1SMARTD(8) 2016-09-28 SMARTD(8)
2
3
4
6 smartd - SMART Disk Monitoring Daemon
7
8
10 smartd [options]
11
12
14 /usr/sbin/smartd
15
16
18 smartmontools-5.43 2016-09-28 r4347
19
20
22 [This man page is generated for the Linux version of smartmontools. It
23 does not contain info specific to other platforms.]
24
25 smartd is a daemon that monitors the Self-Monitoring, Analysis and
26 Reporting Technology (SMART) system built into many ATA-3 and later
27 ATA, IDE and SCSI-3 hard drives. The purpose of SMART is to monitor the
28 reliability of the hard drive and predict drive failures, and to carry
29 out different types of drive self-tests. This version of smartd is
30 compatible with ATA/ATAPI-7 and earlier standards (see REFERENCES
31 below).
32
33 smartd will attempt to enable SMART monitoring on ATA devices (equiva‐
34 lent to smartctl -s on) and polls these and SCSI devices every 30 min‐
35 utes (configurable), logging SMART errors and changes of SMART
36 Attributes via the SYSLOG interface. The default location for these
37 SYSLOG notifications and warnings is system-dependent (typically
38 /var/log/messages or /var/log/syslog). To change this default loca‐
39 tion, please see the ´-l´ command-line option described below.
40
41 In addition to logging to a file, smartd can also be configured to send
42 email warnings if problems are detected. Depending upon the type of
43 problem, you may want to run self-tests on the disk, back up the disk,
44 replace the disk, or use a manufacturer´s utility to force reallocation
45 of bad or unreadable disk sectors. If disk problems are detected,
46 please see the smartctl manual page and the smartmontools web page/FAQ
47 for further guidance.
48
49 If you send a USR1 signal to smartd it will immediately check the sta‐
50 tus of the disks, and then return to polling the disks every 30 min‐
51 utes. See the ´-i´ option below for additional details.
52
53 smartd can be configured at start-up using the configuration file
54 /etc/smartd.conf (Windows: EXEDIR/smartd.conf). If the configuration
55 file is subsequently modified, smartd can be told to re-read the con‐
56 figuration file by sending it a HUP signal, for example with the com‐
57 mand:
58 killall -HUP smartd.
59
60 On startup, if smartd finds a syntax error in the configuration file,
61 it will print an error message and then exit. However if smartd is
62 already running, then is told with a HUP signal to re-read the configu‐
63 ration file, and then find a syntax error in this file, it will print
64 an error message and then continue, ignoring the contents of the
65 (faulty) configuration file, as if the HUP signal had never been
66 received.
67
68 When smartd is running in debug mode, the INT signal (normally gener‐
69 ated from a shell with CONTROL-C) is treated in the same way as a HUP
70 signal: it makes smartd reload its configuration file. To exit smartd
71 use CONTROL-\
72
73 On startup, in the absence of the configuration file /etc/smartd.conf,
74 the smartd daemon first scans for all devices that support SMART. The
75 scanning is done as follows:
76
77 LINUX: Examine all entries "/dev/hd[a-t]" for IDE/ATA devices, and
78 "/dev/sd[a-z]", "/dev/sd[a-c][a-z]" for SCSI or SATA devices.
79
80 smartd then monitors for all possible SMART errors (corresponding to
81 the ´-a´ Directive in the configuration file; see CONFIGURATION FILE
82 below).
83
84
86 -A PREFIX, --attributelog=PREFIX
87 [ATA only] Writes smartd attribute information (normalized and
88 raw attribute values) to files ´PREFIX´´MODEL-SERIAL.ata.csv´.
89 At each check cycle attributes are logged as a line of semicolon
90 separated triplets of the form "attribute-ID;attribute-norm-
91 value;attribute-raw-value;". Each line is led by a date string
92 of the form "yyyy-mm-dd HH:MM:SS" (in UTC).
93
94 MODEL and SERIAL are build from drive identify information,
95 invalid characters are replaced by underline.
96
97 If the PREFIX has the form ´/path/dir/´ (e.g.
98 ´/var/lib/smartd/´), then files ´MODEL-SERIAL.ata.csv´ are cre‐
99 ated in directory ´/path/dir´. If the PREFIX has the form
100 ´/path/name´ (e.g. ´/var/lib/misc/attrlog-´), then files 'nameM‐
101 ODEL-SERIAL.ata.csv' are created in directory '/path/'. The
102 path must be absolute, except if debug mode is enabled.
103
104 -B [+]FILE, --drivedb=[+]FILE
105 [ATA only] Read the drive database from FILE. The new database
106 replaces the built in database by default. If ´+´ is specified,
107 then the new entries prepend the built in entries. Please see
108 the smartctl(8) man page for further details.
109
110 -c FILE, --configfile=FILE
111 Read smartd configuration Directives from FILE, instead of from
112 the default location /etc/smartd.conf (Windows:
113 EXEDIR/smartd.conf). If FILE does not exist, then smartd will
114 print an error message and exit with nonzero status. Thus, ´-c
115 /etc/smartd.conf´ can be used to verify the existence of the
116 default configuration file.
117
118 By using ´-´ for FILE, the configuration is read from standard
119 input. This is useful for commands like:
120 echo /dev/hdb -m user@home -M test | smartd -c - -q onecheck
121 to perform quick and simple checks without a configuration file.
122
123 -C, --capabilities
124 Use capabilities(7).
125
126 Warning: Mail notification does not work when used.
127
128 -d, --debug
129 Runs smartd in "debug" mode. In this mode, it displays status
130 information to STDOUT rather than logging it to SYSLOG and does
131 not fork(2) into the background and detach from the controlling
132 terminal. In this mode, smartd also prints more verbose infor‐
133 mation about what it is doing than when operating in "daemon"
134 mode. In this mode, the INT signal (normally generated from a
135 terminal with CONTROL-C) makes smartd reload its configuration
136 file. Please use CONTROL-\ to exit
137
138 -D, --showdirectives
139 Prints a list (to STDOUT) of all the possible Directives which
140 may appear in the configuration file /etc/smartd.conf, and then
141 exits. These Directives are also described later in this man
142 page. They may appear in the configuration file following the
143 device name.
144
145 -h, --help, --usage
146 Prints usage message to STDOUT and exits.
147
148 -i N, --interval=N
149 Sets the interval between disk checks to N seconds, where N is a
150 decimal integer. The minimum allowed value is ten and the maxi‐
151 mum is the largest positive integer that can be represented on
152 your system (often 2^31-1). The default is 1800 seconds.
153
154 Note that the superuser can make smartd check the status of the
155 disks at any time by sending it the SIGUSR1 signal, for example
156 with the command:
157 kill -SIGUSR1 <pid>
158 where <pid> is the process id number of smartd. One may also
159 use:
160 killall -USR1 smartd
161 for the same purpose.
162
163 -l FACILITY, --logfacility=FACILITY
164 Uses syslog facility FACILITY to log the messages from smartd.
165 Here FACILITY is one of local0, local1, ..., local7, or daemon
166 [default]. If this command-line option is not used, then by
167 default messages from smartd are logged to the facility daemon.
168
169 If you would like to have smartd messages logged somewhere other
170 than the default location, this can typically be accomplished
171 with (for example) the following steps:
172
173 [1] Modify the script that starts smartd to include the smartd
174 command-line argument ´-l local3´. This tells smartd to log
175 its messages to facility local3.
176
177 [2] Modify the syslogd configuration file (typically /etc/sys‐
178 log.conf) by adding a line of the form:
179 local3.* /var/log/smartd.log
180 This tells syslogd to log all the messages from facility
181 local3 to the designated file: /var/log/smartd.log.
182
183 [3] Tell syslogd to re-read its configuration file, typically by
184 sending the syslogd process a SIGHUP hang-up signal.
185
186 [4] Start (or restart) the smartd daemon.
187
188 For more detailed information, please refer to the man pages for
189 syslog.conf, syslogd, and syslog. You may also want to modify
190 the log rotation configuration files; see the man pages for
191 logrotate and examine your system´s /etc/logrotate.conf file.
192
193 -n, --no-fork
194 Do not fork into background; this is useful when executed from
195 modern init methods like initng, minit or supervise.
196
197 -p NAME, --pidfile=NAME
198 Writes pidfile NAME containing the smartd Process ID number
199 (PID). To avoid symlink attacks make sure the directory to
200 which pidfile is written is only writable for root. Without
201 this option, or if the --debug option is given, no PID file is
202 written on startup. If smartd is killed with a maskable signal
203 then the pidfile is removed.
204
205 -q WHEN, --quit=WHEN
206 Specifies when, if ever, smartd should exit. The valid argu‐
207 ments are to this option are:
208
209 nodev - Exit if there are no devices to monitor, or if any
210 errors are found at startup in the configuration file. This is
211 the default.
212
213 errors - Exit if there are no devices to monitor, or if any
214 errors are found in the configuration file /etc/smartd.conf at
215 startup or whenever it is reloaded.
216
217 nodevstartup - Exit if there are no devices to monitor at
218 startup. But continue to run if no devices are found whenever
219 the configuration file is reloaded.
220
221 never - Only exit if a fatal error occurs (no remaining system
222 memory, invalid command line arguments). In this mode, even if
223 there are no devices to monitor, or if the configuration file
224 /etc/smartd.conf has errors, smartd will continue to run, wait‐
225 ing to load a configuration file listing valid devices.
226
227 onecheck - Start smartd in debug mode, then register devices,
228 then check device´s SMART status once, and then exit with zero
229 exit status if all of these steps worked correctly.
230
231 This last option is intended for ´distribution-writers´ who want
232 to create automated scripts to determine whether or not to auto‐
233 matically start up smartd after installing smartmontools. After
234 starting smartd with this command-line option, the distribu‐
235 tion´s install scripts should wait a reasonable length of time
236 (say ten seconds). If smartd has not exited with zero status by
237 that time, the script should send smartd a SIGTERM or SIGKILL
238 and assume that smartd will not operate correctly on the host.
239 Conversely, if smartd exits with zero status, then it is safe to
240 run smartd in normal daemon mode. If smartd is unable to monitor
241 any devices or encounters other problems then it will return
242 with non-zero exit status.
243
244 showtests - Start smartd in debug mode, then register devices,
245 then write a list of future scheduled self tests to stdout, and
246 then exit with zero exit status if all of these steps worked
247 correctly. Device's SMART status is not checked.
248
249 This option is intended to test whether the '-s REGEX' direc‐
250 tives in smartd.conf will have the desired effect. The output
251 lists the next test schedules, limited to 5 tests per type and
252 device. This is followed by a summary of all tests of each
253 device within the next 90 days.
254
255 -r TYPE, --report=TYPE
256 Intended primarily to help smartmontools developers understand
257 the behavior of smartmontools on non-conforming or poorly-con‐
258 forming hardware. This option reports details of smartd trans‐
259 actions with the device. The option can be used multiple times.
260 When used just once, it shows a record of the ioctl() transac‐
261 tions with the device. When used more than once, the detail of
262 these ioctl() transactions are reported in greater detail. The
263 valid arguments to this option are:
264
265 ioctl - report all ioctl() transactions.
266
267 ataioctl - report only ioctl() transactions with ATA devices.
268
269 scsiioctl - report only ioctl() transactions with SCSI devices.
270
271 Any argument may include a positive integer to specify the level
272 of detail that should be reported. The argument should be fol‐
273 lowed by a comma then the integer with no spaces. For example,
274 ataioctl,2 The default level is 1, so ´-r ataioctl,1´ and ´-r
275 ataioctl´ are equivalent.
276
277 -s PREFIX, --savestates=PREFIX
278 [ATA only] Reads/writes smartd state information from/to files
279 ´PREFIX´´MODEL-SERIAL.ata.state´. This preserves SMART
280 attributes, drive min and max temperatures (-W directive), info
281 about last sent warning email (-m directive), and the time of
282 next check of the self-test REGEXP (-s directive) across boot
283 cycles.
284
285 MODEL and SERIAL are build from drive identify information,
286 invalid characters are replaced by underline.
287
288 If the PREFIX has the form ´/path/dir/´ (e.g.
289 ´/var/lib/smartd/´), then files ´MODEL-SERIAL.ata.state´ are
290 created in directory ´/path/dir´. If the PREFIX has the form
291 ´/path/name´ (e.g. ´/var/lib/misc/smartd-´), then files 'nameMO‐
292 DEL-SERIAL.ata.state' are created in directory '/path/'. The
293 path must be absolute, except if debug mode is enabled.
294
295 The state information files are read on smartd startup. The
296 files are always (re)written after reading the configuration
297 file, before rereading the configuration file (SIGHUP), before
298 smartd shutdown, and after a check forced by SIGUSR1. After a
299 normal check cycle, a file is only rewritten if an important
300 change (which usually results in a SYSLOG output) occurred.
301
302 -V, --version, --license, --copyright
303 Prints version, copyright, license, home page and SVN revision
304 information for your copy of smartd to STDOUT and then exits.
305 Please include this information if you are reporting bugs or
306 problems.
307
308
310 smartd
311 Runs the daemon in forked mode. This is the normal way to run smartd.
312 Entries are logged to SYSLOG.
313
314 smartd -d -i 30
315 Run in foreground (debug) mode, checking the disk status every 30 sec‐
316 onds.
317
318 smartd -q onecheck
319 Registers devices, and checks the status of the devices exactly once.
320 The exit status (the bash $? variable) will be zero if all went well,
321 and nonzero if no devices were detected or some other problem was
322 encountered.
323
324 Note that smartmontools provides a start-up script in
325 /etc/rc.d/init.d/smartd which is responsible for starting and stopping
326 the daemon via the normal init interface. Using this script, you can
327 start smartd by giving the command:
328 /etc/rc.d/init.d/smartd start
329 and stop it by using the command:
330 /etc/rc.d/init.d/smartd stop
331
333 The syntax of the smartd.conf(5) file is discussed separately.
334
336 smartd will make log entries at loglevel LOG_INFO if the Normalized
337 SMART Attribute values have changed, as reported using the ´-t´, ´-p´,
338 or ´-u´ Directives. For example:
339 ´Device: /dev/hda, SMART Attribute: 194 Temperature_Celsius changed from 94 to 93´
340 Note that in this message, the value given is the ´Normalized´ not the
341 ´Raw´ Attribute value (the disk temperature in this case is about 22
342 Celsius). The ´-R´ and ´-r´ Directives modify this behavior, so that
343 the information is printed with the Raw values as well, for example:
344 ´Device: /dev/hda, SMART Attribute: 194 Temperature_Celsius changed from 94 [Raw 22] to 93 [Raw 23]´
345 Here the Raw values are the actual disk temperatures in Celsius. The
346 way in which the Raw values are printed, and the names under which the
347 Attributes are reported, is governed by the various ´-v Num,Descrip‐
348 tion´ Directives described previously.
349
350 Please see the smartctl manual page for further explanation of the dif‐
351 ferences between Normalized and Raw Attribute values.
352
353 smartd will make log entries at loglevel LOG_CRIT if a SMART Attribute
354 has failed, for example:
355 ´Device: /dev/hdc, Failed SMART Attribute: 5 Reallocated_Sector_Ct´
356 This loglevel is used for reporting enabled by the ´-H´, -f´,
357 ´-l selftest´, and ´-l error´ Directives. Entries reporting failure of
358 SMART Prefailure Attributes should not be ignored: they mean that the
359 disk is failing. Use the smartctl utility to investigate.
360
361
363 When smartd makes log entries, these are time-stamped. The time stamps
364 are in the computer's local time zone, which is generally set using
365 either the environment variable ´TZ´ or using a time-zone file such as
366 /etc/localtime. You may wish to change the timezone while smartd is
367 running (for example, if you carry a laptop to a new time-zone and
368 don't reboot it). Due to a bug in the tzset(3) function of many unix
369 standard C libraries, the time-zone stamps of smartd might not change.
370 For some systems, smartd will work around this problem if the time-zone
371 is set using /etc/localtime. The work-around fails if the time-zone is
372 set using the ´TZ´ variable (or a file that it points to).
373
374
375
377 The return value (exit status) of smartd can have the following values:
378
379 0: Daemon startup successful, or smartd was killed by a SIGTERM (or
380 in debug mode, a SIGQUIT).
381
382 1: Commandline did not parse.
383
384 2: There was a syntax error in the config file.
385
386 3: Forking the daemon failed.
387
388 4: Couldn´t create PID file.
389
390 5: Config file does not exist (only returned in conjunction with
391 the ´-c´ option).
392
393 6: Config file exists, but cannot be read.
394
395 8: smartd ran out of memory during startup.
396
397 9: A compile time constant of smartd was too small. This can be
398 caused by an excessive number of disks, or by lines in
399 /etc/smartd.conf that are too long. Please report this problem
400 to smartmontools-support@lists.sourceforge.net.
401
402 10: An inconsistency was found in smartd´s internal data structures.
403 This should never happen. It must be due to either a coding or
404 compiler bug. Please report such failures to smartmontools-sup‐
405 port@lists.sourceforge.net.
406
407 16: A device explicitly listed in /etc/smartd.conf can´t be moni‐
408 tored.
409
410 17: smartd didn´t find any devices to monitor.
411
412 254: When in daemon mode, smartd received a SIGINT or SIGQUIT. (Note
413 that in debug mode, SIGINT has the same effect as SIGHUP, and
414 makes smartd reload its configuration file. SIGQUIT has the same
415 effect as SIGTERM and causes smartd to exit with zero exit sta‐
416 tus.
417
418 132 and above
419 smartd was killed by a signal that is not explicitly listed
420 above. The exit status is then 128 plus the signal number. For
421 example if smartd is killed by SIGKILL (signal 9) then the exit
422 status is 137.
423
424
426 Bruce Allen smartmontools-support@lists.sourceforge.net
427 University of Wisconsin - Milwaukee Physics Department
428
429
431 The following have made large contributions to smartmontools:
432 Casper Dik (Solaris SCSI interface)
433 Christian Franke (Windows interface, C++ redesign, USB support, ...)
434 Douglas Gilbert (SCSI subsystem)
435 Guido Guenther (Autoconf/Automake packaging)
436 Geoffrey Keating (Darwin ATA interface)
437 Eduard Martinescu (FreeBSD interface)
438 Frédéric L. W. Meunier (Web site and Mailing list)
439 Gabriele Pohl (Web site and Wiki, conversion from CVS to SVN)
440 Keiji Sawada (Solaris ATA interface)
441 Manfred Schwarb (Drive database)
442 Sergey Svishchev (NetBSD interface)
443 David Snyder and Sergey Svishchev (OpenBSD interface)
444 Phil Williams (User interface and drive database)
445 Shengfeng Zhou (Linux/FreeBSD HighPoint RocketRAID interface)
446 Many other individuals have made smaller contributions and corrections.
447
448
450 This code was derived from the smartsuite package, written by Michael
451 Cornwell, and from the previous UCSC smartsuite package. It extends
452 these to cover ATA-5 disks. This code was originally developed as a
453 Senior Thesis by Michael Cornwell at the Concurrent Systems Laboratory
454 (now part of the Storage Systems Research Center), Jack Baskin School
455 of Engineering, University of California, Santa Cruz.
456 http://ssrc.soe.ucsc.edu/ .
457
459 Please see the following web site for updates, further documentation,
460 bug reports and patches: http://smartmontools.sourceforge.net/
461
462
464 smartd.conf(5), smartctl(8), syslogd(8), syslog.conf(5), badblocks(8),
465 ide-smart(8), regex(7).
466
467
469 An introductory article about smartmontools is Monitoring Hard Disks
470 with SMART, by Bruce Allen, Linux Journal, January 2004, pages 74-77.
471 This is http://www.linuxjournal.com/article/6983 online.
472
473 If you would like to understand better how SMART works, and what it
474 does, a good place to start is with Sections 4.8 and 6.54 of the first
475 volume of the ´AT Attachment with Packet Interface-7´ (ATA/ATAPI-7)
476 specification Revision 4b. This documents the SMART functionality
477 which the smartmontools utilities provide access to.
478
479 The functioning of SMART was originally defined by the SFF-8035i revi‐
480 sion 2 and the SFF-8055i revision 1.4 specifications. These are publi‐
481 cations of the Small Form Factors (SFF) Committee.
482
483 Links to these and other documents may be found on the Links page of
484 the smartmontools Wiki at http://sourceforge.net/apps/trac/smartmon‐
485 tools/wiki/Links .
486
487
489 $Id: smartd.8.in 3561 2012-06-05 19:49:31Z chrfranke $
490
491
492
493smartmontools-5.43 2016-09-28 SMARTD(8)