X-Git-Url: http://git.ipfire.org/?a=blobdiff_plain;f=Grow.c;h=39110b852e7c8979eb45023df072ae536f5a5221;hb=fc54fe7a7e77fdb6316d332d4d41d0ed2293d6be;hp=0588590e877b8febd2aa31e6d7596846f89cd6e3;hpb=b0b67933dc05d6af0a7d3c8ed356ea81021ff70a;p=thirdparty%2Fmdadm.git diff --git a/Grow.c b/Grow.c index 0588590e..39110b85 100644 --- a/Grow.c +++ b/Grow.c @@ -24,6 +24,7 @@ #include "mdadm.h" #include "dlink.h" #include +#include #include #include #include @@ -34,10 +35,6 @@ #include "md_u.h" #include "md_p.h" -#ifndef offsetof -#define offsetof(t,f) ((size_t)&(((t*)0)->f)) -#endif - int restore_backup(struct supertype *st, struct mdinfo *content, int working_disks, @@ -90,11 +87,9 @@ int restore_backup(struct supertype *st, } free(fdlist); if (err) { - pr_err("Failed to restore critical" - " section for reshape - sorry.\n"); + pr_err("Failed to restore critical section for reshape - sorry.\n"); if (!backup_file) - pr_err("Possibly you need" - " to specify a --backup-file\n"); + pr_err("Possibly you need to specify a --backup-file\n"); return 1; } @@ -114,13 +109,13 @@ int Grow_Add_device(char *devname, int fd, char *newdev) */ struct mdinfo info; - struct stat stb; + dev_t rdev; int nfd, fd2; int d, nd; struct supertype *st = NULL; char *subarray = NULL; - if (ioctl(fd, GET_ARRAY_INFO, &info.array) < 0) { + if (md_get_array_info(fd, &info.array) < 0) { pr_err("cannot get array info for %s\n", devname); return 1; } @@ -150,9 +145,7 @@ int Grow_Add_device(char *devname, int fd, char *newdev) free(st); return 1; } - fstat(nfd, &stb); - if ((stb.st_mode & S_IFMT) != S_IFBLK) { - pr_err("%s is not a block device!\n", newdev); + if (!fstat_is_blkdev(nfd, newdev, &rdev)) { close(nfd); free(st); return 1; @@ -166,7 +159,7 @@ int Grow_Add_device(char *devname, int fd, char *newdev) st->ss->free_super(st); disk.number = d; - if (ioctl(fd, GET_DISK_INFO, &disk) < 0) { + if (md_get_disk_info(fd, &disk) < 0) { pr_err("cannot get device detail for device %d\n", d); close(nfd); @@ -203,8 +196,8 @@ int Grow_Add_device(char *devname, int fd, char *newdev) */ info.disk.number = d; - info.disk.major = major(stb.st_rdev); - info.disk.minor = minor(stb.st_rdev); + info.disk.major = major(rdev); + info.disk.minor = minor(rdev); info.disk.raid_disk = d; info.disk.state = (1 << MD_DISK_SYNC) | (1 << MD_DISK_ACTIVE); st->ss->update_super(st, &info, "linear-grow-new", newdev, @@ -226,7 +219,7 @@ int Grow_Add_device(char *devname, int fd, char *newdev) * Now go through and update all superblocks */ - if (ioctl(fd, GET_ARRAY_INFO, &info.array) < 0) { + if (md_get_array_info(fd, &info.array) < 0) { pr_err("cannot get array info for %s\n", devname); return 1; } @@ -237,7 +230,7 @@ int Grow_Add_device(char *devname, int fd, char *newdev) char *dv; disk.number = d; - if (ioctl(fd, GET_DISK_INFO, &disk) < 0) { + if (md_get_disk_info(fd, &disk) < 0) { pr_err("cannot get device detail for device %d\n", d); return 1; @@ -293,16 +286,18 @@ int Grow_addbitmap(char *devname, int fd, struct context *c, struct shape *s) struct supertype *st; char *subarray = NULL; int major = BITMAP_MAJOR_HI; - int vers = md_get_version(fd); unsigned long long bitmapsize, array_size; + struct mdinfo *mdi; - if (vers < 9003) { - major = BITMAP_MAJOR_HOSTENDIAN; - pr_err("Warning - bitmaps created on this kernel" - " are not portable\n" - " between different architectures. Consider upgrading" - " the Linux kernel.\n"); - } + /* + * We only ever get called if s->bitmap_file is != NULL, so this check + * is just here to quiet down static code checkers. + */ + if (!s->bitmap_file) + return 1; + + if (strcmp(s->bitmap_file, "clustered") == 0) + major = BITMAP_MAJOR_CLUSTERED; if (ioctl(fd, GET_BITMAP_FILE, &bmf) != 0) { if (errno == ENOMEM) @@ -312,8 +307,8 @@ int Grow_addbitmap(char *devname, int fd, struct context *c, struct shape *s) return 1; } if (bmf.pathname[0]) { - if (strcmp(s->bitmap_file,"none")==0) { - if (ioctl(fd, SET_BITMAP_FILE, -1)!= 0) { + if (strcmp(s->bitmap_file,"none") == 0) { + if (ioctl(fd, SET_BITMAP_FILE, -1) != 0) { pr_err("failed to remove bitmap %s\n", bmf.pathname); return 1; @@ -324,21 +319,23 @@ int Grow_addbitmap(char *devname, int fd, struct context *c, struct shape *s) devname, bmf.pathname); return 1; } - if (ioctl(fd, GET_ARRAY_INFO, &array) != 0) { + if (md_get_array_info(fd, &array) != 0) { pr_err("cannot get array status for %s\n", devname); return 1; } - if (array.state & (1<bitmap_file, "none")==0) { - array.state &= ~(1< (0x7fffffffULL<<9)) { + array_size > (0x7fffffffULL << 9)) { /* Array is big enough that we cannot trust array.size * try other approaches */ @@ -366,7 +363,9 @@ int Grow_addbitmap(char *devname, int fd, struct context *c, struct shape *s) } if (array.level == 10) { - int ncopies = (array.layout&255)*((array.layout>>8)&255); + int ncopies; + + ncopies = (array.layout & 255) * ((array.layout >> 8) & 255); bitmapsize = bitmapsize * array.raid_disks / ncopies; } @@ -382,66 +381,87 @@ int Grow_addbitmap(char *devname, int fd, struct context *c, struct shape *s) free(st); return 1; } - if (strcmp(s->bitmap_file, "internal") == 0) { + + mdi = sysfs_read(fd, NULL, GET_CONSISTENCY_POLICY); + if (mdi) { + if (mdi->consistency_policy == CONSISTENCY_POLICY_PPL) { + pr_err("Cannot add bitmap to array with PPL\n"); + free(mdi); + free(st); + return 1; + } + free(mdi); + } + + if (strcmp(s->bitmap_file, "internal") == 0 || + strcmp(s->bitmap_file, "clustered") == 0) { int rv; int d; int offset_setable = 0; - struct mdinfo *mdi; if (st->ss->add_internal_bitmap == NULL) { - pr_err("Internal bitmaps not supported " - "with %s metadata\n", st->ss->name); + pr_err("Internal bitmaps not supported with %s metadata\n", st->ss->name); return 1; } + st->nodes = c->nodes; + st->cluster_name = c->homecluster; mdi = sysfs_read(fd, NULL, GET_BITMAP_LOCATION); if (mdi) offset_setable = 1; - for (d=0; d< st->max_devs; d++) { + for (d = 0; d < st->max_devs; d++) { mdu_disk_info_t disk; char *dv; + int fd2; + disk.number = d; - if (ioctl(fd, GET_DISK_INFO, &disk) < 0) + if (md_get_disk_info(fd, &disk) < 0) continue; - if (disk.major == 0 && - disk.minor == 0) + if (disk.major == 0 && disk.minor == 0) continue; - if ((disk.state & (1<ss->load_super(st, fd2, NULL)==0) { - if (st->ss->add_internal_bitmap( - st, - &s->bitmap_chunk, c->delay, s->write_behind, - bitmapsize, offset_setable, - major) - ) - st->ss->write_bitmap(st, fd2); - else { - pr_err("failed to create internal bitmap" - " - chunksize problem.\n"); - close(fd2); - return 1; - } + if (!dv) + continue; + fd2 = dev_open(dv, O_RDWR); + if (fd2 < 0) + continue; + rv = st->ss->load_super(st, fd2, NULL); + if (!rv) { + rv = st->ss->add_internal_bitmap( + st, &s->bitmap_chunk, c->delay, + s->write_behind, bitmapsize, + offset_setable, major); + if (!rv) { + st->ss->write_bitmap(st, fd2, + NodeNumUpdate); + } else { + pr_err("failed to create internal bitmap - chunksize problem.\n"); } - close(fd2); + } else { + pr_err("failed to load super-block.\n"); } + close(fd2); + if (rv) + return 1; } if (offset_setable) { st->ss->getinfo_super(st, mdi, NULL); - sysfs_init(mdi, fd, NULL); + if (sysfs_init(mdi, fd, NULL)) { + pr_err("failed to intialize sysfs.\n"); + free(mdi); + } rv = sysfs_set_num_signed(mdi, NULL, "bitmap/location", mdi->bitmap_offset); + free(mdi); } else { - array.state |= (1<bitmap_file, "clustered") == 0) + array.state |= (1 << MD_SB_CLUSTERED); + array.state |= (1 << MD_SB_BITMAP_PRESENT); + rv = md_set_array_info(fd, &array); } if (rv < 0) { if (errno == EBUSY) - pr_err("Cannot add bitmap while array is" - " resyncing or reshaping etc.\n"); + pr_err("Cannot add bitmap while array is resyncing or reshaping etc.\n"); pr_err("failed to set internal bitmap.\n"); return 1; } @@ -457,10 +477,10 @@ int Grow_addbitmap(char *devname, int fd, struct context *c, struct shape *s) char *dv; int fd2; disk.number = d; - if (ioctl(fd, GET_DISK_INFO, &disk) < 0) + if (md_get_disk_info(fd, &disk) < 0) continue; - if ((disk.major==0 && disk.minor==0) || - (disk.state & (1<bitmap_file, c->force, (char*)uuid, s->bitmap_chunk, - c->delay, s->write_behind, bitmapsize, major)) { + if (CreateBitmap(s->bitmap_file, c->force, (char*)uuid, + s->bitmap_chunk, c->delay, s->write_behind, + bitmapsize, major)) { return 1; } bitmap_fd = open(s->bitmap_file, O_RDWR); if (bitmap_fd < 0) { - pr_err("weird: %s cannot be opened\n", - s->bitmap_file); + pr_err("weird: %s cannot be opened\n", s->bitmap_file); return 1; } if (ioctl(fd, SET_BITMAP_FILE, bitmap_fd) < 0) { int err = errno; if (errno == EBUSY) - pr_err("Cannot add bitmap while array is" - " resyncing or reshaping etc.\n"); + pr_err("Cannot add bitmap while array is resyncing or reshaping etc.\n"); pr_err("Cannot set bitmap file for %s: %s\n", devname, strerror(err)); return 1; @@ -503,6 +522,178 @@ int Grow_addbitmap(char *devname, int fd, struct context *c, struct shape *s) return 0; } +int Grow_consistency_policy(char *devname, int fd, struct context *c, struct shape *s) +{ + struct supertype *st; + struct mdinfo *sra; + struct mdinfo *sd; + char *subarray = NULL; + int ret = 0; + char container_dev[PATH_MAX]; + + if (s->consistency_policy != CONSISTENCY_POLICY_RESYNC && + s->consistency_policy != CONSISTENCY_POLICY_PPL) { + pr_err("Operation not supported for consistency policy %s\n", + map_num(consistency_policies, s->consistency_policy)); + return 1; + } + + st = super_by_fd(fd, &subarray); + if (!st) + return 1; + + sra = sysfs_read(fd, NULL, GET_CONSISTENCY_POLICY|GET_LEVEL| + GET_DEVS|GET_STATE); + if (!sra) { + ret = 1; + goto free_st; + } + + if (s->consistency_policy == CONSISTENCY_POLICY_PPL && + !st->ss->write_init_ppl) { + pr_err("%s metadata does not support PPL\n", st->ss->name); + ret = 1; + goto free_info; + } + + if (sra->array.level != 5) { + pr_err("Operation not supported for array level %d\n", + sra->array.level); + ret = 1; + goto free_info; + } + + if (sra->consistency_policy == (unsigned)s->consistency_policy) { + pr_err("Consistency policy is already %s\n", + map_num(consistency_policies, s->consistency_policy)); + ret = 1; + goto free_info; + } else if (sra->consistency_policy != CONSISTENCY_POLICY_RESYNC && + sra->consistency_policy != CONSISTENCY_POLICY_PPL) { + pr_err("Current consistency policy is %s, cannot change to %s\n", + map_num(consistency_policies, sra->consistency_policy), + map_num(consistency_policies, s->consistency_policy)); + ret = 1; + goto free_info; + } + + if (subarray) { + char *update; + + if (s->consistency_policy == CONSISTENCY_POLICY_PPL) + update = "ppl"; + else + update = "no-ppl"; + + sprintf(container_dev, "/dev/%s", st->container_devnm); + + ret = Update_subarray(container_dev, subarray, update, NULL, + c->verbose); + if (ret) + goto free_info; + } + + if (s->consistency_policy == CONSISTENCY_POLICY_PPL) { + struct mdinfo info; + + if (subarray) { + struct mdinfo *mdi; + int cfd; + + cfd = open(container_dev, O_RDWR|O_EXCL); + if (cfd < 0) { + pr_err("Failed to open %s\n", container_dev); + ret = 1; + goto free_info; + } + + ret = st->ss->load_container(st, cfd, st->container_devnm); + close(cfd); + + if (ret) { + pr_err("Cannot read superblock for %s\n", + container_dev); + goto free_info; + } + + mdi = st->ss->container_content(st, subarray); + info = *mdi; + free(mdi); + } + + for (sd = sra->devs; sd; sd = sd->next) { + int dfd; + char *devpath; + + if ((sd->disk.state & (1 << MD_DISK_SYNC)) == 0) + continue; + + devpath = map_dev(sd->disk.major, sd->disk.minor, 0); + dfd = dev_open(devpath, O_RDWR); + if (dfd < 0) { + pr_err("Failed to open %s\n", devpath); + ret = 1; + goto free_info; + } + + if (!subarray) { + ret = st->ss->load_super(st, dfd, NULL); + if (ret) { + pr_err("Failed to load super-block.\n"); + close(dfd); + goto free_info; + } + + ret = st->ss->update_super(st, sra, "ppl", devname, + c->verbose, 0, NULL); + if (ret) { + close(dfd); + st->ss->free_super(st); + goto free_info; + } + st->ss->getinfo_super(st, &info, NULL); + } + + ret |= sysfs_set_num(sra, sd, "ppl_sector", info.ppl_sector); + ret |= sysfs_set_num(sra, sd, "ppl_size", info.ppl_size); + + if (ret) { + pr_err("Failed to set PPL attributes for %s\n", + sd->sys_name); + close(dfd); + st->ss->free_super(st); + goto free_info; + } + + ret = st->ss->write_init_ppl(st, &info, dfd); + if (ret) + pr_err("Failed to write PPL\n"); + + close(dfd); + + if (!subarray) + st->ss->free_super(st); + + if (ret) + goto free_info; + } + } + + ret = sysfs_set_str(sra, NULL, "consistency_policy", + map_num(consistency_policies, + s->consistency_policy)); + if (ret) + pr_err("Failed to change array consistency policy\n"); + +free_info: + sysfs_free(sra); +free_st: + free(st); + free(subarray); + + return ret; +} + /* * When reshaping an array we might need to backup some data. * This is written to all spares with a 'super_block' describing it. @@ -622,11 +813,9 @@ static void unfreeze(struct supertype *st) char buf[20]; if (sra && - sysfs_get_str(sra, NULL, "sync_action", buf, 20) > 0 - && strcmp(buf, "frozen\n") == 0) { - printf("unfreeze\n"); + sysfs_get_str(sra, NULL, "sync_action", buf, 20) > 0 && + strcmp(buf, "frozen\n") == 0) sysfs_set_str(sra, NULL, "sync_action", "idle"); - } sysfs_free(sra); } } @@ -729,15 +918,28 @@ int start_reshape(struct mdinfo *sra, int already_running, if (!already_running) sysfs_set_num(sra, NULL, "sync_min", sync_max_to_set); err = err ?: sysfs_set_num(sra, NULL, "sync_max", sync_max_to_set); - if (!already_running) - err = err ?: sysfs_set_str(sra, NULL, "sync_action", "reshape"); - + if (!already_running && err == 0) { + int cnt = 5; + do { + err = sysfs_set_str(sra, NULL, "sync_action", "reshape"); + if (err) + sleep(1); + } while (err && errno == EBUSY && cnt-- > 0); + } return err; } void abort_reshape(struct mdinfo *sra) { sysfs_set_str(sra, NULL, "sync_action", "idle"); + /* + * Prior to kernel commit: 23ddff3792f6 ("md: allow suspend_lo and + * suspend_hi to decrease as well as increase.") + * you could only increase suspend_{lo,hi} unless the region they + * covered was empty. So to reset to 0, you need to push suspend_lo + * up past suspend_hi first. So to maximize the chance of mdadm + * working on all kernels, we want to keep doing that. + */ sysfs_set_num(sra, NULL, "suspend_lo", 0x7FFFFFFFFFFFFFFFULL); sysfs_set_num(sra, NULL, "suspend_hi", 0); sysfs_set_num(sra, NULL, "suspend_lo", 0); @@ -754,6 +956,26 @@ int remove_disks_for_takeover(struct supertype *st, struct mdinfo *remaining; int slot; + if (st->ss->external) { + int rv = 0; + struct mdinfo *arrays = st->ss->container_content(st, NULL); + /* + * containter_content returns list of arrays in container + * If arrays->next is not NULL it means that there are + * 2 arrays in container and operation should be blocked + */ + if (arrays) { + if (arrays->next) + rv = 1; + sysfs_free(arrays); + if (rv) { + pr_err("Error. Cannot perform operation on /dev/%s\n", st->devnm); + pr_err("For this operation it MUST be single array in container\n"); + return rv; + } + } + } + if (sra->array.level == 10) nr_of_copies = layout & 0xff; else if (sra->array.level == 1) @@ -856,7 +1078,8 @@ int reshape_prepare_fdlist(char *devname, for (sd = sra->devs; sd; sd = sd->next) { if (sd->disk.state & (1<disk.state & (1<disk.state & (1<disk.raid_disk < raid_disks) { char *dn = map_dev(sd->disk.major, sd->disk.minor, 1); fdlist[sd->disk.raid_disk] @@ -920,8 +1143,7 @@ int reshape_open_backup_file(char *backup_file, dev = stb.st_dev; fstat(fd, &stb); if (stb.st_rdev == dev) { - pr_err("backup file must NOT be" - " on the array being reshaped.\n"); + pr_err("backup file must NOT be on the array being reshaped.\n"); close(*fdlist); return 0; } @@ -929,8 +1151,7 @@ int reshape_open_backup_file(char *backup_file, memset(buf, 0, 512); for (i=0; i < blocks + 8 ; i++) { if (write(*fdlist, buf, 512) != 512) { - pr_err("%s: cannot create" - " backup file %s: %s\n", + pr_err("%s: cannot create backup file %s: %s\n", devname, backup_file, strerror(errno)); return 0; } @@ -943,7 +1164,9 @@ int reshape_open_backup_file(char *backup_file, if (!restart && strncmp(backup_file, MAP_DIR, strlen(MAP_DIR)) != 0) { char *bu = make_backup(sys_name); - symlink(backup_file, bu); + if (symlink(backup_file, bu)) + pr_err("Recording backup file in " MAP_DIR " failed: %s\n", + strerror(errno)); free(bu); } @@ -1029,21 +1252,23 @@ char *analyse_change(char *devname, struct mdinfo *info, struct reshape *re) switch (info->array.level) { default: - return "Cannot understand this RAID level"; + return "No reshape is possibly for this RAID level"; + case LEVEL_LINEAR: + if (info->delta_disks != UnSet) + return "Only --add is supported for LINEAR, setting --raid-disks is not needed"; + else + return "Only --add is supported for LINEAR, other --grow options are not meaningful"; case 1: /* RAID1 can convert to RAID1 with different disks, or * raid5 with 2 disks, or * raid0 with 1 disk */ - if (info->new_level > 1 && - (info->component_size & 7)) - return "Cannot convert RAID1 of this size - " - "reduce size to multiple of 4K first."; + if (info->new_level > 1 && (info->component_size & 7)) + return "Cannot convert RAID1 of this size - reduce size to multiple of 4K first."; if (info->new_level == 0) { if (info->delta_disks != UnSet && info->delta_disks != 0) - return "Cannot change number of disks " - "with RAID1->RAID0 conversion"; + return "Cannot change number of disks with RAID1->RAID0 conversion"; re->level = 0; re->before.data_disks = 1; re->after.data_disks = 1; @@ -1056,9 +1281,9 @@ char *analyse_change(char *devname, struct mdinfo *info, struct reshape *re) re->level = 1; return NULL; } - if (info->array.raid_disks == 2 && - info->new_level == 5) { - + if (info->array.raid_disks != 2 && info->new_level == 5) + return "Can only convert a 2-device array to RAID5"; + if (info->array.raid_disks == 2 && info->new_level == 5) { re->level = 5; re->before.data_disks = 1; if (info->delta_disks != UnSet && @@ -1129,8 +1354,7 @@ char *analyse_change(char *devname, struct mdinfo *info, struct reshape *re) if (far > 1 && !offset) return "Cannot reshape RAID10 to far-mode"; if (near * far != copies) - return "Cannot change number of copies" - " when reshaping RAID10"; + return "Cannot change number of copies when reshaping RAID10"; } if (info->delta_disks == UnSet) info->delta_disks = 0; @@ -1170,7 +1394,8 @@ char *analyse_change(char *devname, struct mdinfo *info, struct reshape *re) case 0: /* RAID0 can be converted to RAID10, or to RAID456 */ if (info->new_level == 10) { - if (info->new_layout == UnSet && info->delta_disks == UnSet) { + if (info->new_layout == UnSet && + info->delta_disks == UnSet) { /* Assume near=2 layout */ info->new_layout = 0x102; info->delta_disks = info->array.raid_disks; @@ -1180,15 +1405,13 @@ char *analyse_change(char *devname, struct mdinfo *info, struct reshape *re) / info->array.raid_disks); if (info->array.raid_disks * (copies-1) != info->delta_disks) - return "Impossible number of devices" - " for RAID0->RAID10"; + return "Impossible number of devices for RAID0->RAID10"; info->new_layout = 0x100 + copies; } if (info->delta_disks == UnSet) { int copies = info->new_layout & 0xff; if (info->new_layout != 0x100 + copies) - return "New layout impossible" - " for RAID0->RAID10";; + return "New layout impossible for RAID0->RAID10";; info->delta_disks = (copies - 1) * info->array.raid_disks; } @@ -1283,8 +1506,7 @@ char *analyse_change(char *devname, struct mdinfo *info, struct reshape *re) return "Can only convert a 2-device array to RAID1"; if (info->delta_disks != UnSet && info->delta_disks != 0) - return "Cannot set raid_disk when " - "converting RAID5->RAID1"; + return "Cannot set raid_disk when converting RAID5->RAID1"; re->level = 1; info->new_chunk = 0; return NULL; @@ -1341,7 +1563,6 @@ char *analyse_change(char *devname, struct mdinfo *info, struct reshape *re) switch (re->level) { case 4: - re->before.layout = 0; re->after.layout = 0; break; case 5: @@ -1380,8 +1601,7 @@ char *analyse_change(char *devname, struct mdinfo *info, struct reshape *re) strcat(strcpy(layout, ls), "-6"); l = map_name(r6layout, layout); if (l == UnSet) - return "Cannot find RAID6 layout" - " to convert to"; + return "Cannot find RAID6 layout to convert to"; } else { /* Current RAID6 has no equivalent. * If it is already a '-6' layout we @@ -1414,16 +1634,19 @@ char *analyse_change(char *devname, struct mdinfo *info, struct reshape *re) if (info->delta_disks == UnSet) info->delta_disks = delta_parity; - re->after.data_disks = (re->before.data_disks - + info->delta_disks - - delta_parity); + re->after.data_disks = + (re->before.data_disks + info->delta_disks - delta_parity); + switch (re->level) { - case 6: re->parity = 2; + case 6: + re->parity = 2; break; case 4: - case 5: re->parity = 1; + case 5: + re->parity = 1; break; - default: re->parity = 0; + default: + re->parity = 0; break; } /* So we have a restripe operation, we need to calculate the number @@ -1477,7 +1700,7 @@ static int set_array_size(struct supertype *st, struct mdinfo *sra, if (text_version == NULL) text_version = sra->text_version; - subarray = strchr(text_version+1, '/')+1; + subarray = strchr(text_version + 1, '/')+1; info = st->ss->container_content(st, subarray); if (info) { unsigned long long current_size = 0; @@ -1493,8 +1716,8 @@ static int set_array_size(struct supertype *st, struct mdinfo *sra, ret_val = 0; dprintf("Array size changed"); } - dprintf(" from %llu to %llu.\n", - current_size, new_size); + dprintf_cont(" from %llu to %llu.\n", + current_size, new_size); } sysfs_free(info); } else @@ -1555,28 +1778,26 @@ int Grow_reshape(char *devname, int fd, struct mdinfo info; struct mdinfo *sra; - if (ioctl(fd, GET_ARRAY_INFO, &array) < 0) { + if (md_get_array_info(fd, &array) < 0) { pr_err("%s is not an active md array - aborting\n", devname); return 1; } - if (data_offset != INVALID_SECTORS && array.level != 10 - && (array.level < 4 || array.level > 6)) { + if (data_offset != INVALID_SECTORS && array.level != 10 && + (array.level < 4 || array.level > 6)) { pr_err("--grow --data-offset not yet supported\n"); return 1; } if (s->size > 0 && (s->chunk || s->level!= UnSet || s->layout_str || s->raiddisks)) { - pr_err("cannot change component size at the same time " - "as other changes.\n" - " Change size first, then check data is intact before " - "making other changes.\n"); + pr_err("cannot change component size at the same time as other changes.\n" + " Change size first, then check data is intact before making other changes.\n"); return 1; } - if (s->raiddisks && s->raiddisks < array.raid_disks && array.level > 1 && - get_linux_version() < 2006032 && + if (s->raiddisks && s->raiddisks < array.raid_disks && + array.level > 1 && get_linux_version() < 2006032 && !check_env("MDADM_FORCE_FEWER")) { pr_err("reducing the number of devices is not safe before Linux 2.6.32\n" " Please use a newer kernel\n"); @@ -1589,17 +1810,25 @@ int Grow_reshape(char *devname, int fd, return 1; } if (s->raiddisks > st->max_devs) { - pr_err("Cannot increase raid-disks on this array" - " beyond %d\n", st->max_devs); + pr_err("Cannot increase raid-disks on this array beyond %d\n", st->max_devs); return 1; } + if (s->level == 0 && + (array.state & (1<ss->external) { - int rv; + int retval; if (subarray) { container = st->container_devnm; @@ -1617,9 +1846,9 @@ int Grow_reshape(char *devname, int fd, return 1; } - rv = st->ss->load_container(st, cfd, NULL); + retval = st->ss->load_container(st, cfd, NULL); - if (rv) { + if (retval) { pr_err("Cannot read superblock for %s\n", devname); free(subarray); @@ -1638,15 +1867,14 @@ int Grow_reshape(char *devname, int fd, /* check if reshape is allowed based on metadata * indications stored in content.array.status */ - if (content->array.state & (1<array.state & + (1 << MD_SB_BLOCK_VOLUME)) allow_reshape = 0; - if (content->array.state - & (1<array.state & + (1 << MD_SB_BLOCK_CONTAINER_RESHAPE)) allow_reshape = 0; if (!allow_reshape) { - pr_err("cannot reshape arrays in" - " container with unsupported" - " metadata: %s(%s)\n", + pr_err("cannot reshape arrays in container with unsupported metadata: %s(%s)\n", devname, container); sysfs_free(cc); free(subarray); @@ -1663,10 +1891,9 @@ int Grow_reshape(char *devname, int fd, for (dv = devlist; dv; dv = dv->next) added_disks++; if (s->raiddisks > array.raid_disks && - array.spare_disks +added_disks < (s->raiddisks - array.raid_disks) && + array.spare_disks + added_disks < (s->raiddisks - array.raid_disks) && !c->force) { - pr_err("Need %d spare%s to avoid degraded array," - " and only have %d.\n" + pr_err("Need %d spare%s to avoid degraded array, and only have %d.\n" " Use --force to over-ride this check.\n", s->raiddisks - array.raid_disks, s->raiddisks - array.raid_disks == 1 ? "" : "s", @@ -1674,8 +1901,8 @@ int Grow_reshape(char *devname, int fd, return 1; } - sra = sysfs_read(fd, NULL, GET_LEVEL | GET_DISKS | GET_DEVS - | GET_STATE | GET_VERSION); + sra = sysfs_read(fd, NULL, GET_LEVEL | GET_DISKS | GET_DEVS | + GET_STATE | GET_VERSION); if (sra) { if (st->ss->external && subarray == NULL) { array.level = LEVEL_CONTAINER; @@ -1692,14 +1919,14 @@ int Grow_reshape(char *devname, int fd, sysfs_free(sra); return 1; } else if (frozen < 0) { - pr_err("%s is performing resync/recovery and cannot" - " be reshaped\n", devname); + pr_err("%s is performing resync/recovery and cannot be reshaped\n", devname); sysfs_free(sra); return 1; } /* ========= set size =============== */ - if (s->size > 0 && (s->size == MAX_SIZE || s->size != (unsigned)array.size)) { + if (s->size > 0 && + (s->size == MAX_SIZE || s->size != (unsigned)array.size)) { unsigned long long orig_size = get_component_size(fd)/2; unsigned long long min_csize; struct mdinfo *mdi; @@ -1715,7 +1942,8 @@ int Grow_reshape(char *devname, int fd, } if (reshape_super(st, s->size, UnSet, UnSet, 0, 0, UnSet, NULL, - devname, APPLY_METADATA_CHANGES, c->verbose > 0)) { + devname, APPLY_METADATA_CHANGES, + c->verbose > 0)) { rv = 1; goto release; } @@ -1734,8 +1962,8 @@ int Grow_reshape(char *devname, int fd, sizeinfo->array.layout, sizeinfo->array.raid_disks); new_size /= data_disks; - dprintf("Metadata size correction from %llu to " - "%llu (%llu)\n", orig_size, new_size, + dprintf("Metadata size correction from %llu to %llu (%llu)\n", + orig_size, new_size, new_size * data_disks); s->size = new_size; sysfs_free(sizeinfo); @@ -1748,15 +1976,7 @@ int Grow_reshape(char *devname, int fd, * understands '0' to mean 'max'. */ min_csize = 0; - rv = 0; for (mdi = sra->devs; mdi; mdi = mdi->next) { - if (sysfs_set_num(sra, mdi, "size", - s->size == MAX_SIZE ? 0 : s->size) < 0) { - /* Probably kernel refusing to let us - * reduce the size - not an error. - */ - break; - } if (array.not_persistent == 0 && array.major_version == 0 && get_linux_version() < 3001000) { @@ -1771,14 +1991,8 @@ int Grow_reshape(char *devname, int fd, } } } - if (rv) { - pr_err("Cannot set size on " - "array members.\n"); - goto size_change_error; - } if (min_csize && s->size > min_csize) { - pr_err("Cannot safely make this array " - "use more than 2TB per device on this kernel.\n"); + pr_err("Cannot safely make this array use more than 2TB per device on this kernel.\n"); rv = 1; goto size_change_error; } @@ -1786,8 +2000,7 @@ int Grow_reshape(char *devname, int fd, /* Don't let the kernel choose a size - it will get * it wrong */ - pr_err("Limited v0.90 array to " - "2TB per device\n"); + pr_err("Limited v0.90 array to 2TB per device\n"); s->size = min_csize; } if (st->ss->external) { @@ -1796,10 +2009,10 @@ int Grow_reshape(char *devname, int fd, "raid5"); if (!rv) { raid0_takeover = 1; - /* get array parametes after takeover - * to chane one parameter at time only + /* get array parameters after takeover + * to change one parameter at time only */ - rv = ioctl(fd, GET_ARRAY_INFO, &array); + rv = md_get_array_info(fd, &array); } } /* make sure mdmon is @@ -1815,7 +2028,7 @@ int Grow_reshape(char *devname, int fd, if (s->size == MAX_SIZE) s->size = 0; array.size = s->size; - if ((unsigned)array.size != s->size) { + if (s->size & ~INT32_MAX) { /* got truncated to 32bit, write to * component_size instead */ @@ -1825,7 +2038,7 @@ int Grow_reshape(char *devname, int fd, else rv = -1; } else { - rv = ioctl(fd, SET_ARRAY_INFO, &array); + rv = md_set_array_info(fd, &array); /* manage array size when it is managed externally */ @@ -1841,7 +2054,7 @@ int Grow_reshape(char *devname, int fd, /* go back to raid0, drop parity disk */ sysfs_set_str(sra, NULL, "level", "raid0"); - ioctl(fd, GET_ARRAY_INFO, &array); + md_get_array_info(fd, &array); } size_change_error: @@ -1870,18 +2083,16 @@ size_change_error: sysfs_set_str(sra, NULL, "resync_start", "none") < 0) pr_err("--assume-clean not supported with --grow on this kernel\n"); } - ioctl(fd, GET_ARRAY_INFO, &array); + md_get_array_info(fd, &array); s->size = get_component_size(fd)/2; if (s->size == 0) s->size = array.size; if (c->verbose >= 0) { if (s->size == orig_size) - pr_err("component size of %s " - "unchanged at %lluK\n", + pr_err("component size of %s unchanged at %lluK\n", devname, s->size); else - pr_err("component size of %s " - "has been set to %lluK\n", + pr_err("component size of %s has been set to %lluK\n", devname, s->size); } changed = 1; @@ -1916,7 +2127,7 @@ size_change_error: int err; err = remove_disks_for_takeover(st, sra, array.layout); if (err) { - dprintf(Name": Array cannot be reshaped\n"); + dprintf("Array cannot be reshaped\n"); if (cfd > -1) close(cfd); rv = 1; @@ -1932,7 +2143,11 @@ size_change_error: memset(&info, 0, sizeof(info)); info.array = array; - sysfs_init(&info, fd, NULL); + if (sysfs_init(&info, fd, NULL)) { + pr_err("failed to intialize sysfs.\n"); + rv = 1; + goto release; + } strcpy(info.text_version, sra->text_version); info.component_size = s->size*2; info.new_level = s->level; @@ -1949,12 +2164,9 @@ size_change_error: if (info.array.level == 6 && (info.new_level == 6 || info.new_level == UnSet) && info.array.layout >= 16) { - pr_err("%s has a non-standard layout. If you" - " wish to preserve this\n", devname); - cont_err("during the reshape, please specify" - " --layout=preserve\n"); - cont_err("If you want to change it, specify a" - " layout or use --layout=normalise\n"); + pr_err("%s has a non-standard layout. If you wish to preserve this\n", devname); + cont_err("during the reshape, please specify --layout=preserve\n"); + cont_err("If you want to change it, specify a layout or use --layout=normalise\n"); rv = 1; goto release; } @@ -1971,8 +2183,7 @@ size_change_error: info.new_layout = map_name(r6layout, l); } } else { - pr_err("%s is only meaningful when reshaping" - " a RAID6 array.\n", s->layout_str); + pr_err("%s is only meaningful when reshaping a RAID6 array.\n", s->layout_str); rv = 1; goto release; } @@ -1994,8 +2205,7 @@ size_change_error: strcat(l, "-6"); info.new_layout = map_name(r6layout, l); } else { - pr_err("%s in only meaningful when reshaping" - " to RAID6\n", s->layout_str); + pr_err("%s in only meaningful when reshaping to RAID6\n", s->layout_str); rv = 1; goto release; } @@ -2017,14 +2227,12 @@ size_change_error: info.new_layout = parse_layout_faulty(s->layout_str); break; default: - pr_err("layout not meaningful" - " with this level\n"); + pr_err("layout not meaningful with this level\n"); rv = 1; goto release; } if (info.new_layout == UnSet) { - pr_err("layout %s not understood" - " for this level\n", + pr_err("layout %s not understood for this level\n", s->layout_str); rv = 1; goto release; @@ -2045,12 +2253,12 @@ size_change_error: rv =1 ; } if (s->layout_str) { - if (ioctl(fd, GET_ARRAY_INFO, &array) != 0) { + if (md_get_array_info(fd, &array) != 0) { dprintf("Cannot get array information.\n"); goto release; } array.layout = info.new_layout; - if (ioctl(fd, SET_ARRAY_INFO, &array) != 0) { + if (md_set_array_info(fd, &array) != 0) { pr_err("failed to set new layout\n"); rv = 1; } else if (c->verbose >= 0) @@ -2066,6 +2274,10 @@ size_change_error: * number of devices (On-Line Capacity Expansion) must be * performed at the level of the container */ + if (fd > 0) { + close(fd); + fd = -1; + } rv = reshape_container(container, devname, -1, st, &info, c->force, c->backup_file, c->verbose, 0, 0, 0); frozen = 0; @@ -2130,27 +2342,22 @@ static int verify_reshape_position(struct mdinfo *info, int level) char *ep; unsigned long long position = strtoull(buf, &ep, 0); - dprintf(Name": Read sync_max sysfs entry is: %s\n", buf); + dprintf("Read sync_max sysfs entry is: %s\n", buf); if (!(ep == buf || (*ep != 0 && *ep != '\n' && *ep != ' '))) { position *= get_data_disks(level, info->new_layout, info->array.raid_disks); if (info->reshape_progress < position) { - dprintf("Corrected reshape progress (%llu) to " - "md position (%llu)\n", + dprintf("Corrected reshape progress (%llu) to md position (%llu)\n", info->reshape_progress, position); info->reshape_progress = position; ret_val = 1; } else if (info->reshape_progress > position) { - pr_err("Fatal error: array " - "reshape was not properly frozen " - "(expected reshape position is %llu, " - "but reshape progress is %llu.\n", + pr_err("Fatal error: array reshape was not properly frozen (expected reshape position is %llu, but reshape progress is %llu.\n", position, info->reshape_progress); ret_val = -1; } else { - dprintf("Reshape position in md and metadata " - "are the same;"); + dprintf("Reshape position in md and metadata are the same;"); ret_val = 1; } } @@ -2252,7 +2459,10 @@ static int set_new_data_offset(struct mdinfo *sra, struct supertype *st, if (info2.space_before == 0 && info2.space_after == 0) { /* Metadata doesn't support data_offset changes */ - return 1; + if (!can_fallback) + pr_err("%s: Metadata version doesn't support data_offset changes\n", + devname); + goto fallback; } if (before > info2.space_before) before = info2.space_before; @@ -2497,8 +2707,7 @@ static int raid10_reshape(char *container, int fd, char *devname, devname); if (err == EBUSY && (info->array.state & (1<cache_size < cache) subarray_set_num(container, sra, "stripe_cache_size", @@ -2607,14 +2816,13 @@ static int impose_reshape(struct mdinfo *sra, * reshape->after.data_disks); } - ioctl(fd, GET_ARRAY_INFO, &array); + md_get_array_info(fd, &array); if (info->array.chunk_size == info->new_chunk && reshape->before.layout == reshape->after.layout && st->ss->external == 0) { /* use SET_ARRAY_INFO but only if reshape hasn't started */ array.raid_disks = reshape->after.data_disks + reshape->parity; - if (!restart && - ioctl(fd, SET_ARRAY_INFO, &array) != 0) { + if (!restart && md_set_array_info(fd, &array) != 0) { int err = errno; pr_err("Cannot set device shape for %s: %s\n", @@ -2622,8 +2830,7 @@ static int impose_reshape(struct mdinfo *sra, if (err == EBUSY && (array.state & (1<= 4 && array.level <= 6)) { /* To convert to RAID0 we need to fail and @@ -2687,44 +2897,37 @@ static int impose_level(int fd, int level, char *devname, int verbose) d++) { mdu_disk_info_t disk; disk.number = d; - if (ioctl(fd, GET_DISK_INFO, &disk) < 0) + if (md_get_disk_info(fd, &disk) < 0) continue; if (disk.major == 0 && disk.minor == 0) continue; found++; - if ((disk.state & (1 << MD_DISK_ACTIVE)) - && disk.raid_disk < data_disks) + if ((disk.state & (1 << MD_DISK_ACTIVE)) && + disk.raid_disk < data_disks) /* keep this */ continue; ioctl(fd, HOT_REMOVE_DISK, makedev(disk.major, disk.minor)); } /* Now fail anything left */ - ioctl(fd, GET_ARRAY_INFO, &array); + md_get_array_info(fd, &array); for (d = 0, found = 0; d < MAX_DISKS && found < array.nr_disks; d++) { - int cnt; mdu_disk_info_t disk; disk.number = d; - if (ioctl(fd, GET_DISK_INFO, &disk) < 0) + if (md_get_disk_info(fd, &disk) < 0) continue; if (disk.major == 0 && disk.minor == 0) continue; found++; - if ((disk.state & (1 << MD_DISK_ACTIVE)) - && disk.raid_disk < data_disks) + if ((disk.state & (1 << MD_DISK_ACTIVE)) && + disk.raid_disk < data_disks) /* keep this */ continue; ioctl(fd, SET_DISK_FAULTY, makedev(disk.major, disk.minor)); - cnt = 5; - while (ioctl(fd, HOT_REMOVE_DISK, - makedev(disk.major, disk.minor)) < 0 - && errno == EBUSY - && cnt--) { - usleep(10000); - } + hot_remove_disk(fd, makedev(disk.major, disk.minor), 1); } } c = map_num(pers, level); @@ -2736,8 +2939,7 @@ static int impose_level(int fd, int level, char *devname, int verbose) devname, c); if (err == EBUSY && (array.state & (1<= 0) @@ -2824,11 +3026,12 @@ static int reshape_array(char *container, int fd, char *devname, unsigned long long array_size; int done; struct mdinfo *sra = NULL; + char buf[20]; /* when reshaping a RAID0, the component_size might be zero. * So try to fix that up. */ - if (ioctl(fd, GET_ARRAY_INFO, &array) != 0) { + if (md_get_array_info(fd, &array) != 0) { dprintf("Cannot get array information.\n"); goto release; } @@ -2866,12 +3069,13 @@ static int reshape_array(char *container, int fd, char *devname, reshape.before.layout != info->array.layout || reshape.before.data_disks + reshape.parity != info->array.raid_disks - max(0, info->delta_disks))) { - pr_err("reshape info is not in native format -" - " cannot continue.\n"); + pr_err("reshape info is not in native format - cannot continue.\n"); goto release; } - if (st->ss->external && restart && (info->reshape_progress == 0)) { + if (st->ss->external && restart && (info->reshape_progress == 0) && + !((sysfs_get_str(info, NULL, "sync_action", buf, sizeof(buf)) > 0) && + (strncmp(buf, "reshape", 7) == 0))) { /* When reshape is restarted from '0', very begin of array * it is possible that for external metadata reshape and array * configuration doesn't happen. @@ -2922,8 +3126,7 @@ static int reshape_array(char *container, int fd, char *devname, if (!force && info->new_level > 1 && info->array.level > 1 && spares_needed > info->array.spare_disks + added_disks) { - pr_err("Need %d spare%s to avoid degraded array," - " and only have %d.\n" + pr_err("Need %d spare%s to avoid degraded array, and only have %d.\n" " Use --force to over-ride this check.\n", spares_needed, spares_needed == 1 ? "" : "s", @@ -2936,8 +3139,7 @@ static int reshape_array(char *container, int fd, char *devname, - array.raid_disks; if ((info->new_level > 1 || info->new_level == 0) && spares_needed > info->array.spare_disks +added_disks) { - pr_err("Need %d spare%s to create working array," - " and only have %d.\n", + pr_err("Need %d spare%s to create working array, and only have %d.\n", spares_needed, spares_needed == 1 ? "" : "s", info->array.spare_disks + added_disks); @@ -2978,7 +3180,12 @@ static int reshape_array(char *container, int fd, char *devname, struct mdinfo *d; if (info2) { - sysfs_init(info2, fd, st->devnm); + if (sysfs_init(info2, fd, st->devnm)) { + pr_err("unable to initialize sysfs for %s", + st->devnm); + free(info2); + goto release; + } /* When increasing number of devices, we need to set * new raid_disks before adding these, or they might * be rejected. @@ -3004,9 +3211,11 @@ static int reshape_array(char *container, int fd, char *devname, * array. Now that the array has been changed to the right * level and frozen, we can safely add them. */ - if (devlist) - Manage_subdevs(devname, fd, devlist, verbose, - 0,NULL, 0); + if (devlist) { + if (Manage_subdevs(devname, fd, devlist, verbose, + 0, NULL, 0)) + goto release; + } if (reshape.backup_blocks == 0 && data_offset != INVALID_SECTORS) reshape.backup_blocks = reshape.before.data_disks * info->array.chunk_size/512; @@ -3015,7 +3224,7 @@ static int reshape_array(char *container, int fd, char *devname, * some more changes: layout, raid_disks, chunk_size */ /* read current array info */ - if (ioctl(fd, GET_ARRAY_INFO, &array) != 0) { + if (md_get_array_info(fd, &array) != 0) { dprintf("Cannot get array information.\n"); goto release; } @@ -3024,7 +3233,7 @@ static int reshape_array(char *container, int fd, char *devname, if (info->new_layout != UnSet && info->new_layout != array.layout) { array.layout = info->new_layout; - if (ioctl(fd, SET_ARRAY_INFO, &array) != 0) { + if (md_set_array_info(fd, &array) != 0) { pr_err("failed to set new layout\n"); goto release; } else if (verbose >= 0) @@ -3035,7 +3244,7 @@ static int reshape_array(char *container, int fd, char *devname, info->delta_disks != 0 && array.raid_disks != (info->array.raid_disks + info->delta_disks)) { array.raid_disks += info->delta_disks; - if (ioctl(fd, SET_ARRAY_INFO, &array) != 0) { + if (md_set_array_info(fd, &array) != 0) { pr_err("failed to set raid disks\n"); goto release; } else if (verbose >= 0) { @@ -3144,8 +3353,19 @@ static int reshape_array(char *container, int fd, char *devname, devname, container, &reshape) < 0) goto release; if (sysfs_set_str(sra, NULL, "sync_action", "reshape") < 0) { - pr_err("Failed to initiate reshape!\n"); - goto release; + struct mdinfo *sd; + if (errno != EINVAL) { + pr_err("Failed to initiate reshape!\n"); + goto release; + } + /* revert data_offset and try the old way */ + for (sd = sra->devs; sd; sd = sd->next) { + sysfs_set_num(sra, sd, "new_offset", + sd->data_offset); + sysfs_set_str(sra, NULL, "reshape_direction", + "forwards"); + } + break; } if (info->new_level == reshape.level) return 0; @@ -3188,12 +3408,10 @@ started: blocks < 16*1024*2) blocks *= 2; } else - pr_err("Need to backup %luK of critical " - "section..\n", blocks/2); + pr_err("Need to backup %luK of critical section..\n", blocks/2); if (blocks >= sra->component_size/2) { - pr_err("%s: Something wrong" - " - reshape aborted\n", + pr_err("%s: Something wrong - reshape aborted\n", devname); goto release; } @@ -3210,7 +3428,7 @@ started: d = reshape_prepare_fdlist(devname, sra, odisks, nrdisks, blocks, backup_file, fdlist, offsets); - if (d < 0) { + if (d < odisks) { goto release; } if ((st->ss->manage_reshape == NULL) || @@ -3222,10 +3440,8 @@ started: devname); pr_err(" Please provide one with \"--backup=...\"\n"); goto release; - } else if (sra->array.spare_disks == 0) { - pr_err("%s: Cannot grow - " - "need a spare or backup-file to backup " - "critical section\n", devname); + } else if (d == odisks) { + pr_err("%s: Cannot grow - need a spare or backup-file to backup critical section\n", devname); goto release; } } else { @@ -3268,8 +3484,7 @@ started: free(fdlist); free(offsets); sysfs_free(sra); - pr_err("Reshape has to be continued from" - " location %llu when root filesystem has been mounted.\n", + pr_err("Reshape has to be continued from location %llu when root filesystem has been mounted.\n", sra->reshape_progress); return 1; } @@ -3366,7 +3581,7 @@ started: bul = make_backup(sra->sys_name); if (bul) { char buf[1024]; - int l = readlink(bul, buf, sizeof(buf)); + int l = readlink(bul, buf, sizeof(buf) - 1); if (l > 0) { buf[l]=0; unlink(buf); @@ -3382,9 +3597,8 @@ started: } if (!st->ss->external && - !(reshape.before.data_disks != reshape.after.data_disks - && info->custom_array_size) && - info->new_level == reshape.level && + !(reshape.before.data_disks != reshape.after.data_disks && + info->custom_array_size) && info->new_level == reshape.level && !forked) { /* no need to wait for the reshape to finish as * there is nothing more to do. @@ -3475,6 +3689,10 @@ int reshape_container(char *container, char *devname, */ ping_monitor(container); + if (!forked && !freeze_reshape && !check_env("MDADM_NO_SYSTEMCTL")) + if (continue_via_systemd(container)) + return 0; + switch (forked ? 0 : fork()) { case -1: /* error */ perror("Cannot fork to complete reshape\n"); @@ -3482,8 +3700,7 @@ int reshape_container(char *container, char *devname, return 1; default: /* parent */ if (!freeze_reshape) - printf(Name ": multi-array reshape continues" - " in background\n"); + printf("%s: multi-array reshape continues in background\n", Name); return 0; case 0: /* child */ map_fork(); @@ -3511,7 +3728,7 @@ int reshape_container(char *container, char *devname, int fd; struct mdstat_ent *mdstat; char *adev; - int devid; + dev_t devid; sysfs_free(cc); @@ -3545,8 +3762,7 @@ int reshape_container(char *container, char *devname, fd = open_dev(mdstat->devnm); if (fd < 0) { - printf(Name ": Device %s cannot be opened for reshape.", - adev); + pr_err("Device %s cannot be opened for reshape.\n", adev); break; } @@ -3561,14 +3777,18 @@ int reshape_container(char *container, char *devname, * This is possibly interim until the behaviour of * reshape_array is resolved(). */ - printf(Name ": Multiple reshape execution detected for " - "device %s.", adev); + printf("%s: Multiple reshape execution detected for device %s.\n", Name, adev); close(fd); break; } strcpy(last_devnm, mdstat->devnm); - sysfs_init(content, fd, mdstat->devnm); + if (sysfs_init(content, fd, mdstat->devnm)) { + pr_err("Unable to initialize sysfs for %s\n", + mdstat->devnm); + rv = 1; + break; + } if (mdmon_running(container)) flush_mdmon(container); @@ -3871,33 +4091,45 @@ int progress_reshape(struct mdinfo *info, struct reshape *reshape, * before setting 'sync_action' to 'idle'. * So we need these extra tests. */ - if (completed == 0 && advancing - && info->reshape_progress > 0) + if (completed == 0 && advancing && + strncmp(action, "idle", 4) == 0 && + info->reshape_progress > 0) break; - if (completed == 0 && !advancing - && info->reshape_progress < (info->component_size - * reshape->after.data_disks)) + if (completed == 0 && !advancing && + strncmp(action, "idle", 4) == 0 && + info->reshape_progress < (info->component_size + * reshape->after.data_disks)) break; sysfs_wait(fd, NULL); if (sysfs_fd_get_ll(fd, &completed) < 0) goto check_progress; } /* Some kernels reset 'sync_completed' to zero, - * we need to have real point we are in md + * we need to have real point we are in md. + * So in that case, read 'reshape_position' from sysfs. */ - if (completed == 0) - completed = max_progress; - - /* some kernels can give an incorrectly high 'completed' number */ - completed /= (info->new_chunk/512); - completed *= (info->new_chunk/512); - /* Convert 'completed' back in to a 'progress' number */ - completed *= reshape->after.data_disks; - if (!advancing) { - completed = info->component_size * reshape->after.data_disks - - completed; + if (completed == 0) { + unsigned long long reshapep; + char action[20]; + if (sysfs_get_str(info, NULL, "sync_action", + action, 20) > 0 && + strncmp(action, "idle", 4) == 0 && + sysfs_get_ll(info, NULL, + "reshape_position", &reshapep) == 0) + *reshape_completed = reshapep; + } else { + /* some kernels can give an incorrectly high + * 'completed' number, so round down */ + completed /= (info->new_chunk/512); + completed *= (info->new_chunk/512); + /* Convert 'completed' back in to a 'progress' number */ + completed *= reshape->after.data_disks; + if (!advancing) + completed = (info->component_size + * reshape->after.data_disks + - completed); + *reshape_completed = completed; } - *reshape_completed = completed; close(fd); @@ -3917,7 +4149,6 @@ check_progress: * it was just a device failure that leaves us degraded but * functioning. */ - strcpy(buf, "hi"); if (sysfs_get_str(info, NULL, "reshape_position", buf, sizeof(buf)) < 0 || strncmp(buf, "none", 4) != 0) { /* The abort might only be temporary. Wait up to 10 @@ -3997,8 +4228,10 @@ static int grow_backup(struct mdinfo *sra, if (sd->disk.state & (1<disk.state & (1<array.utime > (int)__le64_to_cpu(bsb.mtime) + 2*60*60 || - info->array.utime < (int)__le64_to_cpu(bsb.mtime) - 10*60) { + if(time_after(info->array.utime, (unsigned int)__le64_to_cpu(bsb.mtime) + 2*60*60) || + time_before(info->array.utime, (unsigned int)__le64_to_cpu(bsb.mtime) - 10*60)) { if (check_env("MDADM_GROW_ALLOW_OLD")) { - pr_err("accepting backup with timestamp %lu " - "for array with timestamp %lu\n", + pr_err("accepting backup with timestamp %lu for array with timestamp %lu\n", (unsigned long)__le64_to_cpu(bsb.mtime), (unsigned long)info->array.utime); } else { @@ -4554,8 +4786,7 @@ int Grow_restart(struct supertype *st, struct mdinfo *info, int *fdlist, int cnt /* reshape_progress is increasing */ if ((__le64_to_cpu(bsb.arraystart) + __le64_to_cpu(bsb.length) - < info->reshape_progress) - && + < info->reshape_progress) && (__le64_to_cpu(bsb.arraystart2) + __le64_to_cpu(bsb.length2) < info->reshape_progress)) @@ -4599,7 +4830,7 @@ int Grow_restart(struct supertype *st, struct mdinfo *info, int *fdlist, int cnt st->ss->free_super(st); offsets[j] = dinfo.data_offset * 512; } - printf(Name ": restoring critical section\n"); + printf("%s: restoring critical section\n", Name); if (restore_stripes(fdlist, offsets, info->array.raid_disks, @@ -4744,7 +4975,7 @@ int Grow_continue_command(char *devname, int fd, struct mdinfo *cc = NULL; struct mdstat_ent *mdstat = NULL; int cfd = -1; - int fd2 = -1; + int fd2; dprintf("Grow continue from command line called for %s\n", devname); @@ -4758,67 +4989,72 @@ int Grow_continue_command(char *devname, int fd, dprintf("Grow continue is run for "); if (st->ss->external == 0) { int d; - dprintf("native array (%s)\n", devname); - if (ioctl(fd, GET_ARRAY_INFO, &array.array) < 0) { - pr_err("%s is not an active md array -" - " aborting\n", devname); + int cnt = 5; + dprintf_cont("native array (%s)\n", devname); + if (md_get_array_info(fd, &array.array) < 0) { + pr_err("%s is not an active md array - aborting\n", + devname); ret_val = 1; goto Grow_continue_command_exit; } content = &array; + sysfs_init(content, fd, NULL); /* Need to load a superblock. * FIXME we should really get what we need from * sysfs */ - for (d = 0; d < MAX_DISKS; d++) { - mdu_disk_info_t disk; - char *dv; - int err; - disk.number = d; - if (ioctl(fd, GET_DISK_INFO, &disk) < 0) - continue; - if (disk.major == 0 && disk.minor == 0) - continue; - if ((disk.state & (1 << MD_DISK_ACTIVE)) == 0) - continue; - dv = map_dev(disk.major, disk.minor, 1); - if (!dv) - continue; - fd2 = dev_open(dv, O_RDONLY); - if (fd2 < 0) - continue; - err = st->ss->load_super(st, fd2, NULL); - close(fd2); - /* invalidate fd2 to avoid possible double close() */ - fd2 = -1; - if (err) - continue; - break; - } - if (d == MAX_DISKS) { - pr_err("Unable to load metadata for %s\n", - devname); - ret_val = 1; - goto Grow_continue_command_exit; - } - st->ss->getinfo_super(st, content, NULL); + do { + for (d = 0; d < MAX_DISKS; d++) { + mdu_disk_info_t disk; + char *dv; + int err; + disk.number = d; + if (md_get_disk_info(fd, &disk) < 0) + continue; + if (disk.major == 0 && disk.minor == 0) + continue; + if ((disk.state & (1 << MD_DISK_ACTIVE)) == 0) + continue; + dv = map_dev(disk.major, disk.minor, 1); + if (!dv) + continue; + fd2 = dev_open(dv, O_RDONLY); + if (fd2 < 0) + continue; + err = st->ss->load_super(st, fd2, NULL); + close(fd2); + if (err) + continue; + break; + } + if (d == MAX_DISKS) { + pr_err("Unable to load metadata for %s\n", + devname); + ret_val = 1; + goto Grow_continue_command_exit; + } + st->ss->getinfo_super(st, content, NULL); + if (!content->reshape_active) + sleep(3); + else + break; + } while (cnt-- > 0); } else { char *container; if (subarray) { - dprintf("subarray (%s)\n", subarray); + dprintf_cont("subarray (%s)\n", subarray); container = st->container_devnm; cfd = open_dev_excl(st->container_devnm); } else { container = st->devnm; close(fd); cfd = open_dev_excl(st->devnm); - dprintf("container (%s)\n", container); + dprintf_cont("container (%s)\n", container); fd = cfd; } if (cfd < 0) { - pr_err("Unable to open container " - "for %s\n", devname); + pr_err("Unable to open container for %s\n", devname); ret_val = 1; goto Grow_continue_command_exit; } @@ -4835,7 +5071,7 @@ int Grow_continue_command(char *devname, int fd, cc = st->ss->container_content(st, subarray); for (content = cc; content ; content = content->next) { - char *array; + char *array_name; int allow_reshape = 1; if (content->reshape_active == 0) @@ -4854,16 +5090,14 @@ int Grow_continue_command(char *devname, int fd, allow_reshape = 0; if (!allow_reshape) { - pr_err("cannot continue reshape of an array" - " in container with unsupported" - " metadata: %s(%s)\n", + pr_err("cannot continue reshape of an array in container with unsupported metadata: %s(%s)\n", devname, container); ret_val = 1; goto Grow_continue_command_exit; } - array = strchr(content->text_version+1, '/')+1; - mdstat = mdstat_by_subdev(array, container); + array_name = strchr(content->text_version+1, '/')+1; + mdstat = mdstat_by_subdev(array_name, container); if (!mdstat) continue; if (mdstat->active == 0) { @@ -4876,8 +5110,7 @@ int Grow_continue_command(char *devname, int fd, break; } if (!content) { - pr_err("Unable to determine reshaped " - "array for %s\n", devname); + pr_err("Unable to determine reshaped array for %s\n", devname); ret_val = 1; goto Grow_continue_command_exit; } @@ -4888,7 +5121,15 @@ int Grow_continue_command(char *devname, int fd, goto Grow_continue_command_exit; } - sysfs_init(content, fd2, mdstat->devnm); + if (sysfs_init(content, fd2, mdstat->devnm)) { + pr_err("Unable to initialize sysfs for %s, Grow cannot continue", + mdstat->devnm); + ret_val = 1; + close(fd2); + goto Grow_continue_command_exit; + } + + close(fd2); /* start mdmon in case it is not running */ @@ -4899,8 +5140,7 @@ int Grow_continue_command(char *devname, int fd, if (mdmon_running(container)) st->update_tail = &st->updates; else { - pr_err("No mdmon found. " - "Grow cannot continue.\n"); + pr_err("No mdmon found. Grow cannot continue.\n"); ret_val = 1; goto Grow_continue_command_exit; } @@ -4916,11 +5156,9 @@ int Grow_continue_command(char *devname, int fd, /* continue reshape */ - ret_val = Grow_continue(fd, st, content, backup_file, 0); + ret_val = Grow_continue(fd, st, content, backup_file, 1, 0); Grow_continue_command_exit: - if (fd2 > -1) - close(fd2); if (cfd > -1) close(cfd); st->ss->free_super(st); @@ -4932,7 +5170,7 @@ Grow_continue_command_exit: } int Grow_continue(int mdfd, struct supertype *st, struct mdinfo *info, - char *backup_file, int freeze_reshape) + char *backup_file, int forked, int freeze_reshape) { int ret_val = 2; @@ -4949,13 +5187,13 @@ int Grow_continue(int mdfd, struct supertype *st, struct mdinfo *info, close(cfd); ret_val = reshape_container(st->container_devnm, NULL, mdfd, st, info, 0, backup_file, - 0, 1, + 0, forked, 1 | info->reshape_active, freeze_reshape); } else ret_val = reshape_array(NULL, mdfd, "array", st, info, 1, NULL, INVALID_SECTORS, - backup_file, 0, 1, + backup_file, 0, forked, 1 | info->reshape_active, freeze_reshape);