-
Notifications
You must be signed in to change notification settings - Fork 0
/
Create.c
1078 lines (1003 loc) · 28.9 KB
/
Create.c
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
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
* mdadm - manage Linux "md" devices aka RAID arrays.
*
* Copyright (C) 2001-2013 Neil Brown <[email protected]>
*
*
* 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: <[email protected]>
*/
#include "mdadm.h"
#include "md_u.h"
#include "md_p.h"
#include <ctype.h>
static int default_layout(struct supertype *st, int level, int verbose)
{
int layout = UnSet;
if (st && st->ss->default_geometry)
st->ss->default_geometry(st, &level, &layout, NULL);
if (layout == UnSet)
switch(level) {
default: /* no layout */
layout = 0;
break;
case 10:
layout = 0x102; /* near=2, far=1 */
if (verbose > 0)
pr_err("layout defaults to n2\n");
break;
case 5:
case 6:
layout = map_name(r5layout, "default");
if (verbose > 0)
pr_err("layout defaults to %s\n", map_num(r5layout, layout));
break;
case LEVEL_FAULTY:
layout = map_name(faultylayout, "default");
if (verbose > 0)
pr_err("layout defaults to %s\n", map_num(faultylayout, layout));
break;
}
return layout;
}
int Create(struct supertype *st, char *mddev,
char *name, int *uuid,
int subdevs, struct mddev_dev *devlist,
struct shape *s,
struct context *c, unsigned long long data_offset)
{
/*
* 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
*/
int mdfd;
unsigned long long minsize=0, maxsize=0;
char *mindisc = NULL;
char *maxdisc = NULL;
int dnum, raid_disk_num;
struct mddev_dev *dv;
int fail=0, warn=0;
struct stat stb;
int first_missing = subdevs * 2;
int second_missing = subdevs * 2;
int missing_disks = 0;
int insert_point = subdevs * 2; /* where to insert a missing drive */
int total_slots;
int pass;
int vers;
int rv;
int bitmap_fd;
int have_container = 0;
int container_fd = -1;
int need_mdmon = 0;
unsigned long long bitmapsize;
struct mdinfo info, *infos;
int did_default = 0;
int do_default_layout = 0;
int do_default_chunk = 0;
unsigned long safe_mode_delay = 0;
char chosen_name[1024];
struct map_ent *map = NULL;
unsigned long long newsize;
int major_num = BITMAP_MAJOR_HI;
if (s->bitmap_file && strcmp(s->bitmap_file, "clustered") == 0) {
major_num = BITMAP_MAJOR_CLUSTERED;
if (c->nodes <= 1) {
pr_err("At least 2 nodes are needed for cluster-md\n");
return 1;
}
}
memset(&info, 0, sizeof(info));
if (s->level == UnSet && st && st->ss->default_geometry)
st->ss->default_geometry(st, &s->level, NULL, NULL);
if (s->level == UnSet) {
pr_err("a RAID level is needed to create an array.\n");
return 1;
}
if (s->raiddisks < 4 && s->level == 6) {
pr_err("at least 4 raid-devices needed for level 6\n");
return 1;
}
if (s->raiddisks > 256 && s->level == 6) {
pr_err("no more than 256 raid-devices supported for level 6\n");
return 1;
}
if (s->raiddisks < 2 && s->level >= 4) {
pr_err("at least 2 raid-devices needed for level 4 or 5\n");
return 1;
}
if (s->level <= 0 && s->sparedisks) {
pr_err("This level does not support spare devices\n");
return 1;
}
if (subdevs == 1 && strcmp(devlist->devname, "missing") != 0) {
/* If given a single device, it might be a container, and we can
* extract a device list from there
*/
mdu_array_info_t inf;
int fd;
memset(&inf, 0, sizeof(inf));
fd = open(devlist->devname, O_RDONLY);
if (fd >= 0 &&
ioctl(fd, GET_ARRAY_INFO, &inf) == 0 &&
inf.raid_disks == 0) {
/* yep, looks like a container */
if (st) {
rv = st->ss->load_container(st, fd,
devlist->devname);
if (rv == 0)
have_container = 1;
} else {
st = super_by_fd(fd, NULL);
if (st && !(rv = st->ss->
load_container(st, fd,
devlist->devname)))
have_container = 1;
else
st = NULL;
}
if (have_container) {
subdevs = s->raiddisks;
first_missing = subdevs * 2;
second_missing = subdevs * 2;
insert_point = subdevs * 2;
}
}
if (fd >= 0)
close(fd);
}
if (st && st->ss->external && s->sparedisks) {
pr_err("This metadata type does not support spare disks at create time\n");
return 1;
}
if (subdevs > s->raiddisks+s->sparedisks+s->journaldisks) {
pr_err("You have listed more devices (%d) than are in the array(%d)!\n", subdevs, s->raiddisks+s->sparedisks);
return 1;
}
if (!have_container && subdevs < s->raiddisks+s->sparedisks+s->journaldisks) {
pr_err("You haven't given enough devices (real or missing) to create this array\n");
return 1;
}
if (s->bitmap_file && s->level <= 0) {
pr_err("bitmaps not meaningful with level %s\n",
map_num(pers, s->level)?:"given");
return 1;
}
/* now set some defaults */
if (s->layout == UnSet) {
do_default_layout = 1;
s->layout = default_layout(st, s->level, c->verbose);
}
if (s->level == 10)
/* check layout fits in array*/
if ((s->layout&255) * ((s->layout>>8)&255) > s->raiddisks) {
pr_err("that layout requires at least %d devices\n",
(s->layout&255) * ((s->layout>>8)&255));
return 1;
}
switch(s->level) {
case 4:
case 5:
case 10:
case 6:
case 0:
if (s->chunk == 0 || s->chunk == UnSet) {
s->chunk = UnSet;
do_default_chunk = 1;
/* chunk will be set later */
}
break;
case LEVEL_LINEAR:
/* a chunksize of zero 0s perfectly valid (and preferred) since 2.6.16 */
if (get_linux_version() < 2006016 && s->chunk == 0) {
s->chunk = 64;
if (c->verbose > 0)
pr_err("chunk size defaults to 64K\n");
}
break;
case 1:
case LEVEL_FAULTY:
case LEVEL_MULTIPATH:
case LEVEL_CONTAINER:
if (s->chunk) {
s->chunk = 0;
if (c->verbose > 0)
pr_err("chunk size ignored for this level\n");
}
break;
default:
pr_err("unknown level %d\n", s->level);
return 1;
}
if (s->size == MAX_SIZE)
/* use '0' to mean 'max' now... */
s->size = 0;
if (s->size && s->chunk && s->chunk != UnSet)
s->size &= ~(unsigned long long)(s->chunk - 1);
newsize = s->size * 2;
if (st && ! st->ss->validate_geometry(st, s->level, s->layout, s->raiddisks,
&s->chunk, s->size*2,
data_offset, NULL,
&newsize, c->verbose>=0))
return 1;
if (s->chunk && s->chunk != UnSet) {
newsize &= ~(unsigned long long)(s->chunk*2 - 1);
if (do_default_chunk) {
/* default chunk was just set */
if (c->verbose > 0)
pr_err("chunk size defaults to %dK\n", s->chunk);
s->size &= ~(unsigned long long)(s->chunk - 1);
do_default_chunk = 0;
}
}
if (s->size == 0) {
s->size = newsize / 2;
if (s->level == 1)
/* If this is ever reshaped to RAID5, we will
* need a chunksize. So round it off a bit
* now just to be safe
*/
s->size &= ~(64ULL-1);
if (s->size && c->verbose > 0)
pr_err("setting size to %lluK\n", s->size);
}
/* now look at the subdevs */
info.array.active_disks = 0;
info.array.working_disks = 0;
dnum = 0;
for (dv = devlist; dv ; dv = dv->next)
if (data_offset == VARIABLE_OFFSET)
dv->data_offset = INVALID_SECTORS;
else
dv->data_offset = data_offset;
for (dv=devlist; dv && !have_container; dv=dv->next, dnum++) {
char *dname = dv->devname;
unsigned long long freesize;
int dfd;
char *doff;
if (strcasecmp(dname, "missing")==0) {
if (first_missing > dnum)
first_missing = dnum;
if (second_missing > dnum && dnum > first_missing)
second_missing = dnum;
missing_disks ++;
continue;
}
if (data_offset == VARIABLE_OFFSET) {
doff = strchr(dname, ':');
if (doff) {
*doff++ = 0;
dv->data_offset = parse_size(doff);
} else
dv->data_offset = INVALID_SECTORS;
} else
dv->data_offset = data_offset;
dfd = open(dname, O_RDONLY);
if (dfd < 0) {
pr_err("cannot open %s: %s\n",
dname, strerror(errno));
exit(2);
}
if (fstat(dfd, &stb) != 0 ||
(stb.st_mode & S_IFMT) != S_IFBLK) {
close(dfd);
pr_err("%s is not a block device\n",
dname);
exit(2);
}
close(dfd);
info.array.working_disks++;
if (dnum < s->raiddisks && dv->disposition != 'j')
info.array.active_disks++;
if (st == NULL) {
struct createinfo *ci = conf_get_create_info();
if (ci)
st = ci->supertype;
}
if (st == NULL) {
/* Need to choose a default metadata, which is different
* depending on geometry of array.
*/
int i;
char *name = "default";
for(i=0; !st && superlist[i]; i++) {
st = superlist[i]->match_metadata_desc(name);
if (!st)
continue;
if (do_default_layout)
s->layout = default_layout(st, s->level, c->verbose);
switch (st->ss->validate_geometry(
st, s->level, s->layout, s->raiddisks,
&s->chunk, s->size*2,
dv->data_offset, dname,
&freesize, c->verbose > 0)) {
case -1: /* Not valid, message printed, and not
* worth checking any further */
exit(2);
break;
case 0: /* Geometry not valid */
free(st);
st = NULL;
s->chunk = do_default_chunk ? UnSet : s->chunk;
break;
case 1: /* All happy */
break;
}
}
if (!st) {
int dfd = open(dname, O_RDONLY|O_EXCL);
if (dfd < 0) {
pr_err("cannot open %s: %s\n",
dname, strerror(errno));
exit(2);
}
pr_err("device %s not suitable for any style of array\n",
dname);
exit(2);
}
if (st->ss != &super0 ||
st->minor_version != 90)
did_default = 1;
} else {
if (do_default_layout)
s->layout = default_layout(st, s->level, 0);
if (!st->ss->validate_geometry(st, s->level, s->layout,
s->raiddisks,
&s->chunk, s->size*2,
dv->data_offset,
dname, &freesize,
c->verbose >= 0)) {
pr_err("%s is not suitable for this array.\n",
dname);
fail = 1;
continue;
}
}
if (dv->disposition == 'j')
goto skip_size_check; /* skip write journal for size check */
freesize /= 2; /* convert to K */
if (s->chunk && s->chunk != UnSet) {
/* round to chunk size */
freesize = freesize & ~(s->chunk-1);
if (do_default_chunk) {
/* default chunk was just set */
if (c->verbose > 0)
pr_err("chunk size defaults to %dK\n", s->chunk);
s->size &= ~(unsigned long long)(s->chunk - 1);
do_default_chunk = 0;
}
}
if (!freesize) {
pr_err("no free space left on %s\n", dname);
fail = 1;
continue;
}
if (s->size && freesize < s->size) {
pr_err("%s is smaller than given size. %lluK < %lluK + metadata\n",
dname, freesize, s->size);
fail = 1;
continue;
}
if (maxdisc == NULL || (maxdisc && freesize > maxsize)) {
maxdisc = dname;
maxsize = freesize;
}
if (mindisc ==NULL || (mindisc && freesize < minsize)) {
mindisc = dname;
minsize = freesize;
}
skip_size_check:
if (c->runstop != 1 || c->verbose >= 0) {
int fd = open(dname, O_RDONLY);
if (fd <0 ) {
pr_err("Cannot open %s: %s\n",
dname, strerror(errno));
fail=1;
continue;
}
warn |= check_ext2(fd, dname);
warn |= check_reiser(fd, dname);
warn |= check_raid(fd, dname);
if (strcmp(st->ss->name, "1.x") == 0 &&
st->minor_version >= 1)
/* metadata at front */
warn |= check_partitions(fd, dname, 0, 0);
else if (s->level == 1 || s->level == LEVEL_CONTAINER
|| (s->level == 0 && s->raiddisks == 1))
/* partitions could be meaningful */
warn |= check_partitions(fd, dname, freesize*2, s->size*2);
else
/* partitions cannot be meaningful */
warn |= check_partitions(fd, dname, 0, 0);
if (strcmp(st->ss->name, "1.x") == 0 &&
st->minor_version >= 1 &&
did_default &&
s->level == 1 &&
(warn & 1024) == 0) {
warn |= 1024;
pr_err("Note: this array has metadata at the start and\n"
" may not be suitable as a boot device. If you plan to\n"
" store '/boot' on this device please ensure that\n"
" your boot-loader understands md/v1.x metadata, or use\n"
" --metadata=0.90\n");
}
close(fd);
}
}
if (s->raiddisks + s->sparedisks > st->max_devs) {
pr_err("Too many devices: %s metadata only supports %d\n",
st->ss->name, st->max_devs);
return 1;
}
if (have_container)
info.array.working_disks = s->raiddisks;
if (fail) {
pr_err("create aborted\n");
return 1;
}
if (s->size == 0) {
if (mindisc == NULL && !have_container) {
pr_err("no size and no drives given - aborting create.\n");
return 1;
}
if (s->level > 0 || s->level == LEVEL_MULTIPATH
|| s->level == LEVEL_FAULTY
|| st->ss->external ) {
/* size is meaningful */
if (!st->ss->validate_geometry(st, s->level, s->layout,
s->raiddisks,
&s->chunk, minsize*2,
data_offset,
NULL, NULL, 0)) {
pr_err("devices too large for RAID level %d\n", s->level);
return 1;
}
s->size = minsize;
if (s->level == 1)
/* If this is ever reshaped to RAID5, we will
* need a chunksize. So round it off a bit
* now just to be safe
*/
s->size &= ~(64ULL-1);
if (c->verbose > 0)
pr_err("size set to %lluK\n", s->size);
}
}
if (!s->bitmap_file &&
s->level >= 1 &&
st->ss->add_internal_bitmap &&
(s->write_behind || s->size > 100*1024*1024ULL)) {
if (c->verbose > 0)
pr_err("automatically enabling write-intent bitmap on large array\n");
s->bitmap_file = "internal";
}
if (s->bitmap_file && strcmp(s->bitmap_file, "none") == 0)
s->bitmap_file = NULL;
if (!have_container && s->level > 0 && ((maxsize-s->size)*100 > maxsize)) {
if (c->runstop != 1 || c->verbose >= 0)
pr_err("largest drive (%s) exceeds size (%lluK) by more than 1%%\n",
maxdisc, s->size);
warn = 1;
}
if (st->ss->detail_platform && st->ss->detail_platform(0, 1, NULL) != 0) {
if (c->runstop != 1 || c->verbose >= 0)
pr_err("%s unable to enumerate platform support\n"
" array may not be compatible with hardware/firmware\n",
st->ss->name);
warn = 1;
}
st->nodes = c->nodes;
st->cluster_name = c->homecluster;
if (warn) {
if (c->runstop!= 1) {
if (!ask("Continue creating array? ")) {
pr_err("create aborted.\n");
return 1;
}
} else {
if (c->verbose > 0)
pr_err("creation continuing despite oddities due to --run\n");
}
}
/* If this is raid4/5, 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 (st->ss->external == 0 &&
s->assume_clean==0 && c->force == 0 && first_missing >= s->raiddisks) {
switch ( s->level ) {
case 4:
case 5:
insert_point = s->raiddisks-1;
s->sparedisks++;
info.array.active_disks--;
missing_disks++;
break;
default:
break;
}
}
/* For raid6, if creating with 1 missing drive, make a good drive
* into a spare, else the create will fail
*/
if (s->assume_clean == 0 && c->force == 0 && first_missing < s->raiddisks &&
st->ss->external == 0 &&
second_missing >= s->raiddisks && s->level == 6) {
insert_point = s->raiddisks - 1;
if (insert_point == first_missing)
insert_point--;
s->sparedisks ++;
info.array.active_disks--;
missing_disks++;
}
if (s->level <= 0 && first_missing < subdevs * 2) {
pr_err("This level does not support missing devices\n");
return 1;
}
/* We need to create the device */
map_lock(&map);
mdfd = create_mddev(mddev, name, c->autof, LOCAL, chosen_name);
if (mdfd < 0) {
map_unlock(&map);
return 1;
}
/* verify if chosen_name is not in use,
* it could be in conflict with already existing device
* e.g. container, array
*/
if (strncmp(chosen_name, "/dev/md/", 8) == 0
&& map_by_name(&map, chosen_name+8) != NULL) {
pr_err("Array name %s is in use already.\n",
chosen_name);
close(mdfd);
map_unlock(&map);
return 1;
}
mddev = chosen_name;
vers = md_get_version(mdfd);
if (vers < 9000) {
pr_err("Create requires md driver version 0.90.0 or later\n");
goto abort_locked;
} else {
mdu_array_info_t inf;
memset(&inf, 0, sizeof(inf));
ioctl(mdfd, GET_ARRAY_INFO, &inf);
if (inf.working_disks != 0) {
pr_err("another array by this name is already running.\n");
goto abort_locked;
}
}
/* Ok, lets try some ioctls */
info.array.level = s->level;
info.array.size = s->size;
info.array.raid_disks = s->raiddisks;
/* The kernel should *know* what md_minor we are dealing
* with, but it chooses to trust me instead. Sigh
*/
info.array.md_minor = 0;
if (fstat(mdfd, &stb)==0)
info.array.md_minor = minor(stb.st_rdev);
info.array.not_persistent = 0;
if ( ( (s->level == 4 || s->level == 5) &&
(insert_point < s->raiddisks || first_missing < s->raiddisks) )
||
( s->level == 6 && (insert_point < s->raiddisks
|| second_missing < s->raiddisks))
||
( s->level <= 0 )
||
s->assume_clean
) {
info.array.state = 1; /* clean, but one+ drive will be missing*/
info.resync_start = MaxSector;
} else {
info.array.state = 0; /* not clean, but no errors */
info.resync_start = 0;
}
if (s->level == 10) {
/* for raid10, the bitmap size is the capacity of the array,
* which is array.size * raid_disks / ncopies;
* .. but convert to sectors.
*/
int ncopies = ((s->layout>>8) & 255) * (s->layout & 255);
bitmapsize = s->size * s->raiddisks / ncopies * 2;
/* printf("bms=%llu as=%d rd=%d nc=%d\n", bitmapsize, s->size, s->raiddisks, ncopies);*/
} else
bitmapsize = s->size * 2;
/* 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.
*/
info.array.spare_disks=s->sparedisks;
info.array.failed_disks=missing_disks;
info.array.nr_disks = info.array.working_disks
+ info.array.failed_disks;
info.array.layout = s->layout;
info.array.chunk_size = s->chunk*1024;
if (name == NULL || *name == 0) {
/* base name on mddev */
/* /dev/md0 -> 0
* /dev/md_d0 -> d0
* /dev/md_foo -> foo
* /dev/md/1 -> 1
* /dev/md/d1 -> d1
* /dev/md/home -> home
* /dev/mdhome -> home
*/
/* FIXME compare this with rules in create_mddev */
name = strrchr(mddev, '/');
if (name) {
name++;
if (strncmp(name, "md_", 3)==0 &&
strlen(name) > 3 &&
(name-mddev) == 5 /* /dev/ */)
name += 3;
else if (strncmp(name, "md", 2)==0 &&
strlen(name) > 2 &&
isdigit(name[2]) &&
(name-mddev) == 5 /* /dev/ */)
name += 2;
}
}
if (!st->ss->init_super(st, &info.array, s->size, name, c->homehost, uuid,
data_offset))
goto abort_locked;
total_slots = info.array.nr_disks;
st->ss->getinfo_super(st, &info, NULL);
sysfs_init(&info, mdfd, NULL);
if (did_default && c->verbose >= 0) {
if (is_subarray(info.text_version)) {
char devnm[32];
char *ep;
struct mdinfo *mdi;
strncpy(devnm, info.text_version+1, 32);
devnm[31] = 0;
ep = strchr(devnm, '/');
if (ep)
*ep = 0;
mdi = sysfs_read(-1, devnm, GET_VERSION);
pr_err("Creating array inside %s container %s\n",
mdi?mdi->text_version:"managed", devnm);
sysfs_free(mdi);
} else
pr_err("Defaulting to version %s metadata\n", info.text_version);
}
map_update(&map, fd2devnm(mdfd), info.text_version,
info.uuid, chosen_name);
/* Keep map locked until devices have been added to array
* to stop another mdadm from finding and using those devices.
*/
if (s->bitmap_file && vers < 9003) {
major_num = BITMAP_MAJOR_HOSTENDIAN;
#ifdef __BIG_ENDIAN
pr_err("Warning - bitmaps created on this kernel are not portable\n"
" between different architectured. Consider upgrading the Linux kernel.\n");
#endif
}
if (s->bitmap_file && (strcmp(s->bitmap_file, "internal")==0 ||
strcmp(s->bitmap_file, "clustered")==0)) {
if ((vers%100) < 2) {
pr_err("internal bitmaps not supported by this kernel.\n");
goto abort_locked;
}
if (!st->ss->add_internal_bitmap) {
pr_err("internal bitmaps not supported with %s metadata\n",
st->ss->name);
goto abort_locked;
}
if (st->ss->add_internal_bitmap(st, &s->bitmap_chunk,
c->delay, s->write_behind,
bitmapsize, 1, major_num)) {
pr_err("Given bitmap chunk size not supported.\n");
goto abort_locked;
}
s->bitmap_file = NULL;
}
sysfs_init(&info, mdfd, NULL);
if (st->ss->external && st->container_devnm[0]) {
/* member */
/* When creating a member, we need to be careful
* to negotiate with mdmon properly.
* If it is already running, we cannot write to
* the devices and must ask it to do that part.
* If it isn't running, we write to the devices,
* and then start it.
* We hold an exclusive open on the container
* device to make sure mdmon doesn't exit after
* we checked that it is running.
*
* For now, fail if it is already running.
*/
container_fd = open_dev_excl(st->container_devnm);
if (container_fd < 0) {
pr_err("Cannot get exclusive open on container - weird.\n");
goto abort_locked;
}
if (mdmon_running(st->container_devnm)) {
if (c->verbose)
pr_err("reusing mdmon for %s.\n",
st->container_devnm);
st->update_tail = &st->updates;
} else
need_mdmon = 1;
}
rv = set_array_info(mdfd, st, &info);
if (rv) {
pr_err("failed to set array info for %s: %s\n",
mddev, strerror(errno));
goto abort_locked;
}
if (s->bitmap_file) {
int uuid[4];
st->ss->uuid_from_super(st, uuid);
if (CreateBitmap(s->bitmap_file, c->force, (char*)uuid, s->bitmap_chunk,
c->delay, s->write_behind,
bitmapsize,
major_num)) {
goto abort_locked;
}
bitmap_fd = open(s->bitmap_file, O_RDWR);
if (bitmap_fd < 0) {
pr_err("weird: %s cannot be openned\n",
s->bitmap_file);
goto abort_locked;
}
if (ioctl(mdfd, SET_BITMAP_FILE, bitmap_fd) < 0) {
pr_err("Cannot set bitmap file for %s: %s\n",
mddev, strerror(errno));
goto abort_locked;
}
}
infos = xmalloc(sizeof(*infos) * total_slots);
enable_fds(total_slots);
for (pass=1; pass <=2 ; pass++) {
struct mddev_dev *moved_disk = NULL; /* the disk that was moved out of the insert point */
for (dnum=0, raid_disk_num=0, dv = devlist ; dv ;
dv=(dv->next)?(dv->next):moved_disk, dnum++) {
int fd;
struct stat stb;
struct mdinfo *inf = &infos[dnum];
if (dnum >= total_slots)
abort();
if (dnum == insert_point) {
raid_disk_num += 1;
moved_disk = dv;
continue;
}
if (strcasecmp(dv->devname, "missing")==0) {
raid_disk_num += 1;
continue;
}
if (have_container)
moved_disk = NULL;
if (have_container && dnum < info.array.raid_disks - 1)
/* repeatedly use the container */
moved_disk = dv;
switch(pass) {
case 1:
*inf = info;
inf->disk.number = dnum;
inf->disk.raid_disk = raid_disk_num++;
if (dv->disposition == 'j') {
inf->disk.raid_disk = MD_DISK_ROLE_JOURNAL;
inf->disk.state = (1<<MD_DISK_JOURNAL);
raid_disk_num--;
} else if (inf->disk.raid_disk < s->raiddisks)
inf->disk.state = (1<<MD_DISK_ACTIVE) |
(1<<MD_DISK_SYNC);
else
inf->disk.state = 0;
if (dv->writemostly == FlagSet)
inf->disk.state |= (1<<MD_DISK_WRITEMOSTLY);
if (dv->failfast == FlagSet)
inf->disk.state |= (1<<MD_DISK_FAILFAST);
if (have_container)
fd = -1;
else {
if (st->ss->external &&
st->container_devnm[0])
fd = open(dv->devname, O_RDWR);
else
fd = open(dv->devname, O_RDWR|O_EXCL);
if (fd < 0) {
pr_err("failed to open %s after earlier success - aborting\n",
dv->devname);
goto abort_locked;
}
fstat(fd, &stb);
inf->disk.major = major(stb.st_rdev);
inf->disk.minor = minor(stb.st_rdev);
}
if (fd >= 0)
remove_partitions(fd);
if (st->ss->add_to_super(st, &inf->disk,
fd, dv->devname,
dv->data_offset)) {
ioctl(mdfd, STOP_ARRAY, NULL);
goto abort_locked;
}
st->ss->getinfo_super(st, inf, NULL);
safe_mode_delay = inf->safe_mode_delay;
if (have_container && c->verbose > 0)
pr_err("Using %s for device %d\n",
map_dev(inf->disk.major,
inf->disk.minor,
0), dnum);
if (!have_container) {
/* getinfo_super might have lost these ... */
inf->disk.major = major(stb.st_rdev);
inf->disk.minor = minor(stb.st_rdev);
}
break;
case 2:
inf->errors = 0;
rv = add_disk(mdfd, st, &info, inf);
if (rv) {
pr_err("ADD_NEW_DISK for %s failed: %s\n",
dv->devname, strerror(errno));
goto abort_locked;
}
break;
}
if (!have_container &&
dv == moved_disk && dnum != insert_point) break;
}
if (pass == 1) {
struct mdinfo info_new;
struct map_ent *me = NULL;
/* check to see if the uuid has changed due to these
* metadata changes, and if so update the member array
* and container uuid. Note ->write_init_super clears
* the subarray cursor such that ->getinfo_super once
* again returns container info.
*/
st->ss->getinfo_super(st, &info_new, NULL);
if (st->ss->external && s->level != LEVEL_CONTAINER &&
!same_uuid(info_new.uuid, info.uuid, 0)) {
map_update(&map, fd2devnm(mdfd),
info_new.text_version,
info_new.uuid, chosen_name);
me = map_by_devnm(&map, st->container_devnm);
}
if (st->ss->write_init_super(st)) {
st->ss->free_super(st);
goto abort_locked;
}
/* update parent container uuid */
if (me) {
char *path = xstrdup(me->path);
st->ss->getinfo_super(st, &info_new, NULL);
map_update(&map, st->container_devnm,
info_new.text_version,
info_new.uuid, path);
free(path);
}
flush_metadata_updates(st);
st->ss->free_super(st);
}
}
map_unlock(&map);
free(infos);
if (s->level == LEVEL_CONTAINER) {
/* No need to start. But we should signal udev to
* create links */
sysfs_uevent(&info, "change");
if (c->verbose >= 0)
pr_err("container %s prepared.\n", mddev);
wait_for(chosen_name, mdfd);