1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
|
# Network UPS Tools (NUT) systemd integration
# Copyright (C) 2011-2025 by NUT contributors
# Distributed under the terms of GPLv2+
# See https://networkupstools.org/
# and https://github.com/networkupstools/nut/
[Unit]
Description=Network UPS Tools - power devices information server
After=local-fs.target network.target nut-driver.target
# We don't Require drivers to be successfully started! This would be
# a change of behavior compared to init SysV, and could prevent from
# accessing successfully started, at least to audit a system.
Wants=nut-driver.target
# The `upsd` is a networked service (even if bound to a `localhost`)
# so it requires that the OS has some notion of networking already.
# Extending the unit does not require *this* file to be edited, you
# can instead drop in an additional piece of configuration, e.g. to
# require that NUT data server only starts after external networking
# is configured (usable IP addresses appear in the system) you can
# add a `/etc/systemd/system/nut-server.service.d/network.conf` with:
# [Unit]
# Requires=network-online.target
# After=network-online.target
Requires=network.target
Before=nut-monitor.service
PartOf=nut.target
Documentation=man:upsd(@MAN_SECTION_CMD_SYS@)
Documentation=@NUT_WEBSITE_BASE@/docs/man/upsd.html
Documentation=man:ups.conf(@MAN_SECTION_CFG@)
Documentation=@NUT_WEBSITE_BASE@/docs/man/ups.conf.html
Documentation=man:upsd.conf(@MAN_SECTION_CFG@)
Documentation=@NUT_WEBSITE_BASE@/docs/man/upsd.conf.html
Documentation=man:upsd.users(@MAN_SECTION_CFG@)
Documentation=@NUT_WEBSITE_BASE@/docs/man/upsd.users.html
Documentation=man:nut.conf(@MAN_SECTION_CFG@)
Documentation=@NUT_WEBSITE_BASE@/docs/man/nut.conf.html
[Service]
# A busy server can be monitoring a lot of devices as well as replying
# to many clients. The "infinity" definition is actually capped by OS
# settings and hard-coded defaults; typically can be 65535+ nowadays.
# On 64-bit distros this can well be set into hundreds of thousands
# as well (though note each connection has a CPU and RAM overhead
# so one can strike physical limits upon deployment and/or bring the
# poorly sized system to a crawl, or worse). On a running system you
# can check /proc/$MAINPID/limits for active ulimits of the process.
# From my experiments, up to 1048576 can be set, but any larger value
# falls back to 65536. The systemd definition of "infinity" is 65536
# too (or maybe it falls back to that); though this may be OS/distro
# limitation and not systemd fault specifically.
#LimitNOFILE=infinity
#LimitNOFILE=65535
LimitNOFILE=1048576
EnvironmentFile=-@CONFPATH@/nut.conf
SyslogIdentifier=%N
# Note: foreground mode "-F" by default skips writing a PID file (and
# needs default Type=simple); we can use "-FF" here to create the file
# anyway, so that old "upsd -c reload" works rather than systemd action:
ExecStartPre=-@SYSTEMD_TMPFILES_PROGRAM@ --create @systemdtmpfilesdir@/nut-common-tmpfiles.conf
ExecStart=@SBINDIR@/upsd @SYSTEMD_DAEMON_ARGS_UPSD@
ExecReload=@SBINDIR@/upsd -c reload -P $MAINPID
ExecStartPost=-/bin/grep -E 'Units|Max open files' /proc/${MAINPID}/limits
# No tracking for PIDFile path and service attribute here (it might not
# even exist).
# If "-FF" or background-daemon mode is used, so that PID file exists
# and "upsd -c stop" in particular can be used from command-line or
# legacy scripts, it causes a clean and intentional exit of the daemon.
# Then systemd should not revive it - hence restart it only on failure:
Restart=on-failure
Type=@SYSTEMD_DAEMON_TYPE_UPSD@
@SYSTEMD_DAEMON_WATCHDOG_UPSD@
@SYSTEMD_DAEMON_NOTIFYACCESS_UPSD@
[Install]
WantedBy=nut.target
Alias=upsd.service
|
