/*
* 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
char *name, int *uuid,
int subdevs, struct mddev_dev *devlist,
struct shape *s,
- struct context *c)
+ struct context *c, unsigned long long data_offset)
{
/*
* Create a new raid array.
unsigned long long minsize=0, maxsize=0;
char *mindisc = NULL;
char *maxdisc = NULL;
- int dnum;
+ int dnum, raid_disk_num;
struct mddev_dev *dv;
int fail=0, warn=0;
struct stat stb;
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;
memset(&info, 0, sizeof(info));
if (s->level == UnSet && st && st->ss->default_geometry)
close(fd);
}
if (st && st->ss->external && s->sparedisks) {
- pr_err("This metadata type does not support "
- "spare disks at create time\n");
+ pr_err("This metadata type does not support spare disks at create time\n");
return 1;
}
- if (subdevs > s->raiddisks+s->sparedisks) {
+ 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) {
+ 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;
}
newsize = s->size * 2;
if (st && ! st->ss->validate_geometry(st, s->level, s->layout, s->raiddisks,
&s->chunk, s->size*2,
- INVALID_SECTORS, NULL,
+ data_offset, NULL,
&newsize, c->verbose>=0))
return 1;
if (do_default_chunk) {
/* default chunk was just set */
if (c->verbose > 0)
- pr_err("chunk size "
- "defaults to %dK\n", s->chunk);
+ pr_err("chunk size defaults to %dK\n", s->chunk);
s->size &= ~(unsigned long long)(s->chunk - 1);
do_default_chunk = 0;
}
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)
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",
}
close(dfd);
info.array.working_disks++;
- if (dnum < s->raiddisks)
+ if (dnum < s->raiddisks && dv->disposition != 'j')
info.array.active_disks++;
if (st == NULL) {
struct createinfo *ci = conf_get_create_info();
switch (st->ss->validate_geometry(
st, s->level, s->layout, s->raiddisks,
&s->chunk, s->size*2,
- INVALID_SECTORS, dname,
+ dv->data_offset, dname,
&freesize, c->verbose > 0)) {
case -1: /* Not valid, message printed, and not
* worth checking any further */
dname, strerror(errno));
exit(2);
}
- pr_err("device %s not suitable "
- "for any style of array\n",
+ pr_err("device %s not suitable for any style of array\n",
dname);
exit(2);
}
if (!st->ss->validate_geometry(st, s->level, s->layout,
s->raiddisks,
&s->chunk, s->size*2,
- INVALID_SECTORS,
+ dv->data_offset,
dname, &freesize,
c->verbose >= 0)) {
- pr_err("%s is not suitable for "
- "this array.\n",
+ 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 */
if (do_default_chunk) {
/* default chunk was just set */
if (c->verbose > 0)
- pr_err("chunk size "
- "defaults to %dK\n", s->chunk);
+ 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",
+ pr_err("%s is smaller than given size. %lluK < %lluK + metadata\n",
dname, freesize, s->size);
fail = 1;
continue;
mindisc = dname;
minsize = freesize;
}
+ skip_size_check:
if (c->runstop != 1 || c->verbose >= 0) {
int fd = open(dname, O_RDONLY);
if (fd <0 ) {
}
}
if (s->raiddisks + s->sparedisks > st->max_devs) {
- pr_err("Too many devices:"
- " %s metadata only supports %d\n",
+ pr_err("Too many devices: %s metadata only supports %d\n",
st->ss->name, st->max_devs);
return 1;
}
if (!st->ss->validate_geometry(st, s->level, s->layout,
s->raiddisks,
&s->chunk, minsize*2,
- INVALID_SECTORS,
+ data_offset,
NULL, NULL, 0)) {
pr_err("devices too large for RAID level %d\n", s->level);
return 1;
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",
st->ss->name);
warn = 1;
}
+ st->nodes = c->nodes;
+ st->cluster_name = c->homecluster;
if (warn) {
if (c->runstop!= 1) {
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");
+ pr_err("another array by this name is already running.\n");
goto abort_locked;
}
}
* including spares
* failed_disks is the number of disks marked failed
*
- * Ideally, the kernel would keep these (except raid_disks)
+ * 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.
/* 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
name = strrchr(mddev, '/');
if (name) {
name++;
- if (strncmp(name, "md_d", 4)==0 &&
- strlen(name) > 4 &&
- isdigit(name[4]) &&
+ if (strncmp(name, "md_", 3)==0 &&
+ strlen(name) > 3 &&
(name-mddev) == 5 /* /dev/ */)
name += 3;
else if (strncmp(name, "md", 2)==0 &&
}
}
if (!st->ss->init_super(st, &info.array, s->size, name, c->homehost, uuid,
- INVALID_SECTORS))
+ data_offset))
goto abort_locked;
total_slots = info.array.nr_disks;
st->ss->getinfo_super(st, &info, NULL);
- sysfs_init(&info, mdfd, 0);
+ sysfs_init(&info, mdfd, NULL);
if (did_default && c->verbose >= 0) {
if (is_subarray(info.text_version)) {
- int dnum = devname2devnum(info.text_version+1);
- char *path;
- int mdp = get_mdp_major();
+ char devnm[32];
+ char *ep;
struct mdinfo *mdi;
- if (dnum > 0)
- path = map_dev(MD_MAJOR, dnum, 1);
- else
- path = map_dev(mdp, (-1-dnum)<< 6, 1);
- mdi = sysfs_read(-1, dnum, GET_VERSION);
+ 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", path);
+ 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);
+ pr_err("Defaulting to version %s metadata\n", info.text_version);
}
- map_update(&map, fd2devnum(mdfd), info.text_version,
+ map_update(&map, fd2devnm(mdfd), info.text_version,
info.uuid, chosen_name);
- map_unlock(&map);
+ /* 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;
#endif
}
- if (s->bitmap_file && strcmp(s->bitmap_file, "internal")==0) {
+ 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;
+ goto abort_locked;
}
if (!st->ss->add_internal_bitmap) {
pr_err("internal bitmaps not supported with %s metadata\n",
st->ss->name);
- goto abort;
+ 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;
+ goto abort_locked;
}
s->bitmap_file = NULL;
}
- sysfs_init(&info, mdfd, 0);
+ sysfs_init(&info, mdfd, NULL);
- if (st->ss->external && st->container_dev != NoMdDev) {
+ if (st->ss->external && st->container_devnm[0]) {
/* member */
/* When creating a member, we need to be careful
*
* For now, fail if it is already running.
*/
- container_fd = open_dev_excl(st->container_dev);
+ container_fd = open_dev_excl(st->container_devnm);
if (container_fd < 0) {
- pr_err("Cannot get exclusive "
- "open on container - weird.\n");
- goto abort;
+ pr_err("Cannot get exclusive open on container - weird.\n");
+ goto abort_locked;
}
- if (mdmon_running(st->container_dev)) {
+ if (mdmon_running(st->container_devnm)) {
if (c->verbose)
- pr_err("reusing mdmon "
- "for %s.\n",
- devnum2devname(st->container_dev));
+ pr_err("reusing mdmon for %s.\n",
+ st->container_devnm);
st->update_tail = &st->updates;
} else
need_mdmon = 1;
if (rv) {
pr_err("failed to set array info for %s: %s\n",
mddev, strerror(errno));
- goto abort;
+ goto abort_locked;
}
if (s->bitmap_file) {
c->delay, s->write_behind,
bitmapsize,
major_num)) {
- goto abort;
+ 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;
+ 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;
+ 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, dv = devlist ; dv ;
+ for (dnum=0, raid_disk_num=0, dv = devlist ; dv ;
dv=(dv->next)?(dv->next):moved_disk, dnum++) {
int fd;
struct stat stb;
if (dnum >= total_slots)
abort();
if (dnum == insert_point) {
+ raid_disk_num += 1;
moved_disk = dv;
continue;
}
- if (strcasecmp(dv->devname, "missing")==0)
+ 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)
*inf = info;
inf->disk.number = dnum;
- inf->disk.raid_disk = dnum;
- if (inf->disk.raid_disk < s->raiddisks)
+ 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
fd = -1;
else {
if (st->ss->external &&
- st->container_dev != NoMdDev)
+ 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",
+ pr_err("failed to open %s after earlier success - aborting\n",
dv->devname);
- goto abort;
+ goto abort_locked;
}
fstat(fd, &stb);
inf->disk.major = major(stb.st_rdev);
if (fd >= 0)
remove_partitions(fd);
if (st->ss->add_to_super(st, &inf->disk,
- fd, dv->devname)) {
+ fd, dv->devname,
+ dv->data_offset)) {
ioctl(mdfd, STOP_ARRAY, NULL);
- goto abort;
+ goto abort_locked;
}
st->ss->getinfo_super(st, inf, NULL);
safe_mode_delay = inf->safe_mode_delay;
rv = add_disk(mdfd, st, &info, inf);
if (rv) {
- pr_err("ADD_NEW_DISK for %s "
- "failed: %s\n",
+ pr_err("ADD_NEW_DISK for %s failed: %s\n",
dv->devname, strerror(errno));
- goto abort;
+ goto abort_locked;
}
break;
}
* the subarray cursor such that ->getinfo_super once
* again returns container info.
*/
- map_lock(&map);
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, fd2devnum(mdfd),
+ map_update(&map, fd2devnm(mdfd),
info_new.text_version,
info_new.uuid, chosen_name);
- me = map_by_devnum(&map, st->container_dev);
+ me = map_by_devnm(&map, st->container_devnm);
}
if (st->ss->write_init_super(st)) {
char *path = xstrdup(me->path);
st->ss->getinfo_super(st, &info_new, NULL);
- map_update(&map, st->container_dev,
+ map_update(&map, st->container_devnm,
info_new.text_version,
info_new.uuid, path);
free(path);
}
- map_unlock(&map);
flush_metadata_updates(st);
st->ss->free_super(st);
}
}
+ map_unlock(&map);
free(infos);
if (s->level == LEVEL_CONTAINER) {
}
sysfs_set_safemode(&info, safe_mode_delay);
if (err) {
- pr_err("failed to"
- " activate array.\n");
+ pr_err("failed to activate array.\n");
ioctl(mdfd, STOP_ARRAY, NULL);
goto abort;
}
pr_err("RUN_ARRAY failed: %s\n",
strerror(errno));
if (info.array.chunk_size & (info.array.chunk_size-1)) {
- cont_err("Problem may be that "
- "chunk size is not a power of 2\n");
+ cont_err("Problem may be that chunk size is not a power of 2\n");
}
ioctl(mdfd, STOP_ARRAY, NULL);
goto abort;
}
+ /* if start_ro module parameter is set, array is
+ * auto-read-only, which is bad as the resync won't
+ * start. So lets make it read-write now.
+ */
+ ioctl(mdfd, RESTART_ARRAY_RW, NULL);
}
if (c->verbose >= 0)
pr_err("array %s started.\n", mddev);
- if (st->ss->external && st->container_dev != NoMdDev) {
+ if (st->ss->external && st->container_devnm[0]) {
if (need_mdmon)
- start_mdmon(st->container_dev);
+ start_mdmon(st->container_devnm);
- ping_monitor_by_id(st->container_dev);
+ ping_monitor(st->container_devnm);
close(container_fd);
}
wait_for(chosen_name, mdfd);
abort:
map_lock(&map);
abort_locked:
- map_remove(&map, fd2devnum(mdfd));
+ map_remove(&map, fd2devnm(mdfd));
map_unlock(&map);
if (mdfd >= 0)