1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
|
/*
* mdadm - manage Linux "md" devices aka RAID arrays.
*
* Copyright (C) 2001-2002 Neil Brown <neilb@cse.unsw.edu.au>
*
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
* Author: Neil Brown
* Email: <neilb@cse.unsw.edu.au>
* Paper: Neil Brown
* School of Computer Science and Engineering
* The University of New South Wales
* Sydney, 2052
* Australia
*/
#include "mdadm.h"
#include "md_u.h"
#include "md_p.h"
int Create(struct supertype *st, char *mddev, int mdfd,
int chunk, int level, int layout, unsigned long size, int raiddisks, int sparedisks,
int subdevs, mddev_dev_t devlist,
int runstop, int verbose, int force,
char *bitmap_file, int bitmap_chunk, int delay)
{
/*
* Create a new raid array.
*
* First check that necessary details are available
* (i.e. level, raid-disks)
*
* Then check each disk to see what might be on it
* and report anything interesting.
*
* If anything looks odd, and runstop not set,
* abort.
*
* SET_ARRAY_INFO and ADD_NEW_DISK, and
* if runstop==run, or raiddisks disks were used,
* RUN_ARRAY
*/
unsigned long long minsize=0, maxsize=0;
char *mindisc = NULL;
char *maxdisc = NULL;
int dnum;
mddev_dev_t dv;
int fail=0, warn=0;
struct stat stb;
int first_missing = subdevs * 2;
int missing_disks = 0;
int insert_point = subdevs * 2; /* where to insert a missing drive */
void *super;
int pass;
int vers;
int rv;
int bitmap_fd;
mdu_array_info_t array;
vers = md_get_version(mdfd);
if (vers < 9000) {
fprintf(stderr, Name ": Create requires md driver version 0.90.0 or later\n");
return 1;
}
if (level == UnSet) {
fprintf(stderr,
Name ": a RAID level is needed to create an array.\n");
return 1;
}
if (raiddisks < 1) {
fprintf(stderr,
Name ": a number of --raid-devices must be given to create an array\n");
return 1;
}
if (raiddisks < 4 && level == 6) {
fprintf(stderr,
Name ": at least 4 raid-devices needed for level 6\n");
return 1;
}
if (raiddisks > 256 && level == 6) {
fprintf(stderr,
Name ": no more than 256 raid-devices supported for level 6\n");
return 1;
}
if (raiddisks < 2 && level >= 4) {
fprintf(stderr,
Name ": at least 2 raid-devices needed for level 4 or 5\n");
return 1;
}
if (subdevs > raiddisks+sparedisks) {
fprintf(stderr, Name ": You have listed more devices (%d) than are in the array(%d)!\n", subdevs, raiddisks+sparedisks);
return 1;
}
if (subdevs < raiddisks+sparedisks) {
fprintf(stderr, Name ": You haven't given enough devices (real or missing) to create this array\n");
return 1;
}
/* now set some defaults */
if (layout == UnSet)
switch(level) {
default: /* no layout */
layout = 0;
break;
case 10:
layout = 0x102; /* near=2, far=1 */
if (verbose)
fprintf(stderr,
Name ": layout defaults to n1\n");
break;
case 5:
case 6:
layout = map_name(r5layout, "default");
if (verbose)
fprintf(stderr,
Name ": layout defaults to %s\n", map_num(r5layout, layout));
break;
case LEVEL_FAULTY:
layout = map_name(faultylayout, "default");
if (verbose)
fprintf(stderr,
Name ": layout defaults to %s\n", map_num(faultylayout, layout));
break;
}
if (level == 10)
/* check layout fits in array*/
if ((layout&255) * ((layout>>8)&255) > raiddisks) {
fprintf(stderr, Name ": that layout requires at least %d devices\n",
(layout&255) * ((layout>>8)&255));
return 1;
}
switch(level) {
case 4:
case 5:
case 10:
case 6:
case 0:
case -1: /* linear */
if (chunk == 0) {
chunk = 64;
if (verbose)
fprintf(stderr, Name ": chunk size defaults to 64K\n");
}
break;
default: /* raid1, multipath */
if (chunk) {
chunk = 0;
if (verbose)
fprintf(stderr, Name ": chunk size ignored for this level\n");
}
break;
}
/* now look at the subdevs */
array.active_disks = 0;
array.working_disks = 0;
dnum = 0;
for (dv=devlist; dv; dv=dv->next, dnum++) {
char *dname = dv->devname;
unsigned long dsize;
unsigned long long ldsize, freesize;
int fd;
if (strcasecmp(dname, "missing")==0) {
if (first_missing > dnum)
first_missing = dnum;
missing_disks ++;
continue;
}
array.working_disks++;
if (dnum < raiddisks)
array.active_disks++;
fd = open(dname, O_RDONLY|O_EXCL, 0);
if (fd <0 ) {
fprintf(stderr, Name ": Cannot open %s: %s\n",
dname, strerror(errno));
fail=1;
continue;
}
#ifdef BLKGETSIZE64
if (ioctl(fd, BLKGETSIZE64, &ldsize)==0)
;
else
#endif
if (ioctl(fd, BLKGETSIZE, &dsize)) {
fprintf(stderr, Name ": Cannot get size of %s: %s\n",
dname, strerror(errno));
fail = 1;
close(fd);
continue;
}
else {
ldsize = dsize;
ldsize <<= 9;
}
freesize = st->ss->avail_size(ldsize >> 9);
if (freesize == 0) {
fprintf(stderr, Name ": %s is too small: %luK\n",
dname, (unsigned long)(ldsize>>10));
fail = 1;
close(fd);
continue;
}
freesize /= 2; /* convert to K */
if (size && freesize < size) {
fprintf(stderr, Name ": %s is smaller that given size."
" %lluK < %luK + superblock\n", dname, freesize, size);
fail = 1;
close(fd);
continue;
}
if (maxdisc == NULL || (maxdisc && freesize > maxsize)) {
maxdisc = dname;
maxsize = freesize;
}
if (mindisc ==NULL || (mindisc && freesize < minsize)) {
mindisc = dname;
minsize = freesize;
}
warn |= check_ext2(fd, dname);
warn |= check_reiser(fd, dname);
warn |= check_raid(fd, dname);
close(fd);
}
if (fail) {
fprintf(stderr, Name ": create aborted\n");
return 1;
}
if (size == 0) {
if (mindisc == NULL) {
fprintf(stderr, Name ": no size and no drives given - aborting create.\n");
return 1;
}
if (level > 0) {
/* size is meaningful */
if (minsize > 0x100000000ULL) {
fprintf(stderr, Name ": devices too large for RAID level %d\n", level);
return 1;
}
size = minsize;
if (verbose)
fprintf(stderr, Name ": size set to %luK\n", size);
}
}
if (level > 0 && ((maxsize-size)*100 > maxsize)) {
fprintf(stderr, Name ": largest drive (%s) exceed size (%luK) by more than 1%%\n",
maxdisc, size);
warn = 1;
}
if (warn) {
if (runstop!= 1) {
if (!ask("Continue creating array? ")) {
fprintf(stderr, Name ": create aborted.\n");
return 1;
}
} else {
if (verbose)
fprintf(stderr, Name ": creation continuing despite oddities due to --run\n");
}
}
/* If this is raid5, we want to configure the last active slot
* as missing, so that a reconstruct happens (faster than re-parity)
* FIX: Can we do this for raid6 as well?
*/
if (force == 0 && first_missing >= raiddisks) {
switch ( level ) {
case 5:
insert_point = raiddisks-1;
sparedisks++;
array.active_disks--;
missing_disks++;
break;
default:
break;
}
}
/* Ok, lets try some ioctls */
array.level = level;
array.size = size;
array.raid_disks = raiddisks;
/* The kernel should *know* what md_minor we are dealing
* with, but it chooses to trust me instead. Sigh
*/
array.md_minor = 0;
if (fstat(mdfd, &stb)==0)
array.md_minor = minor(stb.st_rdev);
array.not_persistent = 0;
/*** FIX: Need to do something about RAID-6 here ***/
if ( ( (level == 5) &&
(insert_point < raiddisks || first_missing < raiddisks) )
||
( level == 6 && missing_disks == 2)
)
array.state = 1; /* clean, but one+ drive will be missing */
else
array.state = 0; /* not clean, but no errors */
/* There is lots of redundancy in these disk counts,
* raid_disks is the most meaningful value
* it describes the geometry of the array
* it is constant
* nr_disks is total number of used slots.
* it should be raid_disks+spare_disks
* spare_disks is the number of extra disks present
* see above
* active_disks is the number of working disks in
* active slots. (With raid_disks)
* working_disks is the total number of working disks,
* including spares
* failed_disks is the number of disks marked failed
*
* Ideally, the kernel would keep these (except raid_disks)
* up-to-date as we ADD_NEW_DISK, but it doesn't (yet).
* So for now, we assume that all raid and spare
* devices will be given.
*/
array.spare_disks=sparedisks;
array.failed_disks=missing_disks;
array.nr_disks = array.working_disks + array.failed_disks;
array.layout = layout;
array.chunk_size = chunk*1024;
printf("VERS = %d\n", vers);
if (!st->ss->init_super(&super, &array))
return 1;
if ((vers % 100) >= 1) { /* can use different versions */
mdu_array_info_t inf;
memset(&inf, 0, sizeof(inf));
inf.major_version = st->ss->major;
inf.minor_version = st->minor_version;
rv = ioctl(mdfd, SET_ARRAY_INFO, &inf);
} else
rv = ioctl(mdfd, SET_ARRAY_INFO, NULL);
if (rv) {
fprintf(stderr, Name ": SET_ARRAY_INFO failed for %s: %s\n",
mddev, strerror(errno));
return 1;
}
if (bitmap_file) {
int uuid[4];
st->ss->uuid_from_super(uuid, super);
if (CreateBitmap(bitmap_file, force, (char*)uuid, bitmap_chunk, delay,
array.size*2ULL /* FIXME wrong for raid10 */)) {
return 1;
}
bitmap_fd = open(bitmap_file, O_RDWR);
if (bitmap_fd < 0) {
fprintf(stderr, Name ": weird: %s cannot be openned\n",
bitmap_file);
return 1;
}
if (ioctl(mdfd, SET_BITMAP_FILE, bitmap_fd) < 0) {
fprintf(stderr, Name ": Cannot set bitmap file for %s: %s\n",
mddev, strerror(errno));
return 1;
}
}
for (pass=1; pass <=2 ; pass++) {
mddev_dev_t moved_disk = NULL; /* the disk that was moved out of the insert point */
for (dnum=0, dv = devlist ; dv ;
dv=(dv->next)?(dv->next):moved_disk, dnum++) {
int fd;
struct stat stb;
mdu_disk_info_t disk;
disk.number = dnum;
if (dnum == insert_point) {
moved_disk = dv;
}
disk.raid_disk = disk.number;
if (disk.raid_disk < raiddisks)
disk.state = 6; /* active and in sync */
else
disk.state = 0;
if (dnum == insert_point ||
strcasecmp(dv->devname, "missing")==0) {
disk.major = 0;
disk.minor = 0;
disk.state = 1; /* faulty */
} else {
fd = open(dv->devname, O_RDONLY|O_EXCL, 0);
if (fd < 0) {
fprintf(stderr, Name ": failed to open %s after earlier success - aborting\n",
dv->devname);
return 1;
}
fstat(fd, &stb);
disk.major = major(stb.st_rdev);
disk.minor = minor(stb.st_rdev);
close(fd);
}
switch(pass){
case 1:
st->ss->add_to_super(super, &disk);
break;
case 2:
if (disk.state == 1) break;
st->ss->write_init_super(st, super, &disk, dv->devname);
if (ioctl(mdfd, ADD_NEW_DISK, &disk)) {
fprintf(stderr, Name ": ADD_NEW_DISK for %s failed: %s\n",
dv->devname, strerror(errno));
free(super);
return 1;
}
break;
}
if (dv == moved_disk && dnum != insert_point) break;
}
}
free(super);
/* param is not actually used */
if (runstop == 1 || subdevs >= raiddisks) {
mdu_param_t param;
if (ioctl(mdfd, RUN_ARRAY, ¶m)) {
fprintf(stderr, Name ": RUN_ARRAY failed: %s\n",
strerror(errno));
Manage_runstop(mddev, mdfd, -1);
return 1;
}
fprintf(stderr, Name ": array %s started.\n", mddev);
} else {
fprintf(stderr, Name ": not starting array - not enough devices.\n");
}
return 0;
}
|