|
| 1 | +# /etc/watchdogd.conf sample |
| 2 | +# Commented out values are program defaults. |
| 3 | +# |
| 4 | +# The checker/monitor `warning` and `critical` levels are 0.00-1.00, |
| 5 | +# i.e. 0-100%, except for load average which can vary a lot between |
| 6 | +# systems and use-cases, not just because of the number of CPU cores. |
| 7 | +# Use the `script = ...` setting to call script when `warning` and |
| 8 | +# `critical` are reached for a monitor. In `critical` the monitor |
| 9 | +# otherwise triggers an unconditional reboot. |
| 10 | +# |
| 11 | +# NOTE: `critical` is optional, omitting it disables the reboot action. |
| 12 | +# |
| 13 | + |
| 14 | +### Watchdogs ########################################################## |
| 15 | +# Global settings that can be overridden per watchdog |
| 16 | + |
| 17 | +# Do not set WDT timeout and kick interval too low, the daemon runs at |
| 18 | +# SCHED_OTHER level with all other tasks, unless the process supervisor |
| 19 | +# is enabled. The monitor plugins (below) need CPU time as well. |
| 20 | +#timeout = 20 |
| 21 | +#interval = 10 |
| 22 | + |
| 23 | +# With safe-exit enabled (true) the daemon will ask the driver disable |
| 24 | +# the WDT before exiting (SIGINT). However, some WDT drivers (or HW) |
| 25 | +# may not support this. |
| 26 | +#safe-exit = true |
| 27 | + |
| 28 | +# Multiple watchdogs can be kicked, the default, even if no .conf file |
| 29 | +# is found or device node given on the command line, is /dev/watchdog |
| 30 | +device /dev/watchdog { |
| 31 | + timeout = 60 |
| 32 | + interval = 5 |
| 33 | + safe-exit = true |
| 34 | +} |
| 35 | + |
| 36 | +#device /dev/watchdog2 { |
| 37 | +# timeout = 20 |
| 38 | +# interval = 10 |
| 39 | +# safe-exit = true |
| 40 | +#} |
| 41 | + |
| 42 | +### Supervisor ######################################################### |
| 43 | +# Instrumented processes can have their main loop supervised. Processes |
| 44 | +# subscribe to this service using the libwdog API, see the docs for more |
| 45 | +# on this. When the supervisor is enabled and the priority is set to a |
| 46 | +# value > 0, watchdogd runs as a SCHED_RR process with elevated realtime |
| 47 | +# priority. When disabled, or the priority is set to zero (0), it runs |
| 48 | +# as a regular SCHED_OTHER process, this is the default. |
| 49 | +# |
| 50 | +# When a supervised process fails to meet its deadline, the daemon will |
| 51 | +# perform an unconditional reset having saved the reset reason. If a |
| 52 | +# script is provided in this section it will be called instead. The |
| 53 | +# script is called as: |
| 54 | +# |
| 55 | +# script.sh supervisor CODE PID LABEL |
| 56 | +# |
| 57 | +# Availabel CODEs for the reset reason are avilable in wdog.h |
| 58 | +# |
| 59 | +#supervisor { |
| 60 | +# !!!REMEMBER TO ENABLE reset-reason (below) AS WELL!!! |
| 61 | +# enabled = true |
| 62 | +# priority = 98 |
| 63 | +# script = "/path/to/supervisor-script.sh" |
| 64 | +#} |
| 65 | + |
| 66 | +### Reset reason ####################################################### |
| 67 | +# The following section controls if/how the reset reason & reset counter |
| 68 | +# is tracked. By default this is disabled, since not all systems allow |
| 69 | +# writing to disk, e.g. embedded systems using MTD devices with limited |
| 70 | +# number of write cycles. |
| 71 | +# |
| 72 | +# The default file setting is a non-volatile path, according to the FHS. |
| 73 | +# It can be changed to another location, but make sure that location is |
| 74 | +# writable first. |
| 75 | +reset-reason { |
| 76 | + enabled = true |
| 77 | + file = "/var/lib/misc/watchdogd.state" |
| 78 | +} |
| 79 | + |
| 80 | +### Checkers/Monitors ################################################## |
| 81 | +# |
| 82 | +# Script or command to run instead of reboot when a monitor plugin |
| 83 | +# reaches any of its critical or warning level. Setting this will |
| 84 | +# disable the built-in reboot on critical, it is therefore up to the |
| 85 | +# script to perform reboot, if needed. The script is called as: |
| 86 | +# |
| 87 | +# script.sh {filenr, fsmon, loadavg, meminfo} {crit, warn} VALUE |
| 88 | +# |
| 89 | +#script = "/path/to/reboot-action.sh" |
| 90 | + |
| 91 | +# Monitors file descriptor leaks based on /proc/sys/fs/file-nr |
| 92 | +filenr { |
| 93 | +# enabled = true |
| 94 | + interval = 300 |
| 95 | + logmark = false |
| 96 | + warning = 0.9 |
| 97 | + critical = 1.0 |
| 98 | +# script = "/path/to/alt-reboot-action.sh" |
| 99 | +} |
| 100 | + |
| 101 | +# Monitors a file system, blocks and inode usage against watermarks |
| 102 | +# The script is called with fsmon as the first argument and there |
| 103 | +# are two environment variables FSMON_NAME, for the monitored path, |
| 104 | +# and FSMON_TYPE indicating either 'blocks' or 'inodes'. |
| 105 | +#fsmon /var { |
| 106 | +# enabled = true |
| 107 | +# interval = 300 |
| 108 | +# logmark = false |
| 109 | +# warning = 0.95 |
| 110 | +# critical = 1.0 |
| 111 | +# script = "/path/to/alt-reboot-action.sh" |
| 112 | +#} |
| 113 | + |
| 114 | +# Monitors load average based on sysinfo() from /proc/loadavg |
| 115 | +# The level is composed from the average of the 1 and 5 min marks. |
| 116 | +loadavg { |
| 117 | +# enabled = true |
| 118 | + interval = 300 |
| 119 | + logmark = false |
| 120 | + warning = 1.0 |
| 121 | + critical = 2.0 |
| 122 | +# script = "/path/to/alt-reboot-action.sh" |
| 123 | +} |
| 124 | + |
| 125 | +# Monitors free RAM based on data from /proc/meminfo |
| 126 | +meminfo { |
| 127 | +# enabled = true |
| 128 | + interval = 300 |
| 129 | + logmark = false |
| 130 | + warning = 0.9 |
| 131 | + critical = 0.95 |
| 132 | +# script = "/path/to/alt-reboot-action.sh" |
| 133 | +} |
| 134 | + |
| 135 | +# Monitor temperature. The critical value is unset by default, so no |
| 136 | +# action is taken at that watermark (by default). Both the critical and |
| 137 | +# warning watermarks are relative to the trip/critical/max value from |
| 138 | +# sysfs. The warning is default 0.9, i.e., 90% of critical. Use script |
| 139 | +# to to reset the fan controller or poweroff(8) the system. |
| 140 | +# |
| 141 | +# Each temp monitor caches the last 10 values, calculates the mean, and |
| 142 | +# compares that to the warning and critical levels. Logging is only |
| 143 | +# done every 10 x interval (if enabled). |
| 144 | +#tempmon /sys/class/hwmon/hwmon0/temp1_input { |
| 145 | +# enabled = true |
| 146 | +# interval = 30 |
| 147 | +# warning = 0.9 |
| 148 | +# critical = 0.95 |
| 149 | +# logmark = true |
| 150 | +# script = "/script/to/log/and/poweroff.sh" |
| 151 | +#} |
| 152 | + |
| 153 | +# Monitor a generic script, executes 'monitor-script' every 'interval' |
| 154 | +# seconds, with a max runtime of 'timeout' seconds. When the exit code |
| 155 | +# of the monitor script is above the critical level watchdogd either |
| 156 | +# starts the reboot, or calls the alternate 'script' to determin the |
| 157 | +# next cause of action. |
| 158 | +#generic /path/to/monitor-script.sh { |
| 159 | +# enabled = true |
| 160 | +# interval = 300 |
| 161 | +# timeout = 60 |
| 162 | +# warning = 1 |
| 163 | +# critical = 10 |
| 164 | +# script = "/path/to/alt-reboot-action.sh" |
| 165 | +#} |
0 commit comments