-
Notifications
You must be signed in to change notification settings - Fork 0
/
Incremental.c
1824 lines (1679 loc) · 49.2 KB
/
Incremental.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
/*
* Incremental.c - support --incremental. Part of:
* mdadm - manage Linux "md" devices aka RAID arrays.
*
* Copyright (C) 2006-2013 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]>
* Paper: Neil Brown
* Novell Inc
* GPO Box Q1283
* QVB Post Office, NSW 1230
* Australia
*/
#include "mdadm.h"
#include <sys/wait.h>
#include <dirent.h>
#include <ctype.h>
static int count_active(struct supertype *st, struct mdinfo *sra,
int mdfd, char **availp,
struct mdinfo *info);
static void find_reject(int mdfd, struct supertype *st, struct mdinfo *sra,
int number, __u64 events, int verbose,
char *array_name);
static int try_spare(char *devname, int *dfdp, struct dev_policy *pol,
struct map_ent *target,
struct supertype *st, int verbose);
static int Incremental_container(struct supertype *st, char *devname,
struct context *c, char *only);
int Incremental(struct mddev_dev *devlist, struct context *c,
struct supertype *st)
{
/* Add this device to an array, creating the array if necessary
* and starting the array if sensible or - if runstop>0 - if possible.
*
* This has several steps:
*
* 1/ Check if device is permitted by mdadm.conf, reject if not.
* 2/ Find metadata, reject if none appropriate (check
* version/name from args)
* 3/ Check if there is a match in mdadm.conf
* 3a/ if not, check for homehost match. If no match, assemble as
* a 'foreign' array.
* 4/ Determine device number.
* - If in mdadm.conf with std name, use that
* - UUID in /var/run/mdadm.map use that
* - If name is suggestive, use that. unless in use with different uuid.
* - Choose a free, high number.
* - Use a partitioned device unless strong suggestion not to.
* e.g. auto=md
* Don't choose partitioned for containers.
* 5/ Find out if array already exists
* 5a/ if it does not
* - choose a name, from mdadm.conf or 'name' field in array.
* - create the array
* - add the device
* 5b/ if it does
* - check one drive in array to make sure metadata is a reasonably
* close match. Reject if not (e.g. different type)
* - add the device
* 6/ Make sure /var/run/mdadm.map contains this array.
* 7/ Is there enough devices to possibly start the array?
* For a container, this means running Incremental_container.
* 7a/ if not, finish with success.
* 7b/ if yes,
* - read all metadata and arrange devices like -A does
* - if number of OK devices match expected, or -R and there are enough,
* start the array (auto-readonly).
*/
dev_t rdev, rdev2;
struct mdinfo info, dinfo;
struct mdinfo *sra = NULL, *d;
struct mddev_ident *match;
char chosen_name[1024];
char *md_devname;
int rv = 1;
struct map_ent *mp, *map = NULL;
int dfd = -1, mdfd = -1;
char *avail = NULL;
int active_disks;
int trustworthy;
char *name_to_use;
struct dev_policy *policy = NULL;
struct map_ent target_array;
int have_target;
char *devname = devlist->devname;
int journal_device_missing = 0;
struct createinfo *ci = conf_get_create_info();
if (!stat_is_blkdev(devname, &rdev))
return rv;
dfd = dev_open(devname, O_RDONLY);
if (dfd < 0) {
if (c->verbose >= 0)
pr_err("cannot open %s: %s.\n",
devname, strerror(errno));
return rv;
}
/* If the device is a container, we do something very different */
if (must_be_container(dfd)) {
if (!st)
st = super_by_fd(dfd, NULL);
if (st && st->ss->load_container)
rv = st->ss->load_container(st, dfd, NULL);
close(dfd);
if (!rv && st->ss->container_content) {
if (map_lock(&map))
pr_err("failed to get exclusive lock on mapfile\n");
if (c->export)
printf("MD_DEVNAME=%s\n", devname);
rv = Incremental_container(st, devname, c, NULL);
map_unlock(&map);
return rv;
}
pr_err("%s is not part of an md array.\n",
devname);
return rv;
}
/* 1/ Check if device is permitted by mdadm.conf */
for (;devlist; devlist = devlist->next)
if (conf_test_dev(devlist->devname))
break;
if (!devlist) {
devlist = conf_get_devs();
for (;devlist; devlist = devlist->next) {
if (stat_is_blkdev(devlist->devname, &rdev2) &&
rdev2 == rdev)
break;
}
}
if (!devlist) {
if (c->verbose >= 0)
pr_err("%s not permitted by mdadm.conf.\n",
devname);
goto out;
}
/* 2/ Find metadata, reject if none appropriate (check
* version/name from args) */
if (!fstat_is_blkdev(dfd, devname, &rdev))
goto out;
dinfo.disk.major = major(rdev);
dinfo.disk.minor = minor(rdev);
policy = disk_policy(&dinfo);
have_target = policy_check_path(&dinfo, &target_array);
if (st == NULL && (st = guess_super_type(dfd, guess_array)) == NULL) {
if (c->verbose >= 0)
pr_err("no recognisable superblock on %s.\n",
devname);
rv = try_spare(devname, &dfd, policy,
have_target ? &target_array : NULL,
NULL, c->verbose);
goto out;
}
st->ignore_hw_compat = 0;
if (st->ss->compare_super == NULL ||
st->ss->load_super(st, dfd, c->verbose >= 0 ? devname : NULL)) {
if (c->verbose >= 0)
pr_err("no RAID superblock on %s.\n",
devname);
rv = try_spare(devname, &dfd, policy,
have_target ? &target_array : NULL,
st, c->verbose);
free(st);
goto out;
}
close (dfd); dfd = -1;
st->ss->getinfo_super(st, &info, NULL);
/* 3/ Check if there is a match in mdadm.conf */
match = conf_match(st, &info, devname, c->verbose, &rv);
if (!match && rv == 2)
goto out;
if (match && match->devname &&
strcasecmp(match->devname, "<ignore>") == 0) {
if (c->verbose >= 0)
pr_err("array containing %s is explicitly ignored by mdadm.conf\n",
devname);
goto out;
}
/* 3a/ if not, check for homehost match. If no match, continue
* but don't trust the 'name' in the array. Thus a 'random' minor
* number will be assigned, and the device name will be based
* on that. */
if (match)
trustworthy = LOCAL;
else if (st->ss->match_home(st, c->homehost) == 1)
trustworthy = LOCAL;
else if (st->ss->match_home(st, "any") == 1)
trustworthy = LOCAL_ANY;
else
trustworthy = FOREIGN;
if (!match && !conf_test_metadata(st->ss->name, policy,
(trustworthy == LOCAL))) {
if (c->verbose >= 1)
pr_err("%s has metadata type %s for which auto-assembly is disabled\n",
devname, st->ss->name);
goto out;
}
if (trustworthy == LOCAL_ANY)
trustworthy = LOCAL;
/* There are three possible sources for 'autof': command line,
* ARRAY line in mdadm.conf, or CREATE line in mdadm.conf.
* ARRAY takes precedence, then command line, then
* CREATE.
*/
if (match && match->autof)
c->autof = match->autof;
if (c->autof == 0)
c->autof = ci->autof;
name_to_use = info.name;
if (name_to_use[0] == 0 && info.array.level == LEVEL_CONTAINER) {
name_to_use = info.text_version;
trustworthy = METADATA;
}
if (name_to_use[0] && trustworthy != LOCAL &&
! c->require_homehost &&
conf_name_is_free(name_to_use))
trustworthy = LOCAL;
/* strip "hostname:" prefix from name if we have decided
* to treat it as LOCAL
*/
if (trustworthy == LOCAL && strchr(name_to_use, ':') != NULL)
name_to_use = strchr(name_to_use, ':')+1;
/* 4/ Check if array exists.
*/
if (map_lock(&map))
pr_err("failed to get exclusive lock on mapfile\n");
/* Now check we can get O_EXCL. If not, probably "mdadm -A" has
* taken over
*/
dfd = dev_open(devname, O_RDONLY|O_EXCL);
if (dfd < 0) {
if (c->verbose >= 0)
pr_err("cannot reopen %s: %s.\n",
devname, strerror(errno));
goto out_unlock;
}
/* Cannot hold it open while we add the device to the array,
* so we must release the O_EXCL and depend on the map_lock()
* So now is the best time to remove any partitions.
*/
remove_partitions(dfd);
close(dfd);
dfd = -1;
mp = map_by_uuid(&map, info.uuid);
if (mp)
mdfd = open_dev(mp->devnm);
else
mdfd = -1;
if (mdfd < 0) {
/* Skip the clustered ones. This should be started by
* clustering resource agents
*/
if (info.array.state & (1 << MD_SB_CLUSTERED))
goto out;
/* Couldn't find an existing array, maybe make a new one */
mdfd = create_mddev(match ? match->devname : NULL,
name_to_use, c->autof, trustworthy, chosen_name, 0);
if (mdfd < 0)
goto out_unlock;
if (sysfs_init(&info, mdfd, NULL)) {
pr_err("unable to initialize sysfs for %s\n",
chosen_name);
rv = 2;
goto out_unlock;
}
if (set_array_info(mdfd, st, &info) != 0) {
pr_err("failed to set array info for %s: %s\n",
chosen_name, strerror(errno));
rv = 2;
goto out_unlock;
}
dinfo = info;
dinfo.disk.major = major(rdev);
dinfo.disk.minor = minor(rdev);
if (add_disk(mdfd, st, &info, &dinfo) != 0) {
pr_err("failed to add %s to new array %s: %s.\n",
devname, chosen_name, strerror(errno));
ioctl(mdfd, STOP_ARRAY, 0);
rv = 2;
goto out_unlock;
}
sra = sysfs_read(mdfd, NULL, (GET_DEVS | GET_STATE |
GET_OFFSET | GET_SIZE));
if (!sra || !sra->devs || sra->devs->disk.raid_disk >= 0) {
/* It really should be 'none' - must be old buggy
* kernel, and mdadm -I may not be able to complete.
* So reject it.
*/
ioctl(mdfd, STOP_ARRAY, NULL);
pr_err("You have an old buggy kernel which cannot support\n --incremental reliably. Aborting.\n");
rv = 2;
goto out_unlock;
}
info.array.working_disks = 1;
/* 6/ Make sure /var/run/mdadm.map contains this array. */
map_update(&map, fd2devnm(mdfd),
info.text_version,
info.uuid, chosen_name);
} else {
/* 5b/ if it does */
/* - check one drive in array to make sure metadata is a reasonably */
/* close match. Reject if not (e.g. different type) */
/* - add the device */
char dn[20];
int dfd2;
int err;
struct supertype *st2;
struct mdinfo info2, *d;
sra = sysfs_read(mdfd, NULL, (GET_DEVS | GET_STATE |
GET_OFFSET | GET_SIZE));
if (mp->path)
strcpy(chosen_name, mp->path);
else
strcpy(chosen_name, mp->devnm);
/* It is generally not OK to add non-spare drives to a
* running array as they are probably missing because
* they failed. However if runstop is 1, then the
* array was possibly started early and our best bet is
* to add this anyway.
* Also if action policy is re-add or better we allow
* re-add.
* This doesn't apply to containers as the 'non-spare'
* flag has a different meaning. The test has to happen
* at the device level there
*/
if (!st->ss->external &&
(info.disk.state & (1 << MD_DISK_SYNC)) != 0 &&
!policy_action_allows(policy, st->ss->name, act_re_add) &&
c->runstop < 1) {
if (md_array_active(mdfd)) {
pr_err("not adding %s to active array (without --run) %s\n",
devname, chosen_name);
rv = 2;
goto out_unlock;
}
}
if (!sra) {
rv = 2;
goto out_unlock;
}
if (sra->devs) {
sprintf(dn, "%d:%d", sra->devs->disk.major,
sra->devs->disk.minor);
dfd2 = dev_open(dn, O_RDONLY);
if (dfd2 < 0) {
pr_err("unable to open %s\n", devname);
rv = 2;
goto out_unlock;
}
st2 = dup_super(st);
if (st2->ss->load_super(st2, dfd2, NULL) ||
st->ss->compare_super(st, st2) != 0) {
pr_err("metadata mismatch between %s and chosen array %s\n",
devname, chosen_name);
close(dfd2);
rv = 2;
goto out_unlock;
}
close(dfd2);
st2->ss->getinfo_super(st2, &info2, NULL);
st2->ss->free_super(st2);
if (info.array.level != info2.array.level ||
memcmp(info.uuid, info2.uuid, 16) != 0 ||
info.array.raid_disks != info2.array.raid_disks) {
pr_err("unexpected difference between %s and %s.\n",
chosen_name, devname);
rv = 2;
goto out_unlock;
}
}
info.disk.major = major(rdev);
info.disk.minor = minor(rdev);
/* add disk needs to know about containers */
if (st->ss->external)
sra->array.level = LEVEL_CONTAINER;
if (info.array.state & (1 << MD_SB_CLUSTERED))
info.disk.state |= (1 << MD_DISK_CLUSTER_ADD);
err = add_disk(mdfd, st, sra, &info);
if (err < 0 && errno == EBUSY) {
/* could be another device present with the same
* disk.number. Find and reject any such
*/
find_reject(mdfd, st, sra, info.disk.number,
info.events, c->verbose, chosen_name);
err = add_disk(mdfd, st, sra, &info);
}
if (err < 0 && errno == EINVAL &&
info.disk.state & (1<<MD_DISK_SYNC)) {
/* Maybe it needs to be added as a spare */
if (policy_action_allows(policy, st->ss->name,
act_force_spare)) {
info.disk.state &= ~(1<<MD_DISK_SYNC);
err = add_disk(mdfd, st, sra, &info);
} else
if (c->verbose >= 0)
pr_err("can only add %s to %s as a spare, and force-spare is not set.\n",
devname, chosen_name);
}
if (err < 0) {
pr_err("failed to add %s to existing array %s: %s.\n",
devname, chosen_name, strerror(errno));
rv = 2;
goto out_unlock;
}
info.array.working_disks = 0;
for (d = sra->devs; d; d=d->next)
info.array.working_disks ++;
}
if (strncmp(chosen_name, "/dev/md/", 8) == 0)
md_devname = chosen_name+8;
else
md_devname = chosen_name;
if (c->export) {
printf("MD_DEVICE=%s\n", fd2devnm(mdfd));
printf("MD_DEVNAME=%s\n", md_devname);
printf("MD_FOREIGN=%s\n", trustworthy == FOREIGN ? "yes" : "no");
}
/* 7/ Is there enough devices to possibly start the array? */
/* 7a/ if not, finish with success. */
if (info.array.level == LEVEL_CONTAINER) {
char devnm[32];
/* Try to assemble within the container */
sysfs_uevent(sra, "change");
if (!c->export && c->verbose >= 0)
pr_err("container %s now has %d device%s\n",
chosen_name, info.array.working_disks,
info.array.working_disks == 1?"":"s");
sysfs_rules_apply(chosen_name, &info);
wait_for(chosen_name, mdfd);
if (st->ss->external)
strcpy(devnm, fd2devnm(mdfd));
if (st->ss->load_container)
rv = st->ss->load_container(st, mdfd, NULL);
close(mdfd);
sysfs_free(sra);
if (!rv)
rv = Incremental_container(st, chosen_name, c, NULL);
map_unlock(&map);
/* after spare is added, ping monitor for external metadata
* so that it can eg. try to rebuild degraded array */
if (st->ss->external)
ping_monitor(devnm);
return rv;
}
/* We have added something to the array, so need to re-read the
* state. Eventually this state should be kept up-to-date as
* things change.
*/
sysfs_free(sra);
sra = sysfs_read(mdfd, NULL, (GET_DEVS | GET_STATE |
GET_OFFSET | GET_SIZE));
active_disks = count_active(st, sra, mdfd, &avail, &info);
journal_device_missing = (info.journal_device_required) && (info.journal_clean == 0);
if (info.consistency_policy == CONSISTENCY_POLICY_PPL)
info.array.state |= 1;
if (enough(info.array.level, info.array.raid_disks,
info.array.layout, info.array.state & 1, avail) == 0) {
if (c->export) {
printf("MD_STARTED=no\n");
} else if (c->verbose >= 0)
pr_err("%s attached to %s, not enough to start (%d).\n",
devname, chosen_name, active_disks);
rv = 0;
goto out_unlock;
}
/* 7b/ if yes, */
/* - if number of OK devices match expected, or -R and there */
/* are enough, */
/* + add any bitmap file */
/* + start the array (auto-readonly). */
if (md_array_active(mdfd)) {
if (c->export) {
printf("MD_STARTED=already\n");
} else if (c->verbose >= 0)
pr_err("%s attached to %s which is already active.\n",
devname, chosen_name);
rv = 0;
goto out_unlock;
}
map_unlock(&map);
if (c->runstop > 0 || (!journal_device_missing && active_disks >= info.array.working_disks)) {
struct mdinfo *dsk;
/* Let's try to start it */
if (journal_device_missing)
pr_err("Trying to run with missing journal device\n");
if (info.reshape_active && !(info.reshape_active & RESHAPE_NO_BACKUP)) {
pr_err("%s: This array is being reshaped and cannot be started\n",
chosen_name);
cont_err("by --incremental. Please use --assemble\n");
goto out;
}
if (match && match->bitmap_file) {
int bmfd = open(match->bitmap_file, O_RDWR);
if (bmfd < 0) {
pr_err("Could not open bitmap file %s.\n",
match->bitmap_file);
goto out;
}
if (ioctl(mdfd, SET_BITMAP_FILE, bmfd) != 0) {
close(bmfd);
pr_err("Failed to set bitmapfile for %s.\n",
chosen_name);
goto out;
}
close(bmfd);
}
/* Need to remove from the array any devices which
* 'count_active' discerned were too old or inappropriate
*/
for (d = sra ? sra->devs : NULL ; d ; d = d->next)
if (d->disk.state & (1<<MD_DISK_REMOVED))
remove_disk(mdfd, st, sra, d);
if ((sra == NULL || active_disks >= info.array.working_disks) &&
trustworthy != FOREIGN)
rv = ioctl(mdfd, RUN_ARRAY, NULL);
else
rv = sysfs_set_str(sra, NULL,
"array_state", "read-auto");
/* Array might be O_EXCL which will interfere with
* fsck and mount. So re-open without O_EXCL.
*/
reopen_mddev(mdfd);
if (rv == 0) {
if (c->export) {
printf("MD_STARTED=yes\n");
} else if (c->verbose >= 0)
pr_err("%s attached to %s, which has been started.\n",
devname, chosen_name);
rv = 0;
wait_for(chosen_name, mdfd);
/* We just started the array, so some devices
* might have been evicted from the array
* because their event counts were too old.
* If the action=re-add policy is in-force for
* those devices we should re-add them now.
*/
for (dsk = sra->devs; dsk ; dsk = dsk->next) {
if (disk_action_allows(dsk, st->ss->name,
act_re_add) &&
add_disk(mdfd, st, sra, dsk) == 0)
pr_err("%s re-added to %s\n",
dsk->sys_name, chosen_name);
}
} else {
pr_err("%s attached to %s, but failed to start: %s.\n",
devname, chosen_name, strerror(errno));
rv = 1;
}
} else {
if (c->export) {
printf("MD_STARTED=unsafe\n");
} else if (journal_device_missing) {
pr_err("Journal device is missing, not safe to start yet.\n");
} else if (c->verbose >= 0)
pr_err("%s attached to %s, not enough to start safely.\n",
devname, chosen_name);
rv = 0;
}
out:
free(avail);
if (dfd >= 0)
close(dfd);
if (mdfd >= 0)
close(mdfd);
if (policy)
dev_policy_free(policy);
sysfs_free(sra);
return rv;
out_unlock:
map_unlock(&map);
goto out;
}
static void find_reject(int mdfd, struct supertype *st, struct mdinfo *sra,
int number, __u64 events, int verbose,
char *array_name)
{
/* Find a device attached to this array with a disk.number of number
* and events less than the passed events, and remove the device.
*/
struct mdinfo *d;
if (md_array_active(mdfd))
return; /* not safe to remove from active arrays
* without thinking more */
for (d = sra->devs; d ; d = d->next) {
char dn[24]; // 2*11 bytes for ints (including sign) + colon + null byte
int dfd;
struct mdinfo info;
sprintf(dn, "%d:%d", d->disk.major, d->disk.minor);
dfd = dev_open(dn, O_RDONLY);
if (dfd < 0)
continue;
if (st->ss->load_super(st, dfd, NULL)) {
close(dfd);
continue;
}
st->ss->getinfo_super(st, &info, NULL);
st->ss->free_super(st);
close(dfd);
if (info.disk.number != number || info.events >= events)
continue;
if (d->disk.raid_disk > -1)
sysfs_set_str(sra, d, "slot", "none");
if (sysfs_set_str(sra, d, "state", "remove") == 0)
if (verbose >= 0)
pr_err("removing old device %s from %s\n",
d->sys_name+4, array_name);
}
}
static int count_active(struct supertype *st, struct mdinfo *sra,
int mdfd, char **availp,
struct mdinfo *bestinfo)
{
/* count how many devices in sra think they are active */
struct mdinfo *d;
int cnt = 0;
int replcnt = 0;
__u64 max_events = 0;
__u64 max_journal_events = 0;
char *avail = NULL;
int *best = NULL;
char *devmap = NULL;
int numdevs = 0;
int devnum;
int b, i;
int raid_disks = 0;
if (!sra)
return 0;
for (d = sra->devs ; d ; d = d->next)
numdevs++;
for (d = sra->devs, devnum = 0 ; d ; d = d->next, devnum++) {
char dn[30];
int dfd;
int ok;
struct mdinfo info;
sprintf(dn, "%d:%d", d->disk.major, d->disk.minor);
dfd = dev_open(dn, O_RDONLY);
if (dfd < 0)
continue;
ok = st->ss->load_super(st, dfd, NULL);
close(dfd);
if (ok != 0)
continue;
info.array.raid_disks = raid_disks;
st->ss->getinfo_super(st, &info, devmap + raid_disks * devnum);
if (info.disk.raid_disk == MD_DISK_ROLE_JOURNAL &&
info.events > max_journal_events)
max_journal_events = info.events;
if (!avail) {
raid_disks = info.array.raid_disks;
avail = xcalloc(raid_disks, 1);
*availp = avail;
best = xcalloc(raid_disks, sizeof(int));
devmap = xcalloc(raid_disks, numdevs);
st->ss->getinfo_super(st, &info, devmap);
}
if (info.disk.state & (1<<MD_DISK_SYNC))
{
if (cnt == 0) {
cnt++;
max_events = info.events;
avail[info.disk.raid_disk] = 2;
best[info.disk.raid_disk] = devnum;
st->ss->getinfo_super(st, bestinfo, NULL);
} else if (info.events == max_events) {
avail[info.disk.raid_disk] = 2;
best[info.disk.raid_disk] = devnum;
} else if (info.events == max_events-1) {
if (avail[info.disk.raid_disk] == 0) {
avail[info.disk.raid_disk] = 1;
best[info.disk.raid_disk] = devnum;
}
} else if (info.events < max_events - 1)
;
else if (info.events == max_events+1) {
int i;
max_events = info.events;
for (i = 0; i < raid_disks; i++)
if (avail[i])
avail[i]--;
avail[info.disk.raid_disk] = 2;
best[info.disk.raid_disk] = devnum;
st->ss->getinfo_super(st, bestinfo, NULL);
} else { /* info.events much bigger */
memset(avail, 0, raid_disks);
max_events = info.events;
avail[info.disk.raid_disk] = 2;
best[info.disk.raid_disk] = devnum;
st->ss->getinfo_super(st, bestinfo, NULL);
}
} else if (info.disk.state & (1<<MD_DISK_REPLACEMENT))
replcnt++;
st->ss->free_super(st);
}
if (max_journal_events >= max_events - 1)
bestinfo->journal_clean = 1;
if (!avail)
return 0;
/* We need to reject any device that thinks the best device is
* failed or missing */
for (b = 0; b < raid_disks; b++)
if (avail[b] == 2)
break;
cnt = 0;
for (i = 0 ; i < raid_disks ; i++) {
if (i != b && avail[i])
if (devmap[raid_disks * best[i] + b] == 0) {
/* This device thinks 'b' is failed -
* don't use it */
devnum = best[i];
for (d=sra->devs ; devnum; d = d->next)
devnum--;
d->disk.state |= (1 << MD_DISK_REMOVED);
avail[i] = 0;
}
if (avail[i])
cnt++;
}
/* Also need to reject any spare device with an event count that
* is too high
*/
for (d = sra->devs; d; d = d->next) {
if (!(d->disk.state & (1<<MD_DISK_SYNC)) &&
d->events > max_events)
d->disk.state |= (1 << MD_DISK_REMOVED);
}
free(best);
free(devmap);
return cnt + replcnt;
}
/* test if container has degraded member(s) */
static int
container_members_max_degradation(struct map_ent *map, struct map_ent *me)
{
struct mdinfo *sra;
int degraded, max_degraded = 0;
for(; map; map = map->next) {
if (!metadata_container_matches(map->metadata, me->devnm))
continue;
/* most accurate information regarding array degradation */
sra = sysfs_read(-1, map->devnm,
GET_DISKS | GET_DEVS | GET_STATE);
if (!sra)
continue;
degraded = sra->array.raid_disks - sra->array.active_disks -
sra->array.spare_disks;
if (degraded > max_degraded)
max_degraded = degraded;
sysfs_free(sra);
}
return max_degraded;
}
static int array_try_spare(char *devname, int *dfdp, struct dev_policy *pol,
struct map_ent *target, int bare,
struct supertype *st, int verbose)
{
/* This device doesn't have any md metadata
* The device policy allows 'spare' and if !bare, it allows spare-same-slot.
* If 'st' is not set, then we only know that some metadata allows this,
* others possibly don't.
* So look for a container or array to attach the device to.
* Prefer 'target' if that is set and the array is found.
*
* If st is set, then only arrays of that type are considered
* Return 0 on success, or some exit code on failure, probably 1.
*/
int rv = 1;
dev_t rdev;
struct map_ent *mp, *map = NULL;
struct mdinfo *chosen = NULL;
int dfd = *dfdp;
if (!fstat_is_blkdev(dfd, devname, &rdev))
return 1;
/*
* Now we need to find a suitable array to add this to.
* We only accept arrays that:
* - match 'st'
* - are in the same domains as the device
* - are of an size for which the device will be useful
* and we choose the one that is the most degraded
*/
if (map_lock(&map)) {
pr_err("failed to get exclusive lock on mapfile\n");
return 1;
}
for (mp = map ; mp ; mp = mp->next) {
struct supertype *st2;
struct domainlist *dl = NULL;
struct mdinfo *sra;
unsigned long long devsize, freesize = 0;
struct spare_criteria sc = {0, 0};
if (is_subarray(mp->metadata))
continue;
if (st) {
st2 = st->ss->match_metadata_desc(mp->metadata);
if (!st2 ||
(st->minor_version >= 0 &&
st->minor_version != st2->minor_version)) {
if (verbose > 1)
pr_err("not adding %s to %s as metadata type doesn't match\n",
devname, mp->path);
free(st2);
continue;
}
free(st2);
}
sra = sysfs_read(-1, mp->devnm,
GET_DEVS|GET_OFFSET|GET_SIZE|GET_STATE|
GET_COMPONENT|GET_VERSION);
if (sra)
sra->array.failed_disks = -1;
else
continue;
if (st == NULL) {
int i;
st2 = NULL;
for(i = 0; !st2 && superlist[i]; i++)
st2 = superlist[i]->match_metadata_desc(
sra->text_version);
if (!st2) {
if (verbose > 1)
pr_err("not adding %s to %s as metadata not recognised.\n",
devname, mp->path);
goto next;
}
/* Need to double check the 'act_spare' permissions applies
* to this metadata.
*/
if (!policy_action_allows(pol, st2->ss->name, act_spare))
goto next;
if (!bare && !policy_action_allows(pol, st2->ss->name,
act_spare_same_slot))
goto next;
} else
st2 = st;
/* update number of failed disks for mostly degraded
* container member */
if (sra->array.failed_disks == -1)
sra->array.failed_disks = container_members_max_degradation(map, mp);
get_dev_size(dfd, NULL, &devsize);
if (sra->component_size == 0) {
/* true for containers, here we must read superblock
* to obtain minimum spare size */
struct supertype *st3 = dup_super(st2);
int mdfd = open_dev(mp->devnm);
if (mdfd < 0) {
free(st3);
goto next;
}
if (st3->ss->load_container &&
!st3->ss->load_container(st3, mdfd, mp->path)) {
if (st3->ss->get_spare_criteria)
st3->ss->get_spare_criteria(st3, &sc);
st3->ss->free_super(st3);
}
free(st3);
close(mdfd);
}
if ((sra->component_size > 0 &&
st2->ss->validate_geometry(st2, sra->array.level, sra->array.layout,
sra->array.raid_disks, &sra->array.chunk_size,
sra->component_size,
sra->devs ? sra->devs->data_offset : INVALID_SECTORS,
devname, &freesize, sra->consistency_policy,
0) &&
freesize < sra->component_size) ||
(sra->component_size == 0 && devsize < sc.min_size)) {
if (verbose > 1)
pr_err("not adding %s to %s as it is too small\n",
devname, mp->path);
goto next;
}
/* test against target.
* If 'target' is set and 'bare' is false, we only accept
* arrays/containers that match 'target'.
* If 'target' is set and 'bare' is true, we prefer the
* array which matches 'target'.
* target is considered only if we deal with degraded array
*/
if (target && policy_action_allows(pol, st2->ss->name,
act_spare_same_slot)) {
if (strcmp(target->metadata, mp->metadata) == 0 &&
memcmp(target->uuid, mp->uuid,
sizeof(target->uuid)) == 0 &&
sra->array.failed_disks > 0) {
/* This is our target!! */
sysfs_free(chosen);
chosen = sra;
sra = NULL;
/* skip to end so we don't check any more */
while (mp->next)
mp = mp->next;
goto next;
}
/* not our target */
if (!bare)
goto next;
}
dl = domain_from_array(sra, st2->ss->name);
if (domain_test(dl, pol, st2->ss->name) != 1) {
/* domain test fails */
if (verbose > 1)
pr_err("not adding %s to %s as it is not in a compatible domain\n",
devname, mp->path);
goto next;
}
/* all tests passed, OK to add to this array */
if (!chosen) {
chosen = sra;
sra = NULL;
} else if (chosen->array.failed_disks < sra->array.failed_disks) {
sysfs_free(chosen);