemd.timer(5) for details. AccuracySec= set in individual units
205 override the global default for the specific unit. Defaults to
206 1min. Note that the accuracy of timer units is also affected by the
207 configured timer slack for PID 1, see TimerSlackNSec= above.
208 DefaultTimeoutStartSec=, DefaultTimeoutStopSec=,
210 DefaultTimeoutAbortSec=, DefaultRestartSec=
211 Configures the default timeouts for starting, stopping and aborting
212 of units, as well as the default time to sleep between automatic
213 restarts of units, as configured per-unit in TimeoutStartSec=,
214 TimeoutStopSec=, TimeoutAbortSec= and RestartSec= (for services,
215 see systemd.service(5) for details on the per-unit settings).
216 Disabled by default, when service with Type=oneshot is used. For
217 non-service units, DefaultTimeoutStartSec= sets the default
218 TimeoutSec= value. DefaultTimeoutStartSec= and
219 DefaultTimeoutStopSec= default to 90s. DefaultTimeoutAbortSec= is
220 not set by default so that all units fall back to TimeoutStopSec=.
221 DefaultRestartSec= defaults to 100ms.
222 DefaultStartLimitIntervalSec=, DefaultStartLimitBurst=
224 Configure the default unit start rate limiting, as configured
225 per-service by StartLimitIntervalSec= and StartLimitBurst=. See
226 systemd.service(5) for details on the per-service settings.
227 DefaultStartLimitIntervalSec= defaults to 10s.
228 DefaultStartLimitBurst= defaults to 5.
229 DefaultEnvironment=
231 Sets manager environment variables passed to all executed
232 processes. Takes a space-separated list of variable assignments.
233 See environ(7) for details about environment variables.
234 Example:
236 DefaultEnvironment="VAR1=word1 word2" VAR2=word3 "VAR3=word 5 6"
238 Sets three variables "VAR1", "VAR2", "VAR3".
240 DefaultCPUAccounting=, DefaultBlockIOAccounting=,
242 DefaultMemoryAccounting=, DefaultTasksAccounting=,
243 DefaultIOAccounting=, DefaultIPAccounting=
244 Configure the default resource accounting settings, as configured
245 per-unit by CPUAccounting=, BlockIOAccounting=, MemoryAccounting=,
246 TasksAccounting=, IOAccounting= and IPAccounting=. See
247 systemd.resource-control(5) for details on the per-unit settings.
248 DefaultTasksAccounting= defaults to yes, DefaultMemoryAccounting=
249 to yes. DefaultCPUAccounting= defaults to yes if enabling CPU
250 accounting doesn't require the CPU controller to be enabled (Linux
251 4.15+ using the unified hierarchy for resource control), otherwise
252 it defaults to no. The other three settings default to no.
253 DefaultTasksMax=
255 Configure the default value for the per-unit TasksMax= setting. See
256 systemd.resource-control(5) for details. This setting applies to
257 all unit types that support resource control settings, with the
258 exception of slice units. Defaults to 15% of the sysctl setting
259 kernel.pid_max= or root cgroup pids.max. Kernel has a default value
260 for kernel.pid_max= and an algorithm of counting in case of more
261 than 32 cores. For example with the default kernel.pid_max=,
262 DefaultTasksMax= defaults to 4915, but might be greater in other
263 systems or smaller in OS containers.
264 DefaultLimitCPU=, DefaultLimitFSIZE=, DefaultLimitDATA=,
266 DefaultLimitSTACK=, DefaultLimitCORE=, DefaultLimitRSS=,
267 DefaultLimitNOFILE=, DefaultLimitAS=, DefaultLimitNPROC=,
268 DefaultLimitMEMLOCK=, DefaultLimitLOCKS=, DefaultLimitSIGPENDING=,
269 DefaultLimitMSGQUEUE=, DefaultLimitNICE=, DefaultLimitRTPRIO=,
270 DefaultLimitRTTIME=
271 These settings control various default resource limits for
272 processes executed by units. See setrlimit(2) for details. These
273 settings may be overridden in individual units using the
274 corresponding LimitXXX= directives and they accept the same
275 parameter syntax, see systemd.exec(5) for details. Note that these
276 resource limits are only defaults for units, they are not applied
277 to the service manager process (i.e. PID 1) itself.
278 DefaultOOMPolicy=
280 Configure the default policy for reacting to processes being killed
281 by the Linux Out-Of-Memory (OOM) killer. This may be used to pick a
282 global default for the per-unit OOMPolicy= setting. See
283 systemd.service(5) for details. Note that this default is not used
284 for services that have Delegate= turned on.
285
287 systemd(1), systemd.directives(7), systemd.exec(5), systemd.service(5),
288 environ(7), capabilities(7)
289
291 1. No New Privileges Flag
292 https://www.kernel.org/doc/html/latest/userspace-api/no_new_privs.html
293systemd 246 SYSTEMD-SYSTEM.CONF(5)