/*
* mdadm - manage Linux "md" devices aka RAID arrays.
*
- * Copyright (C) 2001-2012 Neil Brown <neilb@suse.de>
+ * Copyright (C) 2001-2013 Neil Brown <neilb@suse.de>
*
*
* This program is free software; you can redistribute it and/or modify
static int is_member_busy(char *metadata_version)
{
/* check if the given member array is active */
- struct mdstat_ent *mdstat = mdstat_read(1, 0);
+ struct mdstat_ent *mdstat = mdstat_read(0, 0);
struct mdstat_ent *ent;
int busy = 0;
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",
if (tmpdev->used > 1)
continue;
- if (ident->devices &&
- !match_oneof(ident->devices, devname)) {
+ 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;
devname);
tmpdev->used = 2;
} else if (auto_assem &&
- !conf_test_metadata(tst->ss->name, (pol = devnum_policy(stb.st_rdev)),
+ !conf_test_metadata(tst->ss->name, (pol = devid_policy(stb.st_rdev)),
tst->ss->match_home(tst, c->homehost) == 1)) {
if (report_mismatch)
pr_err("%s has metadata type %s for which "
tst->ss->name, devname);
tmpdev->used = 2;
} else if (auto_assem && st == NULL &&
- !conf_test_metadata(tst->ss->name, (pol = devnum_policy(stb.st_rdev)),
+ !conf_test_metadata(tst->ss->name, (pol = devid_policy(stb.st_rdev)),
tst->ss->match_home(tst, c->homehost) == 1)) {
if (report_mismatch)
pr_err("%s has metadata type %s for which "
}
close(dfd);
- if (ident->container) {
- if (ident->container[0] == '/' &&
- !same_dev(ident->container, devname)) {
+ 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 is not the container required (%s)\n",
- devname, ident->container);
+ pr_err("%s has wrong UUID to be required container\n",
+ devname);
goto loop;
}
- if (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.
*/
/* Collect domain information from members only */
if (tmpdev && tmpdev->used == 1) {
if (!pol)
- pol = devnum_policy(stb.st_rdev);
+ pol = devid_policy(stb.st_rdev);
domain_merge(&domains, pol, tst?tst->ss->name:NULL);
}
dev_policy_free(pol);
tmpdev->devname, strerror(errno));
tmpdev->used = 2;
} else {
- struct dev_policy *pol = devnum_policy(stb.st_rdev);
+ struct dev_policy *pol = devid_policy(stb.st_rdev);
int dt = domain_test(domains, pol, NULL);
if (inargv && dt != 0)
/* take this spare as domains match
};
static int load_devices(struct devs *devices, char *devmap,
- struct mddev_ident *ident, struct supertype *st,
+ struct mddev_ident *ident, struct supertype **stp,
struct mddev_dev *devlist, struct context *c,
struct mdinfo *content,
int mdfd, char *mddev,
#ifndef MDASSEMBLE
int bitmap_done = 0;
#endif
- int most_recent = 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;
if (tmpdev->used != 1)
int dfd;
/* prepare useful information in info structures */
struct stat stb2;
- struct supertype *tst;
int err;
fstat(mdfd, &stb2);
close(mdfd);
free(devices);
free(devmap);
+ tst->ss->free_super(tst);
+ free(tst);
+ *stp = st;
return -1;
}
tst->ss->getinfo_super(tst, content, devmap + devcnt * content->array.raid_disks);
ident->uuid_set,
c->homehost);
if (err < 0) {
- pr_err("--update=%s not understood"
- " for %s metadata\n",
- c->update, tst->ss->name);
+ if (err == -1)
+ pr_err("--update=%s not understood"
+ " for %s metadata\n",
+ c->update, tst->ss->name);
tst->ss->free_super(tst);
free(tst);
close(mdfd);
close(dfd);
free(devices);
free(devmap);
+ *stp = st;
return -1;
}
if (strcmp(c->update, "uuid")==0 &&
else
bitmap_done = 1;
}
- tst->ss->free_super(tst);
} else
#endif
{
- struct supertype *tst = dup_super(st);
- int dfd;
- dfd = dev_open(devname,
- tmpdev->disposition == 'I'
- ? O_RDWR : (O_RDWR|O_EXCL));
+ int 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",
close(mdfd);
free(devices);
free(devmap);
+ tst->ss->free_super(tst);
+ free(tst);
+ *stp = st;
return -1;
}
tst->ss->getinfo_super(tst, content, devmap + devcnt * content->array.raid_disks);
- tst->ss->free_super(tst);
close(dfd);
}
devices[devcnt].i = *content;
devices[devcnt].i.disk.major = major(stb.st_rdev);
devices[devcnt].i.disk.minor = minor(stb.st_rdev);
- if (most_recent < devcnt) {
- if (devices[devcnt].i.events
- > devices[most_recent].i.events)
+
+ if (devices[devcnt].i.disk.state == 6) {
+ 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;
close(mdfd);
free(devices);
free(devmap);
+ *stp = st;
return -1;
}
if (best[i] == -1
}
devcnt++;
}
- *most_recentp = most_recent;
+ if (most_recent >= 0)
+ *most_recentp = most_recent;
*bestcntp = bestcnt;
*bestp = best;
+ *stp = st;
return devcnt;
}
int chosen_drive = -1;
int i;
- for (i = 0; i < content->array.raid_disks && i < bestcnt; i++) {
+ for (i = 0;
+ i < content->array.raid_disks * 2 && i < bestcnt;
+ i += 2) {
int j = best[i];
if (j>=0 &&
!devices[j].uptodate &&
/* 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 && i < bestcnt ; i++) {
+ for (i = 0;
+ i < content->array.raid_disks * 2 && i < bestcnt ;
+ i += 2) {
int j = best[i];
if (j >= 0 &&
!devices[j].uptodate &&
unsigned int rebuilding_cnt,
struct context *c,
int clean, char *avail,
- int start_partial_ok
+ int start_partial_ok,
+ int err_ok,
+ int was_forced
)
{
int rv;
unsigned int req_cnt;
rv = set_array_info(mdfd, st, content);
- if (rv) {
+ if (rv && !err_ok) {
pr_err("failed to set array info for %s: %s\n",
mddev, strerror(errno));
return 1;
}
st->ss->free_super(st);
sysfs_uevent(content, "change");
+ if (err_ok && okcnt < (unsigned)content->array.raid_disks)
+ /* Was partial, is still partial, so signal an error
+ * to ensure we don't retry */
+ return 1;
return 0;
}
if (content->reshape_active &&
!(content->reshape_active & RESHAPE_NO_BACKUP) &&
content->delta_disks <= 0) {
+ if (!c->backup_file) {
+ pr_err("%s: Need a backup file to complete reshape of this array.\n",
+ mddev);
+ pr_err("Please provided one with \"--backup-file=...\"\n");
+ if (c->update &&
+ strcmp(c->update, "revert-reshape") == 0)
+ pr_err("(Don't specify --update=revert-reshape again, that part succeeded.)\n");
+ return 1;
+ }
rv = sysfs_set_str(content, NULL,
"array_state", "readonly");
if (rv == 0)
} else
#endif
rv = ioctl(mdfd, RUN_ARRAY, NULL);
+ reopen_mddev(mdfd); /* drop O_EXCL */
if (rv == 0) {
if (c->verbose >= 0) {
pr_err("%s has been started with %d drive%s",
* of the stripe cache - default is 256
*/
if (256 < 4 * (content->array.chunk_size/4096)) {
- struct mdinfo *sra = sysfs_read(mdfd, 0, 0);
+ struct mdinfo *sra = sysfs_read(mdfd, NULL, 0);
if (sra)
sysfs_set_num(sra, NULL,
"stripe_cache_size",
}
}
}
+ if (content->array.level == 6 &&
+ okcnt + 1 == (unsigned)content->array.raid_disks &&
+ was_forced) {
+ struct mdinfo *sra = sysfs_read(mdfd, NULL, 0);
+ if (sra)
+ sysfs_set_str(sra, NULL,
+ "sync_action", "repair");
+ sysfs_free(sra);
+ }
return 0;
}
pr_err("failed to RUN_ARRAY %s: %s\n",
int devcnt;
unsigned int okcnt, sparecnt, rebuilding_cnt, replcnt;
int i;
+ int was_forced = 0;
int most_recent = 0;
int chosen_drive;
int change = 0;
inargv, auto_assem);
if (num_devs < 0)
return 1;
-
+
if (!st || !st->sb || !content)
return 2;
if (mp) {
struct mdinfo *dv;
/* array already exists. */
- pre_exist = sysfs_read(-1, mp->devnum, GET_LEVEL|GET_DEVS);
+ pre_exist = sysfs_read(-1, mp->devnm, GET_LEVEL|GET_DEVS);
if (pre_exist->array.level != UnSet) {
pr_err("Found some drive for an array that is already active: %s\n",
mp->path);
strcmp(mddev, chosen_name) != 0)
pr_err("Merging with already-assembled %s\n",
chosen_name);
- mdfd = open_dev_excl(mp->devnum);
+ mdfd = open_dev_excl(mp->devnm);
} else {
int trustworthy = FOREIGN;
name = content->name;
return 1;
}
if (pre_exist == NULL) {
- if (mddev_busy(fd2devnum(mdfd))) {
+ if (mddev_busy(fd2devnm(mdfd))) {
pr_err("%s already active, cannot restart it!\n",
mddev);
for (tmpdev = devlist ;
/* This is a member of a container. Try starting the array. */
int err;
err = assemble_container_content(st, mdfd, content, c,
- chosen_name);
+ chosen_name, NULL);
close(mdfd);
return err;
}
/* Ok, no bad inconsistancy, we can try updating etc */
devices = xcalloc(num_devs, sizeof(*devices));
devmap = xcalloc(num_devs, content->array.raid_disks);
- devcnt = load_devices(devices, devmap, ident, st, devlist,
+ devcnt = load_devices(devices, devmap, ident, &st, devlist,
c, content, mdfd, mddev,
&most_recent, &bestcnt, &best, inargv);
if (devcnt < 0)
/* If this device thinks that 'most_recent' has failed, then
* we must reject this device.
*/
- if (j != most_recent &&
+ if (j != most_recent && !c->force &&
content->array.raid_disks > 0 &&
devices[most_recent].i.disk.raid_disk >= 0 &&
devmap[j * content->array.raid_disks + devices[most_recent].i.disk.raid_disk] == 0) {
best[i] = -1;
continue;
}
+ /* Require event counter to be same as, or just less than,
+ * most recent. If it is bigger, it must be a stray spare and
+ * should be ignored.
+ */
if (devices[j].i.events+event_margin >=
- devices[most_recent].i.events) {
+ devices[most_recent].i.events &&
+ devices[j].i.events <=
+ devices[most_recent].i.events
+ ) {
devices[j].uptodate = 1;
if (i < content->array.raid_disks * 2) {
if (devices[j].i.recovery_start == MaxSector ||
}
}
free(devmap);
- if (c->force)
- okcnt += force_array(content, devices, best, bestcnt,
- avail, most_recent, st, c);
-
+ if (c->force) {
+ int force_ok = force_array(content, devices, best, bestcnt,
+ avail, most_recent, st, c);
+ okcnt += force_ok;
+ if (force_ok)
+ was_forced = 1;
+ }
/* Now we want to look at the superblock which the kernel will base things on
* and compare the devices that we think are working with the devices that the
* superblock thinks are working.
*/
chosen_drive = -1;
st->ss->free_super(st);
- for (i=0; chosen_drive < 0 && i<bestcnt; i++) {
+ for (i=0; chosen_drive < 0 && i<bestcnt; i+=2) {
int j = best[i];
int fd;
}
st->ss->getinfo_super(st, content, NULL);
#ifndef MDASSEMBLE
- sysfs_init(content, mdfd, 0);
+ sysfs_init(content, mdfd, NULL);
#endif
for (i=0; i<bestcnt; i++) {
int j = best[i];
change += st->ss->update_super(st, content, "force-array",
devices[chosen_drive].devname, c->verbose,
0, NULL);
+ was_forced = 1;
clean = 1;
}
pr_err(":%s has an active reshape - checking "
"if critical section needs to be restored\n",
chosen_name);
+ enable_fds(bestcnt/2);
for (i = 0; i < bestcnt/2; i++) {
int j = best[i*2];
if (j >= 0) {
/* First, fill in the map, so that udev can find our name
* as soon as we become active.
*/
- map_update(&map, fd2devnum(mdfd), content->text_version,
+ if (c->update && strcmp(c->update, "metadata")==0) {
+ content->array.major_version = 1;
+ content->array.minor_version = 0;
+ strcpy(content->text_version, "1.0");
+ }
+
+ map_update(&map, fd2devnm(mdfd), content->text_version,
content->uuid, chosen_name);
rv = start_array(mdfd, mddev, content,
chosen_drive, devices, okcnt, sparecnt,
rebuilding_cnt,
c,
- clean, avail, start_partial_ok);
+ clean, avail, start_partial_ok,
+ pre_exist != NULL,
+ was_forced);
if (rv == 1 && !pre_exist)
ioctl(mdfd, STOP_ARRAY, NULL);
free(devices);
#ifndef MDASSEMBLE
int assemble_container_content(struct supertype *st, int mdfd,
struct mdinfo *content, struct context *c,
- char *chosen_name)
+ char *chosen_name, int *result)
{
struct mdinfo *dev, *sra;
int working = 0, preexist = 0;
struct map_ent *map = NULL;
int old_raid_disks;
int start_reshape;
+ char *avail = NULL;
+ int err;
- sysfs_init(content, mdfd, 0);
+ sysfs_init(content, mdfd, NULL);
- sra = sysfs_read(mdfd, 0, GET_VERSION);
+ sra = sysfs_read(mdfd, NULL, GET_VERSION);
if (sra == NULL || strcmp(sra->text_version, content->text_version) != 0) {
if (content->array.major_version == -1 &&
content->array.minor_version == -2 &&
if (sra)
sysfs_free(sra);
old_raid_disks = content->array.raid_disks - content->delta_disks;
- for (dev = content->devs; dev; dev = dev->next)
+ avail = xcalloc(content->array.raid_disks, 1);
+ for (dev = content->devs; dev; dev = dev->next) {
+ if (dev->disk.raid_disk >= 0)
+ avail[dev->disk.raid_disk] = 1;
if (sysfs_add_disk(content, dev, 1) == 0) {
if (dev->disk.raid_disk >= old_raid_disks &&
content->reshape_active)
working++;
} else if (errno == EEXIST)
preexist++;
- if (working + expansion == 0)
+ }
+ if (working + expansion == 0 && c->runstop <= 0) {
+ free(avail);
return 1;/* Nothing new, don't try to start */
-
- map_update(&map, fd2devnum(mdfd),
+ }
+ map_update(&map, fd2devnm(mdfd),
content->text_version,
content->uuid, chosen_name);
- if (c->runstop > 0 ||
- (working + preexist + expansion) >=
- content->array.working_disks) {
- int err;
-
- if (start_reshape) {
- int spare = content->array.raid_disks + expansion;
- if (restore_backup(st, content,
- working,
- spare, c->backup_file, c->verbose) == 1)
- return 1;
-
- err = sysfs_set_str(content, NULL,
- "array_state", "readonly");
- if (err)
- return 1;
-
- if (st->ss->external) {
- if (!mdmon_running(st->container_dev))
- start_mdmon(st->container_dev);
- ping_monitor_by_id(st->container_dev);
- if (mdmon_running(st->container_dev) &&
- st->update_tail == NULL)
- st->update_tail = &st->updates;
- }
-
- err = Grow_continue(mdfd, st, content, c->backup_file,
- c->freeze_reshape);
- } else switch(content->array.level) {
- case LEVEL_LINEAR:
- case LEVEL_MULTIPATH:
- case 0:
- err = sysfs_set_str(content, NULL, "array_state",
- c->readonly ? "readonly" : "active");
- break;
- default:
- err = sysfs_set_str(content, NULL, "array_state",
- "readonly");
- /* start mdmon if needed. */
- if (!err) {
- if (!mdmon_running(st->container_dev))
- start_mdmon(st->container_dev);
- ping_monitor_by_id(st->container_dev);
- }
- break;
- }
- if (!err)
- sysfs_set_safemode(content, content->safe_mode_delay);
-
- /* Block subarray here if it is not reshaped now
- * It has be blocked a little later to allow mdmon to switch in
- * in to R/W state
- */
- if (st->ss->external && content->recovery_blocked &&
- !start_reshape)
- block_subarray(content);
- if (c->verbose >= 0) {
- if (err)
- pr_err("array %s now has %d device%s",
- chosen_name, working + preexist,
- working + preexist == 1 ? "":"s");
- else
- pr_err("Started %s with %d device%s",
- chosen_name, working + preexist,
- working + preexist == 1 ? "":"s");
+ if (enough(content->array.level, content->array.raid_disks,
+ content->array.layout, content->array.state & 1, avail) == 0) {
+ if (c->export && result)
+ *result |= INCR_NO;
+ else if (c->verbose >= 0) {
+ pr_err("%s assembled with %d device%s",
+ chosen_name, preexist + working,
+ preexist + working == 1 ? "":"s");
if (preexist)
fprintf(stderr, " (%d new)", working);
- if (expansion)
- fprintf(stderr, " ( + %d for expansion)",
- expansion);
- fprintf(stderr, "\n");
+ fprintf(stderr, " but not started\n");
}
- if (!err)
- wait_for(chosen_name, mdfd);
- return err;
- /* FIXME should have an O_EXCL and wait for read-auto */
- } else {
- if (c->verbose >= 0) {
+ free(avail);
+ return 1;
+ }
+ free(avail);
+
+ if (c->runstop <= 0 &&
+ (working + preexist + expansion) <
+ content->array.working_disks) {
+ if (c->export && result)
+ *result |= INCR_UNSAFE;
+ else if (c->verbose >= 0) {
pr_err("%s assembled with %d device%s",
chosen_name, preexist + working,
preexist + working == 1 ? "":"s");
if (preexist)
fprintf(stderr, " (%d new)", working);
- fprintf(stderr, " but not started\n");
+ fprintf(stderr, " but not safe to start\n");
}
return 1;
}
+
+
+ if (start_reshape) {
+ int spare = content->array.raid_disks + expansion;
+ if (restore_backup(st, content,
+ working,
+ spare, c->backup_file, c->verbose) == 1)
+ return 1;
+
+ err = sysfs_set_str(content, NULL,
+ "array_state", "readonly");
+ if (err)
+ return 1;
+
+ if (st->ss->external) {
+ if (!mdmon_running(st->container_devnm))
+ start_mdmon(st->container_devnm);
+ ping_monitor(st->container_devnm);
+ if (mdmon_running(st->container_devnm) &&
+ st->update_tail == NULL)
+ st->update_tail = &st->updates;
+ }
+
+ err = Grow_continue(mdfd, st, content, c->backup_file,
+ c->freeze_reshape);
+ } else switch(content->array.level) {
+ case LEVEL_LINEAR:
+ case LEVEL_MULTIPATH:
+ case 0:
+ err = sysfs_set_str(content, NULL, "array_state",
+ c->readonly ? "readonly" : "active");
+ break;
+ default:
+ err = sysfs_set_str(content, NULL, "array_state",
+ "readonly");
+ /* start mdmon if needed. */
+ if (!err) {
+ if (!mdmon_running(st->container_devnm))
+ start_mdmon(st->container_devnm);
+ ping_monitor(st->container_devnm);
+ }
+ break;
+ }
+ if (!err)
+ sysfs_set_safemode(content, content->safe_mode_delay);
+
+ /* Block subarray here if it is not reshaped now
+ * It has be blocked a little later to allow mdmon to switch in
+ * in to R/W state
+ */
+ if (st->ss->external && content->recovery_blocked &&
+ !start_reshape)
+ block_subarray(content);
+
+ if (c->export && result) {
+ if (err)
+ *result |= INCR_NO;
+ else
+ *result |= INCR_YES;
+ } else if (c->verbose >= 0) {
+ if (err)
+ pr_err("array %s now has %d device%s",
+ chosen_name, working + preexist,
+ working + preexist == 1 ? "":"s");
+ else
+ pr_err("Started %s with %d device%s",
+ chosen_name, working + preexist,
+ working + preexist == 1 ? "":"s");
+ if (preexist)
+ fprintf(stderr, " (%d new)", working);
+ if (expansion)
+ fprintf(stderr, " ( + %d for expansion)",
+ expansion);
+ fprintf(stderr, "\n");
+ }
+ if (!err)
+ wait_for(chosen_name, mdfd);
+ return err;
+ /* FIXME should have an O_EXCL and wait for read-auto */
}
#endif
-
projects / thirdparty / mdadm.git / blobdiff