]> git.ipfire.org Git - thirdparty/mdadm.git/blob - mdadm.conf.5
Allow homehost to be largely ignored when assembling arrays.
[thirdparty/mdadm.git] / mdadm.conf.5
1 .\" Copyright Neil Brown and others.
2 .\" This program is free software; you can redistribute it and/or modify
3 .\" it under the terms of the GNU General Public License as published by
4 .\" the Free Software Foundation; either version 2 of the License, or
5 .\" (at your option) any later version.
6 .\" See file COPYING in distribution for details.
7 .TH MDADM.CONF 5
8 .SH NAME
9 mdadm.conf \- configuration for management of Software RAID with mdadm
10 .SH SYNOPSIS
11 /etc/mdadm.conf
12 .SH DESCRIPTION
13 .PP
14 .I mdadm
15 is a tool for creating, managing, and monitoring RAID devices using the
16 .B md
17 driver in Linux.
18 .PP
19 Some common tasks, such as assembling all arrays, can be simplified
20 by describing the devices and arrays in this configuration file.
21
22 .SS SYNTAX
23 The file should be seen as a collection of words separated by white
24 space (space, tab, or newline).
25 Any word that beings with a hash sign (#) starts a comment and that
26 word together with the remainder of the line is ignored.
27
28 Any line that starts with white space (space or tab) is treated as
29 though it were a continuation of the previous line.
30
31 Empty lines are ignored, but otherwise each (non continuation) line
32 must start with a keyword as listed below. The keywords are case
33 insensitive and can be abbreviated to 3 characters.
34
35 The keywords are:
36 .TP
37 .B DEVICE
38 A
39 .B device
40 line lists the devices (whole devices or partitions) that might contain
41 a component of an MD array. When looking for the components of an
42 array,
43 .I mdadm
44 will scan these devices (or any devices listed on the command line).
45
46 The
47 .B device
48 line may contain a number of different devices (separated by spaces)
49 and each device name can contain wild cards as defined by
50 .BR glob (7).
51
52 Also, there may be several device lines present in the file.
53
54 Alternatively, a
55 .B device
56 line can contain either of both of the words
57 .B containers
58 and
59 .BR partitions .
60 The word
61 .B containers
62 will cause
63 .I mdadm
64 to look for assembled CONTAINER arrays and included them as a source
65 for assembling further arrays.
66 .PP
67 The word
68 .I partitions
69 will cause
70 .I mdadm
71 to read
72 .I /proc/partitions
73 and include all devices and partitions found therein.
74 .I mdadm
75 does not use the names from
76 .I /proc/partitions
77 but only the major and minor device numbers. It scans
78 .I /dev
79 to find the name that matches the numbers.
80
81 If no DEVICE line is present, then "DEVICE partitions containers" is assumed.
82
83 For example:
84 .IP
85 DEVICE /dev/hda* /dev/hdc*
86 .br
87 DEV /dev/sd*
88 .br
89 DEVICE /dev/discs/disc*/disc
90 .br
91 DEVICE partitions
92
93 .TP
94 .B ARRAY
95 The ARRAY lines identify actual arrays. The second word on the line
96 may be the name of the device where the array is normally
97 assembled, such as
98 .B /dev/md1
99 or
100 .BR /dev/md/backup .
101 If the name does not start with a slash
102 .RB (' / '),
103 it is treated as being in
104 .BR /dev/md/ .
105 Alternately the word
106 .B <ignore>
107 (complete with angle brackets) can be given in which case any array
108 which matches the rest of the line will never be automatically assembled.
109 If no device name is given,
110 .I mdadm
111 will use various heuristics to determine an appropriate name.
112 .PP
113 Subsequent words identify the array, or identify the array as a member
114 of a group. If multiple identities are given,
115 then a component device must match ALL identities to be considered a
116 match. Each identity word has a tag, and equals sign, and some value.
117 The tags are:
118
119 .RS 4
120 .TP
121 .B uuid=
122 The value should be a 128 bit uuid in hexadecimal, with punctuation
123 interspersed if desired. This must match the uuid stored in the
124 superblock.
125 .TP
126 .B name=
127 The value should be a simple textual name as was given to
128 .I mdadm
129 when the array was created. This must match the name stored in the
130 superblock on a device for that device to be included in the array.
131 Not all superblock formats support names.
132 .TP
133 .B super\-minor=
134 The value is an integer which indicates the minor number that was
135 stored in the superblock when the array was created. When an array is
136 created as /dev/mdX, then the minor number X is stored.
137 .TP
138 .B devices=
139 The value is a comma separated list of device names or device name
140 patterns.
141 Only devices with names which match one entry in the list will be used
142 to assemble the array. Note that the devices
143 listed there must also be listed on a DEVICE line.
144 .TP
145 .B level=
146 The value is a raid level. This is not normally used to
147 identify an array, but is supported so that the output of
148
149 .B "mdadm \-\-examine \-\-scan"
150
151 can be use directly in the configuration file.
152 .TP
153 .B num\-devices=
154 The value is the number of devices in a complete active array. As with
155 .B level=
156 this is mainly for compatibility with the output of
157
158 .BR "mdadm \-\-examine \-\-scan" .
159
160 .TP
161 .B spares=
162 The value is a number of spare devices to expect the array to have.
163 .B mdadm \-\-monitor
164 will report an array if it is found to have fewer than this number of
165 spares when
166 .B \-\-monitor
167 starts or when
168 .B \-\-oneshot
169 is used.
170
171 .TP
172 .B spare\-group=
173 The value is a textual name for a group of arrays. All arrays with
174 the same
175 .B spare\-group
176 name are considered to be part of the same group. The significance of
177 a group of arrays is that
178 .I mdadm
179 will, when monitoring the arrays, move a spare drive from one array in
180 a group to another array in that group if the first array had a failed
181 or missing drive but no spare.
182
183 .TP
184 .B auto=
185 This option declares to
186 .I mdadm
187 that it should try to create the device file of the array if it
188 doesn't already exist, or exists but with the wrong device number.
189
190 The value of this option can be "yes" or "md" to indicate that a
191 traditional, non-partitionable md array should be created, or "mdp",
192 "part" or "partition" to indicate that a partitionable md array (only
193 available in linux 2.6 and later) should be used. This later set can
194 also have a number appended to indicate how many partitions to create
195 device files for, e.g.
196 .BR auto=mdp5 .
197 The default is 4.
198
199 .TP
200 .B bitmap=
201 The option specifies a file in which a write-intent bitmap should be
202 found. When assembling the array,
203 .I mdadm
204 will provide this file to the
205 .B md
206 driver as the bitmap file. This has the same function as the
207 .B \-\-bitmap\-file
208 option to
209 .BR \-\-assemble .
210
211 .TP
212 .B metadata=
213 Specify the metadata format that the array has. This is mainly
214 recognised for comparability with the output of
215 .BR "mdadm \-Es" .
216
217 .TP
218 .B container=
219 Specify that this array is a member array of some container. The
220 value given can be either a path name in /dev, or a UUID of the
221 container array.
222
223 .IP
224 .B member=
225 Specify that this array is a member array of some container. Each
226 type of container has some way to enumerate member arrays, often a
227 simple sequence number. The value identifies which member of a
228 container the array is. It will usually accompany a 'container=' word.
229 .RE
230
231 .TP
232 .B MAILADDR
233 The
234 .B mailaddr
235 line gives an E-mail address that alerts should be
236 sent to when
237 .I mdadm
238 is running in
239 .B \-\-monitor
240 mode (and was given the
241 .B \-\-scan
242 option). There should only be one
243 .B MAILADDR
244 line and it should have only one address.
245
246
247 .TP
248 .B MAILFROM
249 The
250 .B mailfrom
251 line (which can only be abbreviated to at least 5 characters) gives an
252 address to appear in the "From" address for alert mails. This can be
253 useful if you want to explicitly set a domain, as the default from
254 address is "root" with no domain. All words on this line are
255 catenated with spaces to form the address.
256
257 Note that this value cannot be set via the
258 .I mdadm
259 commandline. It is only settable via the config file.
260
261 .TP
262 .B PROGRAM
263 The
264 .B program
265 line gives the name of a program to be run when
266 .B "mdadm \-\-monitor"
267 detects potentially interesting events on any of the arrays that it
268 is monitoring. This program gets run with two or three arguments, they
269 being the Event, the md device, and possibly the related component
270 device.
271
272 There should only be one
273 .B program
274 line and it should be give only one program.
275
276
277 .TP
278 .B CREATE
279 The
280 .B create
281 line gives default values to be used when creating arrays and device entries for
282 arrays.
283 These include:
284
285 .RS 4
286 .TP
287 .B owner=
288 .TP
289 .B group=
290 These can give user/group ids or names to use instead of system
291 defaults (root/wheel or root/disk).
292 .TP
293 .B mode=
294 An octal file mode such as 0660 can be given to override the default
295 of 0600.
296 .TP
297 .B auto=
298 This corresponds to the
299 .B \-\-auto
300 flag to mdadm. Give
301 .BR yes ,
302 .BR md ,
303 .BR mdp ,
304 .B part
305 \(em possibly followed by a number of partitions \(em to indicate how
306 missing device entries should be created.
307
308 .TP
309 .B metadata=
310 The name of the metadata format to use if none is explicitly given.
311 This can be useful to impose a system-wide default of version-1 superblocks.
312
313 .TP
314 .B symlinks=no
315 Normally when creating devices in
316 .B /dev/md/
317 .I mdadm
318 will create a matching symlink from
319 .B /dev/
320 with a name starting
321 .B md
322 or
323 .BR md_ .
324 Give
325 .B symlinks=no
326 to suppress this symlink creation.
327 .RE
328
329 .TP
330 .B HOMEHOST
331 The
332 .B homehost
333 line gives a default value for the
334 .B --homehost=
335 option to mdadm. There should be exactly one other word on the line.
336 It should either be a host name, or one of the special words
337 .B <system>
338 and
339 .BR <ignore> .
340 If
341 .B <system>
342 is given, then the
343 .BR gethostname ( 2 )
344 systemcall is used to get the host name.
345
346 If
347 .B <ignore>
348 is given, then a flag is set so that when arrays are being
349 auto-assemble the checking of the recorded
350 .I homehost
351 is disabled.
352
353 When arrays are created, this host name will be stored in the
354 metadata. When arrays are assembled using auto-assembly, arrays which
355 do not record the correct homehost name in their metadata will be
356 assembled using a 'foreign' name. A 'foreign' name alway ends with a
357 digit string (possibly preceded by an underscore) to differentiate it
358 from any possible local name. e.g.
359 .B /dev/md/1_1
360 or
361 .BR /dev/md/home0 .
362 .TP
363 .B AUTO
364 A list of names of metadata format can be given, each preceded by a
365 plus or minus sign. Also the word
366 .I all
367 preceded by plus or minus is allowed and is usually last.
368
369 When
370 .I mdadm
371 is auto-assembling an array, with via
372 .I --assemble
373 or
374 .I --incremental
375 and it finds metadata of a given type, it checks that metadata type
376 against those listed in this line. The first match wins, where
377 .I all
378 matches anything.
379 If a match is found that was preceded by a plus sign, the auto
380 assembly is allowed. If the match was preceded by a minus sign, the
381 auto assembly is disallowed. If no match is found, the auto assembly
382 is allowed.
383
384 This can be used to disable all auto-assembly (so that only arrays
385 explicitly listed in mdadm.conf or on the command line are assembled),
386 or to disable assembly of certain metadata types which might be
387 handled by other software.
388
389 The known metadata types are
390 .BR 0.90 ,
391 .BR 1.x ,
392 .BR ddf ,
393 .BR imsm .
394
395 .SH EXAMPLE
396 DEVICE /dev/sd[bcdjkl]1
397 .br
398 DEVICE /dev/hda1 /dev/hdb1
399
400 # /dev/md0 is known by its UUID.
401 .br
402 ARRAY /dev/md0 UUID=3aaa0122:29827cfa:5331ad66:ca767371
403 .br
404 # /dev/md1 contains all devices with a minor number of
405 .br
406 # 1 in the superblock.
407 .br
408 ARRAY /dev/md1 superminor=1
409 .br
410 # /dev/md2 is made from precisely these two devices
411 .br
412 ARRAY /dev/md2 devices=/dev/hda1,/dev/hdb1
413
414 # /dev/md4 and /dev/md5 are a spare-group and spares
415 .br
416 # can be moved between them
417 .br
418 ARRAY /dev/md4 uuid=b23f3c6d:aec43a9f:fd65db85:369432df
419 .br
420 spare\-group=group1
421 .br
422 ARRAY /dev/md5 uuid=19464854:03f71b1b:e0df2edd:246cc977
423 .br
424 spare\-group=group1
425 .br
426 # /dev/md/home is created if need to be a partitionable md array
427 .br
428 # any spare device number is allocated.
429 .br
430 ARRAY /dev/md/home UUID=9187a482:5dde19d9:eea3cc4a:d646ab8b
431 .br
432 auto=part
433
434 MAILADDR root@mydomain.tld
435 .br
436 PROGRAM /usr/sbin/handle\-mdadm\-events
437 .br
438 CREATE group=system mode=0640 auto=part\-8
439 .br
440 HOMEHOST <system>
441 .br
442 AUTO +1.x -all
443
444 .SH SEE ALSO
445 .BR mdadm (8),
446 .BR md (4).
447