-
Notifications
You must be signed in to change notification settings - Fork 17
/
Copy pathAssemble.c
2208 lines (2052 loc) · 62.2 KB
/
Assemble.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) 2001-2016 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 "xmalloc.h"
#include <ctype.h>
mapping_t assemble_statuses[] = {
{ "but cannot be started", INCR_NO },
{ "but not safe to start", INCR_UNSAFE },
{ "and started", INCR_YES },
{ NULL, INCR_ALREADY }
};
/**
* struct assembly_array_info - General, meaningful information for assembly.
* @name: Array name.
* @new_cnt: Count of drives known to be members, recently added.
* @preexist_cnt: Count of member drives in pre-assembled array.
* @exp_cnt: Count of known expansion targets.
*
* FIXME: @exp_new_cnt for recently added expansion targets.
*/
struct assembly_array_info {
char *name;
int new_cnt;
int preexist_cnt;
int exp_cnt;
};
/**
* set_array_assembly_status() - generate status of assembly for an array.
* @c: Global settings.
* @result: Pointer to status mask.
* @status: Status to be set/printed.
* @arr: Array information.
*
* Print status message to user or set it in @result if it is not NULL.
*/
static void set_array_assembly_status(struct context *c,
int *result, int status,
struct assembly_array_info *arr)
{
int raid_disks = arr->preexist_cnt + arr->new_cnt;
char *status_msg = map_num_s(assemble_statuses, status);
if (c->export && result)
*result |= status;
if (c->export || c->verbose < 0)
return;
pr_err("%s has been assembled with %d device%s", arr->name,
raid_disks, raid_disks == 1 ? "":"s");
if (arr->preexist_cnt > 0)
fprintf(stderr, " (%d new)", arr->new_cnt);
if (arr->exp_cnt)
fprintf(stderr, " ( + %d for expansion)", arr->exp_cnt);
fprintf(stderr, " %s.\n", status_msg);
}
static int name_matches(char *found, char *required, char *homehost, int require_homehost)
{
/* See if the name found matches the required name, possibly
* prefixed with 'homehost'
*/
char *sep;
unsigned int l;
if (strcmp(found, required)==0)
return 1;
sep = strchr(found, ':');
if (!sep)
return 0;
l = sep - found;
if (strncmp(found, "any:", 4) == 0 ||
(homehost && strcmp(homehost, "any") == 0) ||
!require_homehost ||
(homehost && strlen(homehost) == l &&
strncmp(found, homehost, l) == 0)) {
/* matching homehost */
if (strcmp(sep+1, required) == 0)
return 1;
}
return 0;
}
static int is_member_busy(char *metadata_version)
{
/* check if the given member array is active */
struct mdstat_ent *mdstat = mdstat_read(0, 0);
struct mdstat_ent *ent;
int busy = 0;
for (ent = mdstat; ent; ent = ent->next) {
if (!is_mdstat_ent_subarray(ent))
continue;
/* Skip first char - it can be '/' or '-' */
if (strcmp(&ent->metadata_version[10], metadata_version + 1) == 0) {
busy = 1;
break;
}
}
free_mdstat(mdstat);
return busy;
}
static int ident_matches(struct mddev_ident *ident,
struct mdinfo *content,
struct supertype *tst,
char *homehost, int require_homehost,
enum update_opt update, char *devname)
{
if (ident->uuid_set && update != UOPT_UUID &&
same_uuid(content->uuid, ident->uuid, tst->ss->swapuuid)==0 &&
memcmp(content->uuid, uuid_zero, sizeof(int[4])) != 0) {
if (devname)
pr_err("%s has wrong uuid.\n", devname);
return 0;
}
if (ident->name[0] && update != UOPT_NAME &&
name_matches(content->name, ident->name, homehost, require_homehost)==0) {
if (devname)
pr_err("%s has wrong name.\n", devname);
return 0;
}
if (ident->super_minor != UnSet &&
ident->super_minor != content->array.md_minor) {
if (devname)
pr_err("%s has wrong super-minor.\n",
devname);
return 0;
}
if (ident->level != UnSet &&
ident->level != content->array.level) {
if (devname)
pr_err("%s has wrong raid level.\n",
devname);
return 0;
}
if (ident->raid_disks != UnSet &&
content->array.raid_disks != 0 && /* metadata doesn't know how many to expect */
ident->raid_disks!= content->array.raid_disks) {
if (devname)
pr_err("%s requires wrong number of drives.\n",
devname);
return 0;
}
if (ident->member && ident->member[0]) {
/* content->text_version must match */
char *s = strchr(content->text_version+1, '/');
if (s == NULL) {
if (devname)
pr_err("%s is not a container and one is required.\n",
devname);
return 0;
} else if (strcmp(ident->member, s+1) != 0) {
if (devname)
pr_err("skipping wrong member %s is %s\n",
content->text_version, devname);
return 0;
}
}
return 1;
}
static int select_devices(struct mddev_dev *devlist,
struct mddev_ident *ident,
struct supertype **stp,
struct mdinfo **contentp,
struct context *c,
int inargv, int auto_assem)
{
struct mddev_dev *tmpdev;
int num_devs;
struct supertype *st = *stp;
struct mdinfo *content = NULL;
int report_mismatch = ((inargv && c->verbose >= 0) || c->verbose > 0);
struct domainlist *domains = NULL;
dev_t rdev;
tmpdev = devlist; num_devs = 0;
while (tmpdev) {
if (tmpdev->used)
tmpdev->used = 2;
else
num_devs++;
tmpdev->disposition = 0;
tmpdev = tmpdev->next;
}
/* first walk the list of devices to find a consistent set
* that match the criterea, if that is possible.
* We flag the ones we like with 'used'.
*/
for (tmpdev = devlist;
tmpdev;
tmpdev = tmpdev ? tmpdev->next : NULL) {
char *devname = tmpdev->devname;
int dfd;
struct supertype *tst;
struct dev_policy *pol = NULL;
int found_container = 0;
if (tmpdev->used > 1)
continue;
if (ident->container) {
if (ident->container[0] == '/' &&
!same_dev(ident->container, devname)) {
if (report_mismatch)
pr_err("%s is not the container required (%s)\n",
devname, ident->container);
continue;
}
} else if (ident->devices &&
!match_oneof(ident->devices, devname)) {
/* Note that we ignore the "device=" identifier if a
* "container=" is given. Checking both is unnecessarily
* complicated.
*/
if (report_mismatch)
pr_err("%s is not one of %s\n", devname, ident->devices);
continue;
}
tst = dup_super(st);
dfd = dev_open(devname, O_RDONLY);
if (dfd < 0) {
if (report_mismatch)
pr_err("cannot open device %s: %s\n",
devname, strerror(errno));
tmpdev->used = 2;
} else if (!fstat_is_blkdev(dfd, devname, &rdev)) {
tmpdev->used = 2;
} else if (must_be_container(dfd)) {
if (st) {
/* already found some components, this cannot
* be another one.
*/
if (report_mismatch)
pr_err("%s is a container, but we are looking for components\n",
devname);
tmpdev->used = 2;
} if (!tst && (tst = super_by_fd(dfd, NULL)) == NULL) {
if (report_mismatch)
pr_err("not a recognisable container: %s\n",
devname);
tmpdev->used = 2;
} else if (!tst->ss->load_container ||
tst->ss->load_container(tst, dfd, NULL)) {
if (report_mismatch)
pr_err("no correct container type: %s\n",
devname);
tmpdev->used = 2;
} else if (auto_assem &&
!conf_test_metadata(tst->ss->name,
(pol = devid_policy(rdev)),
tst->ss->match_home(tst, c->homehost) == 1)) {
if (report_mismatch)
pr_err("%s has metadata type %s for which auto-assembly is disabled\n",
devname, tst->ss->name);
tmpdev->used = 2;
} else
found_container = 1;
} else {
if (!tst && (tst = guess_super(dfd)) == NULL) {
if (report_mismatch)
pr_err("no recogniseable superblock on %s\n",
devname);
tmpdev->used = 2;
} else if ((tst->ignore_hw_compat = 0),
tst->ss->load_super(tst, dfd,
report_mismatch ? devname : NULL)) {
if (report_mismatch)
pr_err("no RAID superblock on %s\n",
devname);
tmpdev->used = 2;
} else if (tst->ss->compare_super == NULL) {
if (report_mismatch)
pr_err("Cannot assemble %s metadata on %s\n",
tst->ss->name, devname);
tmpdev->used = 2;
} else if (auto_assem && st == NULL &&
!conf_test_metadata(tst->ss->name,
(pol = devid_policy(rdev)),
tst->ss->match_home(tst, c->homehost) == 1)) {
if (report_mismatch)
pr_err("%s has metadata type %s for which auto-assembly is disabled\n",
devname, tst->ss->name);
tmpdev->used = 2;
}
}
if (dfd >= 0) close(dfd);
if (tmpdev->used == 2) {
if (auto_assem || !inargv)
/* Ignore unrecognised devices during auto-assembly */
goto loop;
if (ident->name[0] ||
ident->super_minor != UnSet)
/* Ignore unrecognised device if looking for
* specific array */
goto loop;
if (ident->uuid_set)
/* ignore unrecognized device if looking for
* specific uuid
*/
goto loop;
pr_err("%s has no superblock - assembly aborted\n",
devname);
if (st)
st->ss->free_super(st);
dev_policy_free(pol);
domain_free(domains);
if (tst) {
tst->ss->free_super(tst);
free(tst);
}
return -1;
}
if (found_container) {
/* tmpdev is a container. We need to be either
* looking for a member, or auto-assembling
*/
/* should be safe to try an exclusive open now, we
* have rejected anything that some other mdadm might
* be looking at
*/
dfd = dev_open(devname, O_RDONLY | O_EXCL);
if (dfd < 0) {
if (report_mismatch)
pr_err("%s is busy - skipping\n", devname);
goto loop;
}
close(dfd);
if (ident->container && ident->container[0] != '/') {
/* we have a uuid */
int uuid[4];
content = *contentp;
tst->ss->getinfo_super(tst, content, NULL);
if (!parse_uuid(ident->container, uuid) ||
!same_uuid(content->uuid, uuid, tst->ss->swapuuid)) {
if (report_mismatch)
pr_err("%s has wrong UUID to be required container\n",
devname);
goto loop;
}
}
/* It is worth looking inside this container.
*/
if (c->verbose > 0)
pr_err("looking in container %s\n",
devname);
for (content = tst->ss->container_content(tst, NULL);
content;
content = content->next) {
if (!ident_matches(ident, content, tst,
c->homehost, c->require_homehost,
c->update,
report_mismatch ? devname : NULL))
/* message already printed */;
else if (is_member_busy(content->text_version)) {
if (report_mismatch)
pr_err("member %s in %s is already assembled\n",
content->text_version,
devname);
} else if (content->array.state & (1<<MD_SB_BLOCK_VOLUME)) {
/* do not assemble arrays with unsupported configurations */
pr_err("Cannot activate member %s in %s.\n",
content->text_version,
devname);
} else
break;
}
if (!content) {
tmpdev->used = 2;
goto loop; /* empty container */
}
st = tst; tst = NULL;
if (!auto_assem && inargv && tmpdev->next != NULL) {
pr_err("%s is a container, but is not only device given: confused and aborting\n",
devname);
st->ss->free_super(st);
dev_policy_free(pol);
domain_free(domains);
free(st);
return -1;
}
if (c->verbose > 0)
pr_err("found match on member %s in %s\n",
content->text_version, devname);
/* make sure we finished the loop */
tmpdev = NULL;
goto loop;
} else {
content = *contentp;
tst->ss->getinfo_super(tst, content, NULL);
if (!ident_matches(ident, content, tst,
c->homehost, c->require_homehost,
c->update,
report_mismatch ? devname : NULL))
goto loop;
if (auto_assem) {
/* Never auto-assemble things that conflict
* with mdadm.conf in some way
*/
struct mddev_ident *match;
int rv = 0;
match = conf_match(tst, content, devname,
report_mismatch ? c->verbose : -1,
&rv);
if (!match && rv == 2)
goto loop;
if (match && match->devname &&
strcasecmp(match->devname, "<ignore>") == 0) {
if (report_mismatch)
pr_err("%s is a member of an explicitly ignored array\n",
devname);
goto loop;
}
if (match && !ident_matches(match, content, tst,
c->homehost, c->require_homehost,
c->update,
report_mismatch ? devname : NULL))
/* Array exists in mdadm.conf but some
* details don't match, so reject it
*/
goto loop;
}
/* should be safe to try an exclusive open now, we
* have rejected anything that some other mdadm might
* be looking at
*/
dfd = dev_open(devname, O_RDONLY | O_EXCL);
if (dfd < 0) {
if (report_mismatch)
pr_err("%s is busy - skipping\n", devname);
goto loop;
}
close(dfd);
if (st == NULL)
st = dup_super(tst);
if (st->minor_version == -1)
st->minor_version = tst->minor_version;
if (memcmp(content->uuid, uuid_zero,
sizeof(int[4])) == 0) {
/* this is a floating spare. It cannot define
* an array unless there are no more arrays of
* this type to be found. It can be included
* in an array of this type though.
*/
tmpdev->used = 3;
goto loop;
}
if (st->ss != tst->ss ||
st->minor_version != tst->minor_version ||
st->ss->compare_super(st, tst, 1) != 0) {
/* Some mismatch. If exactly one array matches this host,
* we can resolve on that one.
* Or, if we are auto assembling, we just ignore the second
* for now.
*/
if (auto_assem)
goto loop;
if (c->homehost) {
int first = st->ss->match_home(st, c->homehost);
int last = tst->ss->match_home(tst, c->homehost);
if (first != last &&
(first == 1 || last == 1)) {
/* We can do something */
if (first) {/* just ignore this one */
if (report_mismatch)
pr_err("%s misses out due to wrong homehost\n",
devname);
goto loop;
} else { /* reject all those sofar */
struct mddev_dev *td;
if (report_mismatch)
pr_err("%s overrides previous devices due to good homehost\n",
devname);
for (td=devlist; td != tmpdev; td=td->next)
if (td->used == 1)
td->used = 0;
tmpdev->used = 1;
goto loop;
}
}
}
pr_err("superblock on %s doesn't match others - assembly aborted\n",
devname);
tst->ss->free_super(tst);
st->ss->free_super(st);
dev_policy_free(pol);
domain_free(domains);
free(tst);
return -1;
}
tmpdev->used = 1;
}
loop:
/* Collect domain information from members only */
if (tmpdev && tmpdev->used == 1) {
if (!pol)
pol = devid_policy(rdev);
domain_merge(&domains, pol, tst?tst->ss->name:NULL);
}
dev_policy_free(pol);
pol = NULL;
if (tst) {
tst->ss->free_super(tst);
free(tst);
}
}
/* Check if we found some imsm spares but no members */
if ((auto_assem ||
(ident->uuid_set &&
memcmp(uuid_zero, ident->uuid,sizeof(uuid_zero)) == 0)) &&
(!st || !st->sb))
for (tmpdev = devlist; tmpdev; tmpdev = tmpdev->next) {
if (tmpdev->used != 3)
continue;
tmpdev->used = 1;
content = *contentp;
if (!st)
return -1;
if (!st->sb) {
/* we need sb from one of the spares */
int dfd = dev_open(tmpdev->devname, O_RDONLY);
if (dfd < 0 ||
st->ss->load_super(st, dfd, NULL))
tmpdev->used = 2;
close_fd(&dfd);
}
}
/* Now reject spares that don't match domains of identified members */
for (tmpdev = devlist; tmpdev; tmpdev = tmpdev->next) {
if (tmpdev->used != 3)
continue;
if (!stat_is_blkdev(tmpdev->devname, &rdev)) {
tmpdev->used = 2;
} else {
struct dev_policy *pol = devid_policy(rdev);
int dt = domain_test(domains, pol, NULL);
if (inargv && dt != 0)
/* take this spare as domains match
* if there are any */
tmpdev->used = 1;
else if (!inargv && dt == 1)
/* device wasn't explicitly listed, so need
* explicit domain match - which we have */
tmpdev->used = 1;
else
/* if domains don't match mark as unused */
tmpdev->used = 0;
dev_policy_free(pol);
}
}
domain_free(domains);
*stp = st;
if (st && st->sb && content == *contentp)
st->ss->getinfo_super(st, content, NULL);
*contentp = content;
return num_devs;
}
struct devs {
char *devname;
int uptodate; /* set once we decide that this device is as
* recent as everything else in the array.
*/
int included; /* set if the device is already in the array
* due to a previous '-I'
*/
struct mdinfo i;
};
static int load_devices(struct devs *devices, char *devmap,
struct mddev_ident *ident, struct supertype **stp,
struct mddev_dev *devlist, struct context *c,
struct mdinfo *content,
int mdfd, char *mddev,
int *most_recentp, int *bestcntp, int **bestp,
int inargv)
{
struct mddev_dev *tmpdev;
int devcnt = 0;
int nextspare = 0;
int most_recent = -1;
int bestcnt = 0;
int *best = *bestp;
struct supertype *st = *stp;
for (tmpdev = devlist; tmpdev; tmpdev=tmpdev->next) {
char *devname = tmpdev->devname;
struct stat stb;
struct supertype *tst;
int i;
int dfd;
int disk_state;
if (tmpdev->used != 1)
continue;
/* looks like a good enough match to update the super block if needed */
if (c->update) {
/* prepare useful information in info structures */
struct stat stb2;
int err;
if (fstat(mdfd, &stb2) != 0)
goto error;
if (c->update == UOPT_UUID && !ident->uuid_set)
random_uuid((__u8 *)ident->uuid);
if (c->update == UOPT_PPL && ident->btype != BitmapNone) {
pr_err("PPL is not compatible with bitmap\n");
close(mdfd);
free(devices);
free(devmap);
return -1;
}
dfd = dev_open(devname,
tmpdev->disposition == 'I'
? O_RDWR : (O_RDWR|O_EXCL));
tst = dup_super(st);
if (dfd < 0 || tst->ss->load_super(tst, dfd, NULL) != 0) {
pr_err("cannot re-read metadata from %s - aborting\n",
devname);
if (dfd >= 0)
close(dfd);
tst->ss->free_super(tst);
free(tst);
*stp = st;
goto error;
}
tst->ss->getinfo_super(tst, content, devmap + devcnt * content->array.raid_disks);
memcpy(content->uuid, ident->uuid, 16);
strcpy(content->name, ident->name);
content->array.md_minor = minor(stb2.st_rdev);
if (c->update == UOPT_BYTEORDER)
err = 0;
else if (c->update == UOPT_HOME_CLUSTER) {
tst->cluster_name = c->homecluster;
err = tst->ss->write_bitmap(tst, dfd, NameUpdate);
} else if (c->update == UOPT_NODES) {
tst->nodes = c->nodes;
err = tst->ss->write_bitmap(tst, dfd, NodeNumUpdate);
} else if (c->update == UOPT_REVERT_RESHAPE && c->invalid_backup)
err = tst->ss->update_super(tst, content,
UOPT_SPEC_REVERT_RESHAPE_NOBACKUP,
devname, c->verbose,
ident->uuid_set,
c->homehost);
else
err = tst->ss->update_super(tst, content,
c->update,
devname, c->verbose,
ident->uuid_set,
c->homehost);
if (err < 0) {
if (err == -1)
pr_err("--update=%s not understood for %s metadata\n",
map_num(update_options, c->update), tst->ss->name);
tst->ss->free_super(tst);
free(tst);
close(dfd);
*stp = st;
goto error;
}
if (c->update == UOPT_UUID &&
!ident->uuid_set) {
ident->uuid_set = 1;
memcpy(ident->uuid, content->uuid, 16);
}
if (tst->ss->store_super(tst, dfd))
pr_err("Could not re-write superblock on %s.\n",
devname);
} else {
dfd = dev_open(devname,
tmpdev->disposition == 'I'
? O_RDWR : (O_RDWR|O_EXCL));
tst = dup_super(st);
if (dfd < 0 || tst->ss->load_super(tst, dfd, NULL) != 0) {
pr_err("cannot re-read metadata from %s - aborting\n",
devname);
if (dfd >= 0)
close(dfd);
tst->ss->free_super(tst);
free(tst);
*stp = st;
goto error;
}
tst->ss->getinfo_super(tst, content, devmap + devcnt * content->array.raid_disks);
}
if (fstat(dfd, &stb) != 0) {
close(dfd);
free(devices);
free(devmap);
tst->ss->free_super(tst);
free(tst);
*stp = st;
free(best);
return -1;
}
close(dfd);
if (c->verbose > 0)
pr_err("%s is identified as a member of %s, slot %d%s.\n",
devname, mddev, content->disk.raid_disk,
(content->disk.state & (1<<MD_DISK_REPLACEMENT)) ? " replacement":"");
devices[devcnt].devname = devname;
devices[devcnt].uptodate = 0;
devices[devcnt].included = (tmpdev->disposition == 'I');
devices[devcnt].i = *content;
devices[devcnt].i.disk.major = major(stb.st_rdev);
devices[devcnt].i.disk.minor = minor(stb.st_rdev);
disk_state = devices[devcnt].i.disk.state & ~((1<<MD_DISK_FAILFAST) |
(1<<MD_DISK_WRITEMOSTLY));
if (disk_state == ((1<<MD_DISK_ACTIVE) | (1<<MD_DISK_SYNC))) {
if (most_recent < 0 ||
devices[devcnt].i.events
> devices[most_recent].i.events) {
struct supertype *tmp = tst;
tst = st;
st = tmp;
most_recent = devcnt;
}
}
tst->ss->free_super(tst);
free(tst);
if (content->array.level == LEVEL_MULTIPATH)
/* with multipath, the raid_disk from the superblock is meaningless */
i = devcnt;
else
i = devices[devcnt].i.disk.raid_disk;
if (i+1 == 0 || i == MD_DISK_ROLE_JOURNAL) {
if (nextspare < content->array.raid_disks*2)
nextspare = content->array.raid_disks*2;
i = nextspare++;
} else {
/* i is raid_disk - double it so there is room for
* replacements */
i *= 2;
if (devices[devcnt].i.disk.state & (1<<MD_DISK_REPLACEMENT))
i++;
if (i >= content->array.raid_disks*2 &&
i >= nextspare)
nextspare = i+1;
}
if (i < 10000) {
if (i >= bestcnt) {
int newbestcnt = i+10;
int *newbest = xmalloc(sizeof(int)*newbestcnt);
int cc;
for (cc = 0; cc < newbestcnt; cc++)
if (cc < bestcnt)
newbest[cc] = best[cc];
else
newbest[cc] = -1;
if (best)free(best);
best = newbest;
bestcnt = newbestcnt;
}
if (best[i] >=0 &&
devices[best[i]].i.events ==
devices[devcnt].i.events &&
(devices[best[i]].i.disk.minor !=
devices[devcnt].i.disk.minor) &&
st->ss == &super0 &&
content->array.level != LEVEL_MULTIPATH) {
/* two different devices with identical superblock.
* Could be a mis-detection caused by overlapping
* partitions. fail-safe.
*/
pr_err("WARNING %s and %s appear to have very similar superblocks.\n"
" If they are really different, please --zero the superblock on one\n"
" If they are the same or overlap, please remove one from %s.\n",
devices[best[i]].devname, devname,
inargv ? "the list" :
"the\n DEVICE list in mdadm.conf"
);
*stp = st;
goto error;
}
if (best[i] == -1 || (devices[best[i]].i.events
< devices[devcnt].i.events))
best[i] = devcnt;
else if (st->ss == &super_imsm)
best[i+1] = devcnt;
}
devcnt++;
}
if (most_recent >= 0)
*most_recentp = most_recent;
*bestcntp = bestcnt;
*bestp = best;
*stp = st;
return devcnt;
error:
close(mdfd);
free(devices);
free(devmap);
free(best);
return -1;
}
static int force_array(struct mdinfo *content,
struct devs *devices,
int *best, int bestcnt, char *avail,
int most_recent,
struct supertype *st,
struct context *c)
{
int okcnt = 0;
while (!enough(content->array.level, content->array.raid_disks,
content->array.layout, 1,
avail) ||
(content->reshape_active && content->delta_disks > 0 &&
!enough(content->array.level, (content->array.raid_disks
- content->delta_disks),
content->new_layout, 1, avail))) {
/* Choose the newest best drive which is
* not up-to-date, update the superblock
* and add it.
*/
int fd;
struct supertype *tst;
unsigned long long current_events;
int chosen_drive = -1;
int i;
for (i = 0;
i < content->array.raid_disks * 2 && i < bestcnt;
i += 2) {
int j = best[i];
if (j < 0)
continue;
if (devices[j].uptodate)
continue;
if (devices[j].i.recovery_start != MaxSector) {
int delta;
if (!devices[j].i.reshape_active ||
devices[j].i.delta_disks <= 0)
continue;
/* When increasing number of devices, an
* added device also appears to be
* recovering. It is safe to include it
* as long as it won't be a source of
* data.
* For now, just allow for last data
* devices in RAID4 or last devices in RAID4/5/6.
*/
delta = devices[j].i.delta_disks;
if (is_level456(devices[j].i.array.level) &&
i/2 >= content->array.raid_disks - delta)
/* OK */;
else if (devices[j].i.array.level == 4 &&
i/2 >= content->array.raid_disks - delta - 1)
/* OK */;
else
continue;
} else if (devices[j].i.reshape_active !=
content->reshape_active ||
(devices[j].i.reshape_active &&
devices[j].i.reshape_progress !=
content->reshape_progress))
/* Here, it may be a source of data. If two
* devices claim different progresses, it
* means that reshape boundaries differ for
* their own devices. Kernel will only treat
* the first one as reshape progress and
* go on. It may cause disaster, so avoid it.
*/
continue;
if (chosen_drive < 0 ||
devices[j].i.events
> devices[chosen_drive].i.events)
chosen_drive = j;
}
if (chosen_drive < 0)
break;
current_events = devices[chosen_drive].i.events;
add_another:
if (c->verbose >= 0)
pr_err("forcing event count in %s(%d) from %d up to %d\n",
devices[chosen_drive].devname,
devices[chosen_drive].i.disk.raid_disk,
(int)(devices[chosen_drive].i.events),
(int)(devices[most_recent].i.events));
fd = dev_open(devices[chosen_drive].devname,
devices[chosen_drive].included ? O_RDWR
: (O_RDWR|O_EXCL));
if (fd < 0) {
pr_err("Couldn't open %s for write - not updating\n",
devices[chosen_drive].devname);
devices[chosen_drive].i.events = 0;
continue;
}
tst = dup_super(st);
if (tst->ss->load_super(tst,fd, NULL)) {
close(fd);
pr_err("RAID superblock disappeared from %s - not updating.\n",
devices[chosen_drive].devname);
devices[chosen_drive].i.events = 0;
continue;
}
content->events = devices[most_recent].i.events;
tst->ss->update_super(tst, content, UOPT_SPEC_FORCE_ONE,
devices[chosen_drive].devname, c->verbose,
0, NULL);
if (tst->ss->store_super(tst, fd)) {
close(fd);
pr_err("Could not re-write superblock on %s\n",
devices[chosen_drive].devname);
devices[chosen_drive].i.events = 0;
tst->ss->free_super(tst);
continue;
}
close(fd);
devices[chosen_drive].i.events = devices[most_recent].i.events;
devices[chosen_drive].uptodate = 1;
avail[chosen_drive] = 1;
okcnt++;
tst->ss->free_super(tst);
/* If there are any other drives of the same vintage,
* add them in as well. We can't lose and we might gain
*/
for (i = 0;
i < content->array.raid_disks * 2 && i < bestcnt ;
i += 2) {
int j = best[i];
if (j >= 0 &&
!devices[j].uptodate &&
devices[j].i.recovery_start == MaxSector &&
devices[j].i.events == current_events &&
((!devices[j].i.reshape_active &&
!content->reshape_active) ||
(devices[j].i.reshape_active ==
content->reshape_active &&
devices[j].i.reshape_progress ==
content->reshape_progress))) {