]>
Commit | Line | Data |
---|---|---|
1 | .\" -*- nroff -*- | |
2 | .TH MDADM 8 "" v2.4.1 | |
3 | .SH NAME | |
4 | mdadm \- manage MD devices | |
5 | .I aka | |
6 | Linux Software Raid. | |
7 | ||
8 | .SH SYNOPSIS | |
9 | ||
10 | .BI mdadm " [mode] <raiddevice> [options] <component-devices>" | |
11 | ||
12 | .SH DESCRIPTION | |
13 | RAID devices are virtual devices created from two or more | |
14 | real block devices. This allows multiple devices (typically disk | |
15 | drives or partitions there-of) to be combined into a single device to | |
16 | hold (for example) a single filesystem. | |
17 | Some RAID levels include redundancy and so can survive some degree of | |
18 | device failure. | |
19 | ||
20 | Linux Software RAID devices are implemented through the md (Multiple | |
21 | Devices) device driver. | |
22 | ||
23 | Currently, Linux supports | |
24 | .B LINEAR | |
25 | md devices, | |
26 | .B RAID0 | |
27 | (striping), | |
28 | .B RAID1 | |
29 | (mirroring), | |
30 | .BR RAID4 , | |
31 | .BR RAID5 , | |
32 | .BR RAID6 , | |
33 | .BR RAID10 , | |
34 | .BR MULTIPATH , | |
35 | and | |
36 | .BR FAULTY . | |
37 | ||
38 | .B MULTIPATH | |
39 | is not a Software RAID mechanism, but does involve | |
40 | multiple devices. For | |
41 | .B MULTIPATH | |
42 | each device is a path to one common physical storage device. | |
43 | ||
44 | .B FAULTY | |
45 | is also not true RAID, and it only involves one device. It | |
46 | provides a layer over a true device that can be used to inject faults. | |
47 | ||
48 | '''.B mdadm | |
49 | '''is a program that can be used to create, manage, and monitor | |
50 | '''MD devices. As | |
51 | '''such it provides a similar set of functionality to the | |
52 | '''.B raidtools | |
53 | '''packages. | |
54 | '''The key differences between | |
55 | '''.B mdadm | |
56 | '''and | |
57 | '''.B raidtools | |
58 | '''are: | |
59 | '''.IP \(bu 4 | |
60 | '''.B mdadm | |
61 | '''is a single program and not a collection of programs. | |
62 | '''.IP \(bu 4 | |
63 | '''.B mdadm | |
64 | '''can perform (almost) all of its functions without having a | |
65 | '''configuration file and does not use one by default. Also | |
66 | '''.B mdadm | |
67 | '''helps with management of the configuration | |
68 | '''file. | |
69 | '''.IP \(bu 4 | |
70 | '''.B mdadm | |
71 | '''can provide information about your arrays (through Query, Detail, and Examine) | |
72 | '''that | |
73 | '''.B raidtools | |
74 | '''cannot. | |
75 | '''.P | |
76 | '''.I mdadm | |
77 | '''does not use | |
78 | '''.IR /etc/raidtab , | |
79 | '''the | |
80 | '''.B raidtools | |
81 | '''configuration file, at all. It has a different configuration file | |
82 | '''with a different format and a different purpose. | |
83 | ||
84 | .SH MODES | |
85 | mdadm has 7 major modes of operation: | |
86 | .TP | |
87 | .B Assemble | |
88 | Assemble the parts of a previously created | |
89 | array into an active array. Components can be explicitly given | |
90 | or can be searched for. | |
91 | .B mdadm | |
92 | checks that the components | |
93 | do form a bona fide array, and can, on request, fiddle superblock | |
94 | information so as to assemble a faulty array. | |
95 | ||
96 | .TP | |
97 | .B Build | |
98 | Build an array that doesn't have per-device superblocks. For these | |
99 | sorts of arrays, | |
100 | .I mdadm | |
101 | cannot differentiate between initial creation and subsequent assembly | |
102 | of an array. It also cannot perform any checks that appropriate | |
103 | devices have been requested. Because of this, the | |
104 | .B Build | |
105 | mode should only be used together with a complete understanding of | |
106 | what you are doing. | |
107 | ||
108 | .TP | |
109 | .B Create | |
110 | Create a new array with per-device superblocks. | |
111 | '''It can progress | |
112 | '''in several step create-add-add-run or it can all happen with one command. | |
113 | ||
114 | .TP | |
115 | .B "Follow or Monitor" | |
116 | Monitor one or more md devices and act on any state changes. This is | |
117 | only meaningful for raid1, 4, 5, 6, 10 or multipath arrays as | |
118 | only these have interesting state. raid0 or linear never have | |
119 | missing, spare, or failed drives, so there is nothing to monitor. | |
120 | ||
121 | .TP | |
122 | .B "Grow" | |
123 | Grow (or shrink) an array, or otherwise reshape it in some way. | |
124 | Currently supported growth options including changing the active size | |
125 | of component devices in RAID level 1/4/5/6 and changing the number of | |
126 | active devices in RAID1. | |
127 | ||
128 | .TP | |
129 | .B Manage | |
130 | This is for doing things to specific components of an array such as | |
131 | adding new spares and removing faulty devices. | |
132 | ||
133 | .TP | |
134 | .B Misc | |
135 | This is an 'everything else' mode that supports operations on active | |
136 | arrays, operations on component devices such as erasing old superblocks, and | |
137 | information gathering operations. | |
138 | '''This mode allows operations on independent devices such as examine MD | |
139 | '''superblocks, erasing old superblocks and stopping active arrays. | |
140 | ||
141 | .SH OPTIONS | |
142 | ||
143 | .SH Options for selecting a mode are: | |
144 | ||
145 | .TP | |
146 | .BR -A ", " --assemble | |
147 | Assemble a pre-existing array. | |
148 | ||
149 | .TP | |
150 | .BR -B ", " --build | |
151 | Build a legacy array without superblocks. | |
152 | ||
153 | .TP | |
154 | .BR -C ", " --create | |
155 | Create a new array. | |
156 | ||
157 | .TP | |
158 | .BR -F ", " --follow ", " --monitor | |
159 | Select | |
160 | .B Monitor | |
161 | mode. | |
162 | ||
163 | .TP | |
164 | .BR -G ", " --grow | |
165 | Change the size or shape of an active array. | |
166 | .P | |
167 | If a device is given before any options, or if the first option is | |
168 | .BR --add , | |
169 | .BR --fail , | |
170 | or | |
171 | .BR --remove , | |
172 | then the MANAGE mode is assume. | |
173 | Anything other than these will cause the | |
174 | .B Misc | |
175 | mode to be assumed. | |
176 | ||
177 | .SH Options that are not mode-specific are: | |
178 | ||
179 | .TP | |
180 | .BR -h ", " --help | |
181 | Display general help message or, after one of the above options, a | |
182 | mode specific help message. | |
183 | ||
184 | .TP | |
185 | .B --help-options | |
186 | Display more detailed help about command line parsing and some commonly | |
187 | used options. | |
188 | ||
189 | .TP | |
190 | .BR -V ", " --version | |
191 | Print version information for mdadm. | |
192 | ||
193 | .TP | |
194 | .BR -v ", " --verbose | |
195 | Be more verbose about what is happening. This can be used twice to be | |
196 | extra-verbose. | |
197 | The extra verbosity currently only affects | |
198 | .B --detail --scan | |
199 | and | |
200 | .BR "--examine --scan" . | |
201 | ||
202 | .TP | |
203 | .BR -q ", " --quiet | |
204 | Avoid printing purely informative messages. With this, | |
205 | .B mdadm | |
206 | will be silent unless there is something really important to report. | |
207 | ||
208 | .TP | |
209 | .BR -b ", " --brief | |
210 | Be less verbose. This is used with | |
211 | .B --detail | |
212 | and | |
213 | .BR --examine . | |
214 | Using | |
215 | .B --brief | |
216 | with | |
217 | .B --verbose | |
218 | gives an intermediate level of verbosity. | |
219 | ||
220 | .TP | |
221 | .BR -f ", " --force | |
222 | Be more forceful about certain operations. See the various modes of | |
223 | the exact meaning of this option in different contexts. | |
224 | ||
225 | .TP | |
226 | .BR -c ", " --config= | |
227 | Specify the config file. Default is to use | |
228 | .BR /etc/mdadm.conf , | |
229 | or if that is missing, then | |
230 | .BR /etc/mdadm/mdadm.conf . | |
231 | If the config file given is | |
232 | .B partitions | |
233 | then nothing will be read, but | |
234 | .I mdadm | |
235 | will act as though the config file contained exactly | |
236 | .B "DEVICE partitions" | |
237 | and will read | |
238 | .B /proc/partitions | |
239 | to find a list of devices to scan. | |
240 | If the word | |
241 | .B none | |
242 | is given for the config file, then | |
243 | .I mdadm | |
244 | will act as though the config file were empty. | |
245 | ||
246 | .TP | |
247 | .BR -s ", " --scan | |
248 | scan config file or | |
249 | .B /proc/mdstat | |
250 | for missing information. | |
251 | In general, this option gives | |
252 | .B mdadm | |
253 | permission to get any missing information, like component devices, | |
254 | array devices, array identities, and alert destination from the | |
255 | configuration file: | |
256 | .BR /etc/mdadm.conf . | |
257 | One exception is MISC mode when using | |
258 | .B --detail | |
259 | or | |
260 | .B --stop | |
261 | in which case | |
262 | .B --scan | |
263 | says to get a list of array devices from | |
264 | .BR /proc/mdstat . | |
265 | ||
266 | .TP | |
267 | .B -e ", " --metadata= | |
268 | Declare the style of superblock (raid metadata) to be used. The | |
269 | default is 0.90 for --create, and to guess for other operations. | |
270 | ||
271 | Options are: | |
272 | .RS | |
273 | .IP "0, 0.90, default" | |
274 | Use the original 0.90 format superblock. This format limits arrays to | |
275 | 28 componenet devices and limits component devices of levels 1 and | |
276 | greater to 2 terabytes. | |
277 | .IP "1, 1.0, 1.1, 1.2" | |
278 | Use the new version-1 format superblock. This has few restrictions. | |
279 | The different subversion store the superblock at different locations | |
280 | on the device, either at the end (for 1.0), at the start (for 1.1) or | |
281 | 4K from the start (for 1.2). | |
282 | .RE | |
283 | ||
284 | .SH For create, build, or grow: | |
285 | ||
286 | .TP | |
287 | .BR -n ", " --raid-devices= | |
288 | Specify the number of active devices in the array. This, plus the | |
289 | number of spare devices (see below) must equal the number of | |
290 | .I component-devices | |
291 | (including "\fBmissing\fP" devices) | |
292 | that are listed on the command line for | |
293 | .BR --create . | |
294 | Setting a value of 1 is probably | |
295 | a mistake and so requires that | |
296 | .B --force | |
297 | be specified first. A value of 1 will then be allowed for linear, | |
298 | multipath, raid0 and raid1. It is never allowed for raid4 or raid5. | |
299 | .br | |
300 | This number can only be changed using | |
301 | .B --grow | |
302 | for RAID1 arrays, and only on kernels which provide necessary support. | |
303 | ||
304 | .TP | |
305 | .BR -x ", " --spare-devices= | |
306 | Specify the number of spare (eXtra) devices in the initial array. | |
307 | Spares can also be added | |
308 | and removed later. The number of component devices listed | |
309 | on the command line must equal the number of raid devices plus the | |
310 | number of spare devices. | |
311 | ||
312 | ||
313 | .TP | |
314 | .BR -z ", " --size= | |
315 | Amount (in Kibibytes) of space to use from each drive in RAID1/4/5/6. | |
316 | This must be a multiple of the chunk size, and must leave about 128Kb | |
317 | of space at the end of the drive for the RAID superblock. | |
318 | If this is not specified | |
319 | (as it normally is not) the smallest drive (or partition) sets the | |
320 | size, though if there is a variance among the drives of greater than 1%, a warning is | |
321 | issued. | |
322 | ||
323 | This value can be set with | |
324 | .B --grow | |
325 | for RAID level 1/4/5/6. If the array was created with a size smaller | |
326 | than the currently active drives, the extra space can be accessed | |
327 | using | |
328 | .BR --grow . | |
329 | The size can be given as | |
330 | .B max | |
331 | which means to choose the largest size that fits on all current drives. | |
332 | ||
333 | .TP | |
334 | .BR -c ", " --chunk= | |
335 | Specify chunk size of kibibytes. The default is 64. | |
336 | ||
337 | .TP | |
338 | .BR --rounding= | |
339 | Specify rounding factor for linear array (==chunk size) | |
340 | ||
341 | .TP | |
342 | .BR -l ", " --level= | |
343 | Set raid level. When used with | |
344 | .IR --create , | |
345 | options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4, | |
346 | raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty. Obviously some of these are synonymous. | |
347 | ||
348 | When used with | |
349 | .IR --build , | |
350 | only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid. | |
351 | ||
352 | Not yet supported with | |
353 | .IR --grow . | |
354 | ||
355 | .TP | |
356 | .BR -p ", " --layout= | |
357 | This option configures the fine details of data layout for raid5, | |
358 | and raid10 arrays, and controls the failure modes for | |
359 | .IR faulty . | |
360 | ||
361 | The layout of the raid5 parity block can be one of | |
362 | left-asymmetric, | |
363 | left-symmetric, | |
364 | right-asymmetric, | |
365 | right-symmetric, | |
366 | la, ra, ls, rs. The default is left-symmetric. | |
367 | ||
368 | When setting the failure mode for | |
369 | .I faulty | |
370 | the options are: | |
371 | write-transient, | |
372 | wt, | |
373 | read-transient, | |
374 | rt, | |
375 | write-persistent, | |
376 | wp, | |
377 | read-persistent, | |
378 | rp, | |
379 | write-all, | |
380 | read-fixable, | |
381 | rf, | |
382 | clear, | |
383 | flush, | |
384 | none. | |
385 | ||
386 | Each mode can be followed by a number which is used as a period | |
387 | between fault generation. Without a number, the fault is generated | |
388 | once on the first relevant request. With a number, the fault will be | |
389 | generated after that many request, and will continue to be generated | |
390 | every time the period elapses. | |
391 | ||
392 | Multiple failure modes can be current simultaneously by using the | |
393 | "--grow" option to set subsequent failure modes. | |
394 | ||
395 | "clear" or "none" will remove any pending or periodic failure modes, | |
396 | and "flush" will clear any persistent faults. | |
397 | ||
398 | To set the parity with "--grow", the level of the array ("faulty") | |
399 | must be specified before the fault mode is specified. | |
400 | ||
401 | Finally, the layout options for RAID10 are one of 'n', 'o' or 'p' followed | |
402 | by a small number. The default is 'n2'. | |
403 | ||
404 | .I n | |
405 | signals 'near' copies. Multiple copies of one data block are at | |
406 | similar offsets in different devices. | |
407 | ||
408 | .I o | |
409 | signals 'offset' copies. Rather than the chunks being duplicated | |
410 | within a stripe, whole stripes are duplicated but are rotated by one | |
411 | device so duplicate blocks are on different devices. Thus subsequent | |
412 | copies of a block are in the next drive, and are one chunk further | |
413 | down. | |
414 | ||
415 | .I f | |
416 | signals 'far' copies | |
417 | (multiple copies have very different offsets). See md(4) for more | |
418 | detail about 'near' and 'far'. | |
419 | ||
420 | The number is the number of copies of each datablock. 2 is normal, 3 | |
421 | can be useful. This number can be at most equal to the number of | |
422 | devices in the array. It does not need to divide evenly into that | |
423 | number (e.g. it is perfectly legal to have an 'n2' layout for an array | |
424 | with an odd number of devices). | |
425 | ||
426 | .TP | |
427 | .BR --parity= | |
428 | same as --layout (thus explaining the p of | |
429 | .IR -p ). | |
430 | ||
431 | .TP | |
432 | .BR -b ", " --bitmap= | |
433 | Specify a file to store a write-intent bitmap in. The file should not | |
434 | exist unless --force is also given. The same file should be provided | |
435 | when assembling the array. If the word | |
436 | .B internal | |
437 | is given, then the bitmap is stored with the metadata on the array, | |
438 | and so is replicated on all devices. If the word | |
439 | .B none | |
440 | is given with | |
441 | .B --grow | |
442 | mode, then any bitmap that is present is removed. | |
443 | ||
444 | To help catch typing errors, the filename must contain at least one | |
445 | slash ('/') if it is a real file (not 'internal' or 'none'). | |
446 | ||
447 | Note: external bitmaps are only known to work on ext2 and ext3. | |
448 | Storing bitmap files on other filesystems may result in serious problems. | |
449 | ||
450 | .TP | |
451 | .BR --bitmap-chunk= | |
452 | Set the chunksize of the bitmap. Each bit corresponds to that many | |
453 | Kilobytes of storage. | |
454 | When using a file based bitmap, the default is to use the smallest | |
455 | size that is atleast 4 and requires no more than 2^21 chunks. | |
456 | When using an | |
457 | .B internal | |
458 | bitmap, the chunksize is automatically determined to make best use of | |
459 | available space. | |
460 | ||
461 | ||
462 | .TP | |
463 | .BR -W ", " --write-mostly | |
464 | subsequent devices lists in a | |
465 | .BR --build , | |
466 | .BR --create , | |
467 | or | |
468 | .B --add | |
469 | command will be flagged as 'write-mostly'. This is valid for RAID1 | |
470 | only and means that the 'md' driver will avoid reading from these | |
471 | devices if at all possible. This can be useful if mirroring over a | |
472 | slow link. | |
473 | ||
474 | .TP | |
475 | .BR --write-behind= | |
476 | Specify that write-behind mode should be enabled (valid for RAID1 | |
477 | only). If an argument is specified, it will set the maximum number | |
478 | of outstanding writes allowed. The default value is 256. | |
479 | A write-intent bitmap is required in order to use write-behind | |
480 | mode, and write-behind is only attempted on drives marked as | |
481 | .IR write-mostly . | |
482 | ||
483 | .TP | |
484 | .BR --assume-clean | |
485 | Tell | |
486 | .I mdadm | |
487 | that the array pre-existed and is known to be clean. It can be useful | |
488 | when trying to recover from a major failure as you can be sure that no | |
489 | data will be affected unless you actually write to the array. It can | |
490 | also be used when creating a RAID1 or RAID10 if you want to avoid the | |
491 | initial resync, however this practice - while normally safe - is not | |
492 | recommended. Use this ony if you really know what you are doing. | |
493 | ||
494 | .TP | |
495 | .BR --backup-file= | |
496 | This is needed when --grow is used to increase the number of | |
497 | raid-devices in a RAID5 if there are no spare devices available. | |
498 | See the section below on RAID_DEVICE CHANGES. The file should be | |
499 | stored on a separate device, not on the raid array being reshaped. | |
500 | ||
501 | .TP | |
502 | .BR -N ", " --name= | |
503 | Set a | |
504 | .B name | |
505 | for the array. This is currently only effective when creating an | |
506 | array with a version-1 superblock. The name is a simple textual | |
507 | string that can be used to identify array components when assembling. | |
508 | ||
509 | .TP | |
510 | .BR -R ", " --run | |
511 | Insist that | |
512 | .I mdadm | |
513 | run the array, even if some of the components | |
514 | appear to be active in another array or filesystem. Normally | |
515 | .I mdadm | |
516 | will ask for confirmation before including such components in an | |
517 | array. This option causes that question to be suppressed. | |
518 | ||
519 | .TP | |
520 | .BR -f ", " --force | |
521 | Insist that | |
522 | .I mdadm | |
523 | accept the geometry and layout specified without question. Normally | |
524 | .I mdadm | |
525 | will not allow creation of an array with only one device, and will try | |
526 | to create a raid5 array with one missing drive (as this makes the | |
527 | initial resync work faster). With | |
528 | .BR --force , | |
529 | .I mdadm | |
530 | will not try to be so clever. | |
531 | ||
532 | .TP | |
533 | .BR -a ", " "--auto{=no,yes,md,mdp,part,p}{NN}" | |
534 | Instruct mdadm to create the device file if needed, possibly allocating | |
535 | an unused minor number. "md" causes a non-partitionable array | |
536 | to be used. "mdp", "part" or "p" causes a partitionable array (2.6 and | |
537 | later) to be used. "yes" requires the named md device to have | |
538 | a 'standard' format, and the type and minor number will be determined | |
539 | from this. See DEVICE NAMES below. | |
540 | ||
541 | The argument can also come immediately after | |
542 | "-a". e.g. "-ap". | |
543 | ||
544 | If | |
545 | .I --scan | |
546 | is also given, then any | |
547 | .I auto= | |
548 | entries in the config file will over-ride the | |
549 | .I --auto | |
550 | instruction given on the command line. | |
551 | ||
552 | For partitionable arrays, | |
553 | .I mdadm | |
554 | will create the device file for the whole array and for the first 4 | |
555 | partitions. A different number of partitions can be specified at the | |
556 | end of this option (e.g. | |
557 | .BR --auto=p7 ). | |
558 | If the device name ends with a digit, the partition names add a 'p', | |
559 | and a number, e.g. "/dev/home1p3". If there is no | |
560 | trailing digit, then the partition names just have a number added, | |
561 | e.g. "/dev/scratch3". | |
562 | ||
563 | If the md device name is in a 'standard' format as described in DEVICE | |
564 | NAMES, then it will be created, if necessary, with the appropriate | |
565 | number based on that name. If the device name is not in one of these | |
566 | formats, then a unused minor number will be allocated. The minor | |
567 | number will be considered unused if there is no active array for that | |
568 | number, and there is no entry in /dev for that number and with a | |
569 | non-standard name. | |
570 | ||
571 | .SH For assemble: | |
572 | ||
573 | .TP | |
574 | .BR -u ", " --uuid= | |
575 | uuid of array to assemble. Devices which don't have this uuid are | |
576 | excluded | |
577 | ||
578 | .TP | |
579 | .BR -m ", " --super-minor= | |
580 | Minor number of device that array was created for. Devices which | |
581 | don't have this minor number are excluded. If you create an array as | |
582 | /dev/md1, then all superblocks will contain the minor number 1, even if | |
583 | the array is later assembled as /dev/md2. | |
584 | ||
585 | Giving the literal word "dev" for | |
586 | .B --super-minor | |
587 | will cause | |
588 | .I mdadm | |
589 | to use the minor number of the md device that is being assembled. | |
590 | e.g. when assembling | |
591 | .BR /dev/md0 , | |
592 | .M --super-minor=dev | |
593 | will look for super blocks with a minor number of 0. | |
594 | ||
595 | .TP | |
596 | .BR -N ", " --name= | |
597 | Specify the name of the array to assemble. This must be the name | |
598 | that was specified when creating the array. | |
599 | ||
600 | .TP | |
601 | .BR -f ", " --force | |
602 | Assemble the array even if some superblocks appear out-of-date | |
603 | ||
604 | .TP | |
605 | .BR -R ", " --run | |
606 | Attempt to start the array even if fewer drives were given than are | |
607 | needed for a full array. Normally if not all drives are found and | |
608 | .B --scan | |
609 | is not used, then the array will be assembled but not started. | |
610 | With | |
611 | .B --run | |
612 | an attempt will be made to start it anyway. | |
613 | ||
614 | .TP | |
615 | .BR -a ", " "--auto{=no,yes,md,mdp,part}" | |
616 | See this option under Create and Build options. | |
617 | ||
618 | .TP | |
619 | .BR -b ", " --bitmap= | |
620 | Specify the bitmap file that was given when the array was created. If | |
621 | an array has an | |
622 | .B internal | |
623 | bitmap, there is no need to specify this when assembling the array. | |
624 | ||
625 | .TP | |
626 | .BR --backup-file= | |
627 | If | |
628 | .B --backup-file | |
629 | was used to grow the number of raid-devices in a RAID5, and the system | |
630 | crashed during the critical section, then the same | |
631 | .B --backup-file | |
632 | must be presented to --assemble to allow possibly corrupted data to be | |
633 | restored. | |
634 | ||
635 | .TP | |
636 | .BR -U ", " --update= | |
637 | Update the superblock on each device while assembling the array. The | |
638 | argument given to this flag can be one of | |
639 | .BR sparc2.2 , | |
640 | .BR summaries , | |
641 | .BR uuid , | |
642 | .BR resync , | |
643 | .BR byteorder , | |
644 | or | |
645 | .BR super-minor . | |
646 | ||
647 | The | |
648 | .B sparc2.2 | |
649 | option will adjust the superblock of an array what was created on a Sparc | |
650 | machine running a patched 2.2 Linux kernel. This kernel got the | |
651 | alignment of part of the superblock wrong. You can use the | |
652 | .B "--examine --sparc2.2" | |
653 | option to | |
654 | .I mdadm | |
655 | to see what effect this would have. | |
656 | ||
657 | The | |
658 | .B super-minor | |
659 | option will update the | |
660 | .B "preferred minor" | |
661 | field on each superblock to match the minor number of the array being | |
662 | assembled. This is not needed on 2.6 and later kernels as they make | |
663 | this adjustment automatically. | |
664 | ||
665 | The | |
666 | .B uuid | |
667 | option will change the uuid of the array. If a UUID is given with the | |
668 | "--uuid" option that UUID will be used as a new UUID and with | |
669 | .B NOT | |
670 | be used to help identify the devices in the array. | |
671 | If no "--uuid" is given, a random uuid is chosen. | |
672 | ||
673 | The | |
674 | .B resync | |
675 | option will cause the array to be marked | |
676 | .I dirty | |
677 | meaning that any redundancy in the array (e.g. parity for raid5, | |
678 | copies for raid1) may be incorrect. This will cause the raid system | |
679 | to perform a "resync" pass to make sure that all redundant information | |
680 | is correct. | |
681 | ||
682 | The | |
683 | .B byteorder | |
684 | option allows arrays to be moved between machines with different | |
685 | byte-order. | |
686 | When assembling such an array for the first time after a move, giving | |
687 | .B "--update=byteorder" | |
688 | will cause | |
689 | .I mdadm | |
690 | to expect superblocks to have their byteorder reversed, and will | |
691 | correct that order before assembling the array. This is only valid | |
692 | with original (Version 0.90) superblocks. | |
693 | ||
694 | The | |
695 | .B summaries | |
696 | option will correct the summaries in the superblock. That is the | |
697 | counts of total, working, active, failed, and spare devices. | |
698 | ||
699 | .SH For Manage mode: | |
700 | ||
701 | .TP | |
702 | .BR -a ", " --add | |
703 | hot-add listed devices. | |
704 | ||
705 | .TP | |
706 | .BR --re-add | |
707 | re-add a device that was recently removed from an array. | |
708 | ||
709 | .TP | |
710 | .BR -r ", " --remove | |
711 | remove listed devices. They must not be active. i.e. they should | |
712 | be failed or spare devices. | |
713 | ||
714 | .TP | |
715 | .BR -f ", " --fail | |
716 | mark listed devices as faulty. | |
717 | ||
718 | .TP | |
719 | .BR --set-faulty | |
720 | same as --fail. | |
721 | ||
722 | .P | |
723 | Each of these options require that the first device list is the array | |
724 | to be acted upon and the remainder are component devices to be added, | |
725 | removed, or marked as fault. Several different operations can be | |
726 | specified for different devices, e.g. | |
727 | .in +5 | |
728 | mdadm /dev/md0 --add /dev/sda1 --fail /dev/sdb1 --remove /dev/sdb1 | |
729 | .in -5 | |
730 | Each operation applies to all devices listed until the next | |
731 | operations. | |
732 | ||
733 | If an array is using a write-intent bitmap, then devices which have | |
734 | been removed can be re-added in a way that avoids a full | |
735 | reconstruction but instead just updated the blocks that have changed | |
736 | since the device was removed. For arrays with persistent metadata | |
737 | (superblocks) this is done automatically. For arrays created with | |
738 | .B --build | |
739 | mdadm needs to be told that this device we removed recently with | |
740 | .B --re-add. | |
741 | ||
742 | Devices can only be removed from an array if they are not in active | |
743 | use. i.e. that must be spares or failed devices. To remove an active | |
744 | device, it must be marked as | |
745 | .B faulty | |
746 | first. | |
747 | ||
748 | .SH For Misc mode: | |
749 | ||
750 | .TP | |
751 | .BR -Q ", " --query | |
752 | Examine a device to see | |
753 | (1) if it is an md device and (2) if it is a component of an md | |
754 | array. | |
755 | Information about what is discovered is presented. | |
756 | ||
757 | .TP | |
758 | .BR -D ", " --detail | |
759 | Print detail of one or more md devices. | |
760 | ||
761 | .TP | |
762 | .BR -E ", " --examine | |
763 | Print content of md superblock on device(s). | |
764 | .TP | |
765 | .B --sparc2.2 | |
766 | If an array was created on a 2.2 Linux kernel patched with RAID | |
767 | support, the superblock will have been created incorrectly, or at | |
768 | least incompatibly with 2.4 and later kernels. Using the | |
769 | .B --sparc2.2 | |
770 | flag with | |
771 | .B --examine | |
772 | will fix the superblock before displaying it. If this appears to do | |
773 | the right thing, then the array can be successfully assembled using | |
774 | .BR "--assemble --update=sparc2.2" . | |
775 | ||
776 | .TP | |
777 | .BR -X ", " --examine-bitmap | |
778 | Report information about a bitmap file. | |
779 | ||
780 | .TP | |
781 | .BR -R ", " --run | |
782 | start a partially built array. | |
783 | ||
784 | .TP | |
785 | .BR -S ", " --stop | |
786 | deactivate array, releasing all resources. | |
787 | ||
788 | .TP | |
789 | .BR -o ", " --readonly | |
790 | mark array as readonly. | |
791 | ||
792 | .TP | |
793 | .BR -w ", " --readwrite | |
794 | mark array as readwrite. | |
795 | ||
796 | .TP | |
797 | .B --zero-superblock | |
798 | If the device contains a valid md superblock, the block is | |
799 | over-written with zeros. With | |
800 | --force | |
801 | the block where the superblock would be is over-written even if it | |
802 | doesn't appear to be valid. | |
803 | ||
804 | .TP | |
805 | .BR -t ", " --test | |
806 | When used with | |
807 | .BR --detail , | |
808 | the exit status of | |
809 | .I mdadm | |
810 | is set to reflect the status of the device. | |
811 | ||
812 | .SH For Monitor mode: | |
813 | .TP | |
814 | .BR -m ", " --mail | |
815 | Give a mail address to send alerts to. | |
816 | ||
817 | .TP | |
818 | .BR -p ", " --program ", " --alert | |
819 | Give a program to be run whenever an event is detected. | |
820 | ||
821 | .TP | |
822 | .BR -y ", " --syslog | |
823 | Cause all events to be reported through 'syslog'. The messages have | |
824 | facility of 'daemon' and varying priorities. | |
825 | ||
826 | .TP | |
827 | .BR -d ", " --delay | |
828 | Give a delay in seconds. | |
829 | .B mdadm | |
830 | polls the md arrays and then waits this many seconds before polling | |
831 | again. The default is 60 seconds. | |
832 | ||
833 | .TP | |
834 | .BR -f ", " --daemonise | |
835 | Tell | |
836 | .B mdadm | |
837 | to run as a background daemon if it decides to monitor anything. This | |
838 | causes it to fork and run in the child, and to disconnect form the | |
839 | terminal. The process id of the child is written to stdout. | |
840 | This is useful with | |
841 | .B --scan | |
842 | which will only continue monitoring if a mail address or alert program | |
843 | is found in the config file. | |
844 | ||
845 | .TP | |
846 | .BR -i ", " --pid-file | |
847 | When | |
848 | .B mdadm | |
849 | is running in daemon mode, write the pid of the daemon process to | |
850 | the specified file, instead of printing it on standard output. | |
851 | ||
852 | .TP | |
853 | .BR -1 ", " --oneshot | |
854 | Check arrays only once. This will generate | |
855 | .B NewArray | |
856 | events and more significantly | |
857 | .B DegradedArray | |
858 | and | |
859 | .B SparesMissing | |
860 | events. Running | |
861 | .in +5 | |
862 | .B " mdadm --monitor --scan -1" | |
863 | .in -5 | |
864 | from a cron script will ensure regular notification of any degraded arrays. | |
865 | ||
866 | .TP | |
867 | .BR -t ", " --test | |
868 | Generate a | |
869 | .B TestMessage | |
870 | alert for every array found at startup. This alert gets mailed and | |
871 | passed to the alert program. This can be used for testing that alert | |
872 | message do get through successfully. | |
873 | ||
874 | .SH ASSEMBLE MODE | |
875 | ||
876 | .HP 12 | |
877 | Usage: | |
878 | .B mdadm --assemble | |
879 | .I md-device options-and-component-devices... | |
880 | .HP 12 | |
881 | Usage: | |
882 | .B mdadm --assemble --scan | |
883 | .I md-devices-and-options... | |
884 | .HP 12 | |
885 | Usage: | |
886 | .B mdadm --assemble --scan | |
887 | .I options... | |
888 | ||
889 | .PP | |
890 | This usage assembles one or more raid arrays from pre-existing components. | |
891 | For each array, mdadm needs to know the md device, the identity of the | |
892 | array, and a number of component-devices. These can be found in a number of ways. | |
893 | ||
894 | In the first usage example (without the | |
895 | .BR --scan ) | |
896 | the first device given is the md device. | |
897 | In the second usage example, all devices listed are treated as md | |
898 | devices and assembly is attempted. | |
899 | In the third (where no devices are listed) all md devices that are | |
900 | listed in the configuration file are assembled. | |
901 | ||
902 | If precisely one device is listed, but | |
903 | .B --scan | |
904 | is not given, then | |
905 | .I mdadm | |
906 | acts as though | |
907 | .B --scan | |
908 | was given and identify information is extracted from the configuration file. | |
909 | ||
910 | The identity can be given with the | |
911 | .B --uuid | |
912 | option, with the | |
913 | .B --super-minor | |
914 | option, can be found in the config file, or will be taken from the | |
915 | super block on the first component-device listed on the command line. | |
916 | ||
917 | Devices can be given on the | |
918 | .B --assemble | |
919 | command line or in the config file. Only devices which have an md | |
920 | superblock which contains the right identity will be considered for | |
921 | any array. | |
922 | ||
923 | The config file is only used if explicitly named with | |
924 | .B --config | |
925 | or requested with (a possibly implicit) | |
926 | .B --scan. | |
927 | In the later case, | |
928 | .B /etc/mdadm.conf | |
929 | is used. | |
930 | ||
931 | If | |
932 | .B --scan | |
933 | is not given, then the config file will only be used to find the | |
934 | identity of md arrays. | |
935 | ||
936 | Normally the array will be started after it is assembled. However if | |
937 | .B --scan | |
938 | is not given and insufficient drives were listed to start a complete | |
939 | (non-degraded) array, then the array is not started (to guard against | |
940 | usage errors). To insist that the array be started in this case (as | |
941 | may work for RAID1, 4, 5, 6, or 10), give the | |
942 | .B --run | |
943 | flag. | |
944 | ||
945 | If an | |
946 | .B auto | |
947 | option is given, either on the command line (--auto) or in the | |
948 | configuration file (e.g. auto=part), then | |
949 | .I mdadm | |
950 | will create the md device if necessary or will re-create it if it | |
951 | doesn't look usable as it is. | |
952 | ||
953 | This can be useful for handling partitioned devices (which don't have | |
954 | a stable device number - it can change after a reboot) and when using | |
955 | "udev" to manage your | |
956 | .B /dev | |
957 | tree (udev cannot handle md devices because of the unusual device | |
958 | initialisation conventions). | |
959 | ||
960 | If the option to "auto" is "mdp" or "part" or (on the command line | |
961 | only) "p", then mdadm will create a partitionable array, using the | |
962 | first free one that is not in use, and does not already have an entry | |
963 | in /dev (apart from numeric /dev/md* entries). | |
964 | ||
965 | If the option to "auto" is "yes" or "md" or (on the command line) | |
966 | nothing, then mdadm will create a traditional, non-partitionable md | |
967 | array. | |
968 | ||
969 | It is expected that the "auto" functionality will be used to create | |
970 | device entries with meaningful names such as "/dev/md/home" or | |
971 | "/dev/md/root", rather than names based on the numerical array number. | |
972 | ||
973 | When using this option to create a partitionable array, the device | |
974 | files for the first 4 partitions are also created. If a different | |
975 | number is required it can be simply appended to the auto option. | |
976 | e.g. "auto=part8". Partition names are created by appending a digit | |
977 | string to the device name, with an intervening "p" if the device name | |
978 | ends with a digit. | |
979 | ||
980 | The | |
981 | .B --auto | |
982 | option is also available in Build and Create modes. As those modes do | |
983 | not use a config file, the "auto=" config option does not apply to | |
984 | these modes. | |
985 | ||
986 | .SH BUILD MODE | |
987 | ||
988 | .HP 12 | |
989 | Usage: | |
990 | .B mdadm --build | |
991 | .I device | |
992 | .BI --chunk= X | |
993 | .BI --level= Y | |
994 | .BI --raid-devices= Z | |
995 | .I devices | |
996 | ||
997 | .PP | |
998 | This usage is similar to | |
999 | .BR --create . | |
1000 | The difference is that it creates an array without a superblock. With | |
1001 | these arrays there is no difference between initially creating the array and | |
1002 | subsequently assembling the array, except that hopefully there is useful | |
1003 | data there in the second case. | |
1004 | ||
1005 | The level may raid0, linear, multipath, or faulty, or one of their | |
1006 | synonyms. All devices must be listed and the array will be started | |
1007 | once complete. | |
1008 | ||
1009 | .SH CREATE MODE | |
1010 | ||
1011 | .HP 12 | |
1012 | Usage: | |
1013 | .B mdadm --create | |
1014 | .I device | |
1015 | .BI --chunk= X | |
1016 | .BI --level= Y | |
1017 | .br | |
1018 | .BI --raid-devices= Z | |
1019 | .I devices | |
1020 | ||
1021 | .PP | |
1022 | This usage will initialise a new md array, associate some devices with | |
1023 | it, and activate the array. | |
1024 | ||
1025 | If the | |
1026 | .B --auto | |
1027 | option is given (as described in more detail in the section on | |
1028 | Assemble mode), then the md device will be created with a suitable | |
1029 | device number if necessary. | |
1030 | ||
1031 | As devices are added, they are checked to see if they contain raid | |
1032 | superblocks or filesystems. They are also checked to see if the variance in | |
1033 | device size exceeds 1%. | |
1034 | ||
1035 | If any discrepancy is found, the array will not automatically be run, though | |
1036 | the presence of a | |
1037 | .B --run | |
1038 | can override this caution. | |
1039 | ||
1040 | To create a "degraded" array in which some devices are missing, simply | |
1041 | give the word "\fBmissing\fP" | |
1042 | in place of a device name. This will cause | |
1043 | .B mdadm | |
1044 | to leave the corresponding slot in the array empty. | |
1045 | For a RAID4 or RAID5 array at most one slot can be | |
1046 | "\fBmissing\fP"; for a RAID6 array at most two slots. | |
1047 | For a RAID1 array, only one real device needs to be given. All of the | |
1048 | others can be | |
1049 | "\fBmissing\fP". | |
1050 | ||
1051 | When creating a RAID5 array, | |
1052 | .B mdadm | |
1053 | will automatically create a degraded array with an extra spare drive. | |
1054 | This is because building the spare into a degraded array is in general faster than resyncing | |
1055 | the parity on a non-degraded, but not clean, array. This feature can | |
1056 | be over-ridden with the | |
1057 | .I --force | |
1058 | option. | |
1059 | ||
1060 | '''If the | |
1061 | '''.B --size | |
1062 | '''option is given, it is not necessary to list any component-devices in this command. | |
1063 | '''They can be added later, before a | |
1064 | '''.B --run. | |
1065 | '''If no | |
1066 | '''.B --size | |
1067 | '''is given, the apparent size of the smallest drive given is used. | |
1068 | ||
1069 | The General Management options that are valid with --create are: | |
1070 | .TP | |
1071 | .B --run | |
1072 | insist on running the array even if some devices look like they might | |
1073 | be in use. | |
1074 | ||
1075 | .TP | |
1076 | .B --readonly | |
1077 | start the array readonly - not supported yet. | |
1078 | ||
1079 | ||
1080 | .SH MANAGE MODE | |
1081 | .HP 12 | |
1082 | Usage: | |
1083 | .B mdadm | |
1084 | .I device | |
1085 | .I options... devices... | |
1086 | .PP | |
1087 | ||
1088 | This usage will allow individual devices in an array to be failed, | |
1089 | removed or added. It is possible to perform multiple operations with | |
1090 | on command. For example: | |
1091 | .br | |
1092 | .B " mdadm /dev/md0 -f /dev/hda1 -r /dev/hda1 -a /dev/hda1" | |
1093 | .br | |
1094 | will firstly mark | |
1095 | .B /dev/hda1 | |
1096 | as faulty in | |
1097 | .B /dev/md0 | |
1098 | and will then remove it from the array and finally add it back | |
1099 | in as a spare. However only one md array can be affected by a single | |
1100 | command. | |
1101 | ||
1102 | .SH MISC MODE | |
1103 | .HP 12 | |
1104 | Usage: | |
1105 | .B mdadm | |
1106 | .I options ... | |
1107 | .I devices ... | |
1108 | .PP | |
1109 | ||
1110 | MISC mode includes a number of distinct operations that | |
1111 | operate on distinct devices. The operations are: | |
1112 | .TP | |
1113 | --query | |
1114 | The device is examined to see if it is | |
1115 | (1) an active md array, or | |
1116 | (2) a component of an md array. | |
1117 | The information discovered is reported. | |
1118 | ||
1119 | .TP | |
1120 | --detail | |
1121 | The device should be an active md device. | |
1122 | .B mdadm | |
1123 | will display a detailed description of the array. | |
1124 | .B --brief | |
1125 | or | |
1126 | .B --scan | |
1127 | will cause the output to be less detailed and the format to be | |
1128 | suitable for inclusion in | |
1129 | .BR /etc/mdadm.conf . | |
1130 | The exit status of | |
1131 | .I mdadm | |
1132 | will normally be 0 unless | |
1133 | .I mdadm | |
1134 | failed to get useful information about the device(s). However if the | |
1135 | .B --test | |
1136 | option is given, then the exit status will be: | |
1137 | .RS | |
1138 | .TP | |
1139 | 0 | |
1140 | The array is functioning normally. | |
1141 | .TP | |
1142 | 1 | |
1143 | The array has at least one failed device. | |
1144 | .TP | |
1145 | 2 | |
1146 | The array has multiple failed devices and hence is unusable (raid4 or | |
1147 | raid5). | |
1148 | .TP | |
1149 | 4 | |
1150 | There was an error while trying to get information about the device. | |
1151 | .RE | |
1152 | ||
1153 | .TP | |
1154 | --examine | |
1155 | The device should be a component of an md array. | |
1156 | .B mdadm | |
1157 | will read the md superblock of the device and display the contents. | |
1158 | If | |
1159 | .B --brief | |
1160 | is given, or | |
1161 | .B --scan | |
1162 | then multiple devices that are components of the one array | |
1163 | are grouped together and reported in a single entry suitable | |
1164 | for inclusion in | |
1165 | .BR /etc/mdadm.conf . | |
1166 | ||
1167 | Having | |
1168 | .B --scan | |
1169 | without listing any devices will cause all devices listed in the | |
1170 | config file to be examined. | |
1171 | ||
1172 | .TP | |
1173 | --stop | |
1174 | The devices should be active md arrays which will be deactivated, as | |
1175 | long as they are not currently in use. | |
1176 | ||
1177 | .TP | |
1178 | --run | |
1179 | This will fully activate a partially assembled md array. | |
1180 | ||
1181 | .TP | |
1182 | --readonly | |
1183 | This will mark an active array as read-only, providing that it is | |
1184 | not currently being used. | |
1185 | ||
1186 | .TP | |
1187 | --readwrite | |
1188 | This will change a | |
1189 | .B readonly | |
1190 | array back to being read/write. | |
1191 | ||
1192 | .TP | |
1193 | --scan | |
1194 | For all operations except | |
1195 | .BR --examine , | |
1196 | .B --scan | |
1197 | will cause the operation to be applied to all arrays listed in | |
1198 | .BR /proc/mdstat . | |
1199 | For | |
1200 | .BR --examine, | |
1201 | .B --scan | |
1202 | causes all devices listed in the config file to be examined. | |
1203 | ||
1204 | ||
1205 | .SH MONITOR MODE | |
1206 | ||
1207 | .HP 12 | |
1208 | Usage: | |
1209 | .B mdadm --monitor | |
1210 | .I options... devices... | |
1211 | ||
1212 | .PP | |
1213 | This usage causes | |
1214 | .B mdadm | |
1215 | to periodically poll a number of md arrays and to report on any events | |
1216 | noticed. | |
1217 | .B mdadm | |
1218 | will never exit once it decides that there are arrays to be checked, | |
1219 | so it should normally be run in the background. | |
1220 | ||
1221 | As well as reporting events, | |
1222 | .B mdadm | |
1223 | may move a spare drive from one array to another if they are in the | |
1224 | same | |
1225 | .B spare-group | |
1226 | and if the destination array has a failed drive but no spares. | |
1227 | ||
1228 | If any devices are listed on the command line, | |
1229 | .B mdadm | |
1230 | will only monitor those devices. Otherwise all arrays listed in the | |
1231 | configuration file will be monitored. Further, if | |
1232 | .B --scan | |
1233 | is given, then any other md devices that appear in | |
1234 | .B /proc/mdstat | |
1235 | will also be monitored. | |
1236 | ||
1237 | The result of monitoring the arrays is the generation of events. | |
1238 | These events are passed to a separate program (if specified) and may | |
1239 | be mailed to a given E-mail address. | |
1240 | ||
1241 | When passing event to program, the program is run once for each event | |
1242 | and is given 2 or 3 command-line arguments. The first is the | |
1243 | name of the event (see below). The second is the name of the | |
1244 | md device which is affected, and the third is the name of a related | |
1245 | device if relevant, such as a component device that has failed. | |
1246 | ||
1247 | If | |
1248 | .B --scan | |
1249 | is given, then a program or an E-mail address must be specified on the | |
1250 | command line or in the config file. If neither are available, then | |
1251 | .B mdadm | |
1252 | will not monitor anything. | |
1253 | Without | |
1254 | .B --scan | |
1255 | .B mdadm | |
1256 | will continue monitoring as long as something was found to monitor. If | |
1257 | no program or email is given, then each event is reported to | |
1258 | .BR stdout . | |
1259 | ||
1260 | The different events are: | |
1261 | ||
1262 | .RS 4 | |
1263 | .TP | |
1264 | .B DeviceDisappeared | |
1265 | An md array which previously was configured appears to no longer be | |
1266 | configured. (syslog priority: Critical) | |
1267 | ||
1268 | If | |
1269 | .I mdadm | |
1270 | was told to monitor an array which is RAID0 or Linear, then it will | |
1271 | report | |
1272 | .B DeviceDisappeared | |
1273 | with the extra information | |
1274 | .BR Wrong-Level . | |
1275 | This is because RAID0 and Linear do not support the device-failed, | |
1276 | hot-spare and resync operations which are monitored. | |
1277 | ||
1278 | .TP | |
1279 | .B RebuildStarted | |
1280 | An md array started reconstruction. (syslog priority: Warning) | |
1281 | ||
1282 | .TP | |
1283 | .BI Rebuild NN | |
1284 | Where | |
1285 | .I NN | |
1286 | is 20, 40, 60, or 80, this indicates that rebuild has passed that many | |
1287 | percentage of the total. (syslog priority: Warning) | |
1288 | ||
1289 | .TP | |
1290 | .B RebuildFinished | |
1291 | An md array that was rebuilding, isn't any more, either because it | |
1292 | finished normally or was aborted. (syslog priority: Warning) | |
1293 | ||
1294 | .TP | |
1295 | .B Fail | |
1296 | An active component device of an array has been marked as | |
1297 | faulty. (syslog priority: Critical) | |
1298 | ||
1299 | .TP | |
1300 | .B FailSpare | |
1301 | A spare component device which was being rebuilt to replace a faulty | |
1302 | device has failed. (syslog priority: Critial) | |
1303 | ||
1304 | .TP | |
1305 | .B SpareActive | |
1306 | A spare component device which was being rebuilt to replace a faulty | |
1307 | device has been successfully rebuilt and has been made active. | |
1308 | (syslog priority: Info) | |
1309 | ||
1310 | .TP | |
1311 | .B NewArray | |
1312 | A new md array has been detected in the | |
1313 | .B /proc/mdstat | |
1314 | file. (syslog priority: Info) | |
1315 | ||
1316 | .TP | |
1317 | .B DegradedArray | |
1318 | A newly noticed array appears to be degraded. This message is not | |
1319 | generated when | |
1320 | .I mdadm | |
1321 | notices a drive failure which causes degradation, but only when | |
1322 | .I mdadm | |
1323 | notices that an array is degraded when it first sees the array. | |
1324 | (syslog priority: Critial) | |
1325 | ||
1326 | .TP | |
1327 | .B MoveSpare | |
1328 | A spare drive has been moved from one array in a | |
1329 | .B spare-group | |
1330 | to another to allow a failed drive to be replaced. | |
1331 | (syslog priority: Info) | |
1332 | ||
1333 | .TP | |
1334 | .B SparesMissing | |
1335 | If | |
1336 | .I mdadm | |
1337 | has been told, via the config file, that an array should have a certain | |
1338 | number of spare devices, and | |
1339 | .I mdadm | |
1340 | detects that it has fewer that this number when it first sees the | |
1341 | array, it will report a | |
1342 | .B SparesMissing | |
1343 | message. | |
1344 | (syslog priority: Warning) | |
1345 | ||
1346 | .TP | |
1347 | .B TestMessage | |
1348 | An array was found at startup, and the | |
1349 | .B --test | |
1350 | flag was given. | |
1351 | (syslog priority: Info) | |
1352 | .RE | |
1353 | ||
1354 | Only | |
1355 | .B Fail , | |
1356 | .B FailSpare , | |
1357 | .B DegradedArray , | |
1358 | .B SparesMissing , | |
1359 | and | |
1360 | .B TestMessage | |
1361 | cause Email to be sent. All events cause the program to be run. | |
1362 | The program is run with two or three arguments, they being the event | |
1363 | name, the array device and possibly a second device. | |
1364 | ||
1365 | Each event has an associated array device (e.g. | |
1366 | .BR /dev/md1 ) | |
1367 | and possibly a second device. For | |
1368 | .BR Fail , | |
1369 | .BR FailSpare , | |
1370 | and | |
1371 | .B SpareActive | |
1372 | the second device is the relevant component device. | |
1373 | For | |
1374 | .B MoveSpare | |
1375 | the second device is the array that the spare was moved from. | |
1376 | ||
1377 | For | |
1378 | .B mdadm | |
1379 | to move spares from one array to another, the different arrays need to | |
1380 | be labelled with the same | |
1381 | .B spare-group | |
1382 | in the configuration file. The | |
1383 | .B spare-group | |
1384 | name can be any string. It is only necessary that different spare | |
1385 | groups use different names. | |
1386 | ||
1387 | When | |
1388 | .B mdadm | |
1389 | detects that an array which is in a spare group has fewer active | |
1390 | devices than necessary for the complete array, and has no spare | |
1391 | devices, it will look for another array in the same spare group that | |
1392 | has a full complement of working drive and a spare. It will then | |
1393 | attempt to remove the spare from the second drive and add it to the | |
1394 | first. | |
1395 | If the removal succeeds but the adding fails, then it is added back to | |
1396 | the original array. | |
1397 | ||
1398 | .SH GROW MODE | |
1399 | The GROW mode is used for changing the size or shape of an active | |
1400 | array. | |
1401 | For this to work, the kernel must support the necessary change. | |
1402 | Various types of growth are being added during 2.6 development, | |
1403 | including restructuring a raid5 array to have more active devices. | |
1404 | ||
1405 | Currently the only support available is to | |
1406 | .IP \(bu 4 | |
1407 | change the "size" attribute | |
1408 | for RAID1, RAID5 and RAID6. | |
1409 | .IP \(bu 4 | |
1410 | increase the "raid-disks" attribute of RAID1 and RAID5. | |
1411 | .IP \(bu 4 | |
1412 | add a write-intent bitmap to any array which support these bitmaps, or | |
1413 | remove a write-intent bitmap from such an array. | |
1414 | .PP | |
1415 | ||
1416 | .SS SIZE CHANGES | |
1417 | Normally when an array is built the "size" it taken from the smallest | |
1418 | of the drives. If all the small drives in an arrays are, one at a | |
1419 | time, removed and replaced with larger drives, then you could have an | |
1420 | array of large drives with only a small amount used. In this | |
1421 | situation, changing the "size" with "GROW" mode will allow the extra | |
1422 | space to start being used. If the size is increased in this way, a | |
1423 | "resync" process will start to make sure the new parts of the array | |
1424 | are synchronised. | |
1425 | ||
1426 | Note that when an array changes size, any filesystem that may be | |
1427 | stored in the array will not automatically grow to use the space. The | |
1428 | filesystem will need to be explicitly told to use the extra space. | |
1429 | ||
1430 | .SS RAID-DEVICES CHANGES | |
1431 | ||
1432 | A RAID1 array can work with any number of devices from 1 upwards | |
1433 | (though 1 is not very useful). There may be times which you want to | |
1434 | increase or decrease the number of active devices. Note that this is | |
1435 | different to hot-add or hot-remove which changes the number of | |
1436 | inactive devices. | |
1437 | ||
1438 | When reducing the number of devices in a RAID1 array, the slots which | |
1439 | are to be removed from the array must already be vacant. That is, the | |
1440 | devices that which were in those slots must be failed and removed. | |
1441 | ||
1442 | When the number of devices is increased, any hot spares that are | |
1443 | present will be activated immediately. | |
1444 | ||
1445 | Increasing the number of active devices in a RAID5 is much more | |
1446 | effort. Every block in the array will need to be read and written | |
1447 | back to a new location. From 2.6.17, the Linux Kernel is able to do | |
1448 | this safely, including restart and interrupted "reshape". | |
1449 | ||
1450 | When relocating the first few stripes on a raid5, it is not possible | |
1451 | to keep the data on disk completely consistent and crash-proof. To | |
1452 | provide the required safety, mdadm disables writes to the array while | |
1453 | this "critical section" is reshaped, and takes a backup of the data | |
1454 | that is in that section. This backup is normally stored in any spare | |
1455 | devices that the array has, however it can also be stored in a | |
1456 | separate file specified with the | |
1457 | .B --backup-file | |
1458 | option. If this option is used, and the system does crash during the | |
1459 | critical period, the same file must be passed to | |
1460 | .B --assemble | |
1461 | to restore the backup and reassemble the array. | |
1462 | ||
1463 | .SS BITMAP CHANGES | |
1464 | ||
1465 | A write-intent bitmap can be added to, or removed from, an active | |
1466 | array. Either internal bitmaps, or bitmaps stored in a separate file | |
1467 | can be added. Note that if you add a bitmap stored in a file which is | |
1468 | in a filesystem that is on the raid array being affected, the system | |
1469 | will deadlock. The bitmap must be on a separate filesystem. | |
1470 | ||
1471 | .SH EXAMPLES | |
1472 | ||
1473 | .B " mdadm --query /dev/name-of-device" | |
1474 | .br | |
1475 | This will find out if a given device is a raid array, or is part of | |
1476 | one, and will provide brief information about the device. | |
1477 | ||
1478 | .B " mdadm --assemble --scan" | |
1479 | .br | |
1480 | This will assemble and start all arrays listed in the standard config file | |
1481 | file. This command will typically go in a system startup file. | |
1482 | ||
1483 | .B " mdadm --stop --scan" | |
1484 | .br | |
1485 | This will shut down all array that can be shut down (i.e. are not | |
1486 | currently in use). This will typically go in a system shutdown script. | |
1487 | ||
1488 | .B " mdadm --follow --scan --delay=120" | |
1489 | .br | |
1490 | If (and only if) there is an Email address or program given in the | |
1491 | standard config file, then | |
1492 | monitor the status of all arrays listed in that file by | |
1493 | polling them ever 2 minutes. | |
1494 | ||
1495 | .B " mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/hd[ac]1" | |
1496 | .br | |
1497 | Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1. | |
1498 | ||
1499 | .br | |
1500 | .B " echo 'DEVICE /dev/hd*[0-9] /dev/sd*[0-9]' > mdadm.conf" | |
1501 | .br | |
1502 | .B " mdadm --detail --scan >> mdadm.conf" | |
1503 | .br | |
1504 | This will create a prototype config file that describes currently | |
1505 | active arrays that are known to be made from partitions of IDE or SCSI drives. | |
1506 | This file should be reviewed before being used as it may | |
1507 | contain unwanted detail. | |
1508 | ||
1509 | .B " echo 'DEVICE /dev/hd[a-z] /dev/sd*[a-z]' > mdadm.conf" | |
1510 | .br | |
1511 | .B " mdadm --examine --scan --config=mdadm.conf >> mdadm.conf" | |
1512 | .ber | |
1513 | This will find what arrays could be assembled from existing IDE and | |
1514 | SCSI whole drives (not partitions) and store the information is the | |
1515 | format of a config file. | |
1516 | This file is very likely to contain unwanted detail, particularly | |
1517 | the | |
1518 | .B devices= | |
1519 | entries. It should be reviewed and edited before being used as an | |
1520 | actual config file. | |
1521 | ||
1522 | .B " mdadm --examine --brief --scan --config=partitions" | |
1523 | .br | |
1524 | .B " mdadm -Ebsc partitions" | |
1525 | .br | |
1526 | Create a list of devices by reading | |
1527 | .BR /proc/partitions , | |
1528 | scan these for RAID superblocks, and printout a brief listing of all | |
1529 | that was found. | |
1530 | ||
1531 | .B " mdadm -Ac partitions -m 0 /dev/md0" | |
1532 | .br | |
1533 | Scan all partitions and devices listed in | |
1534 | .BR /proc/partitions | |
1535 | and assemble | |
1536 | .B /dev/md0 | |
1537 | out of all such devices with a RAID superblock with a minor number of 0. | |
1538 | ||
1539 | .B " mdadm --monitor --scan --daemonise > /var/run/mdadm" | |
1540 | .br | |
1541 | If config file contains a mail address or alert program, run mdadm in | |
1542 | the background in monitor mode monitoring all md devices. Also write | |
1543 | pid of mdadm daemon to | |
1544 | .BR /var/run/mdadm . | |
1545 | ||
1546 | .B " mdadm --create --help" | |
1547 | .br | |
1548 | Provide help about the Create mode. | |
1549 | ||
1550 | .B " mdadm --config --help" | |
1551 | .br | |
1552 | Provide help about the format of the config file. | |
1553 | ||
1554 | .B " mdadm --help" | |
1555 | .br | |
1556 | Provide general help. | |
1557 | ||
1558 | ||
1559 | .SH FILES | |
1560 | ||
1561 | .SS /proc/mdstat | |
1562 | ||
1563 | If you're using the | |
1564 | .B /proc | |
1565 | filesystem, | |
1566 | .B /proc/mdstat | |
1567 | lists all active md devices with information about them. | |
1568 | .B mdadm | |
1569 | uses this to find arrays when | |
1570 | .B --scan | |
1571 | is given in Misc mode, and to monitor array reconstruction | |
1572 | on Monitor mode. | |
1573 | ||
1574 | ||
1575 | .SS /etc/mdadm.conf | |
1576 | ||
1577 | The config file lists which devices may be scanned to see if | |
1578 | they contain MD super block, and gives identifying information | |
1579 | (e.g. UUID) about known MD arrays. See | |
1580 | .BR mdadm.conf (5) | |
1581 | for more details. | |
1582 | ||
1583 | .SH DEVICE NAMES | |
1584 | ||
1585 | While entries in the /dev directory can have any format you like, | |
1586 | .I mdadm | |
1587 | has an understanding of 'standard' formats which it uses to guide its | |
1588 | behaviour when creating device files via the | |
1589 | .I --auto | |
1590 | option. | |
1591 | ||
1592 | The standard names for non-partitioned arrays (the only sort of md | |
1593 | array available in 2.4 and earlier) either of | |
1594 | .IP | |
1595 | /dev/mdNN | |
1596 | .br | |
1597 | /dev/md/NN | |
1598 | .PP | |
1599 | where NN is a number. | |
1600 | The standard names for partitionable arrays (as available from 2.6 | |
1601 | onwards) is one of | |
1602 | .IP | |
1603 | /dev/md/dNN | |
1604 | .br | |
1605 | /dev/md_dNN | |
1606 | .PP | |
1607 | Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2". | |
1608 | ||
1609 | .SH NOTE | |
1610 | .B mdadm | |
1611 | was previously known as | |
1612 | .BR mdctl . | |
1613 | .P | |
1614 | .B mdadm | |
1615 | is completely separate from the | |
1616 | .B raidtools | |
1617 | package, and does not use the | |
1618 | .I /etc/raidtab | |
1619 | configuration file at all. | |
1620 | ||
1621 | .SH SEE ALSO | |
1622 | For information on the various levels of | |
1623 | RAID, check out: | |
1624 | ||
1625 | .IP | |
1626 | .UR http://ostenfeld.dk/~jakob/Software-RAID.HOWTO/ | |
1627 | http://ostenfeld.dk/~jakob/Software-RAID.HOWTO/ | |
1628 | .UE | |
1629 | '''.PP | |
1630 | '''for new releases of the RAID driver check out: | |
1631 | ''' | |
1632 | '''.IP | |
1633 | '''.UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches | |
1634 | '''ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches | |
1635 | '''.UE | |
1636 | '''.PP | |
1637 | '''or | |
1638 | '''.IP | |
1639 | '''.UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/ | |
1640 | '''http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/ | |
1641 | '''.UE | |
1642 | .PP | |
1643 | The latest version of | |
1644 | .I mdadm | |
1645 | should always be available from | |
1646 | .IP | |
1647 | .UR http://www.kernel.org/pub/linux/utils/raid/mdadm/ | |
1648 | http://www.kernel.org/pub/linux/utils/raid/mdadm/ | |
1649 | .UE | |
1650 | .PP | |
1651 | .IR mdadm.conf (5), | |
1652 | .IR md (4). | |
1653 | .PP | |
1654 | .IR raidtab (5), | |
1655 | .IR raid0run (8), | |
1656 | .IR raidstop (8), | |
1657 | .IR mkraid (8). |