/*
* mdadm - manage Linux "md" devices aka RAID arrays.
*
- * Copyright (C) 2001-2013 Neil Brown <neilb@suse.de>
+ * Copyright (C) 2001-2016 Neil Brown <neilb@suse.com>
*
*
* This program is free software; you can redistribute it and/or modify
struct mdinfo *content = NULL;
int report_mismatch = ((inargv && c->verbose >= 0) || c->verbose > 0);
struct domainlist *domains = NULL;
+ dev_t rdev;
tmpdev = devlist; num_devs = 0;
while (tmpdev) {
tmpdev = tmpdev ? tmpdev->next : NULL) {
char *devname = tmpdev->devname;
int dfd;
- struct stat stb;
struct supertype *tst;
struct dev_policy *pol = NULL;
int found_container = 0;
pr_err("cannot open device %s: %s\n",
devname, strerror(errno));
tmpdev->used = 2;
- } else if (fstat(dfd, &stb)< 0) {
- /* Impossible! */
- pr_err("fstat failed for %s: %s\n",
- devname, strerror(errno));
- tmpdev->used = 2;
- } else if ((stb.st_mode & S_IFMT) != S_IFBLK) {
- pr_err("%s is not a block device.\n",
- devname);
+ } else if (!fstat_is_blkdev(dfd, devname, &rdev)) {
tmpdev->used = 2;
} else if (must_be_container(dfd)) {
if (st) {
pr_err("%s is a container, but we are looking for components\n",
devname);
tmpdev->used = 2;
-#if !defined(MDASSEMBLE) || defined(MDASSEMBLE) && defined(MDASSEMBLE_AUTO)
} if (!tst && (tst = super_by_fd(dfd, NULL)) == NULL) {
if (report_mismatch)
pr_err("not a recognisable container: %s\n",
devname);
tmpdev->used = 2;
-#endif
} else if (!tst->ss->load_container
|| tst->ss->load_container(tst, dfd, NULL)) {
if (report_mismatch)
devname);
tmpdev->used = 2;
} else if (auto_assem &&
- !conf_test_metadata(tst->ss->name, (pol = devid_policy(stb.st_rdev)),
+ !conf_test_metadata(tst->ss->name,
+ (pol = devid_policy(rdev)),
tst->ss->match_home(tst, c->homehost) == 1)) {
if (report_mismatch)
pr_err("%s has metadata type %s for which auto-assembly is disabled\n",
tst->ss->name, devname);
tmpdev->used = 2;
} else if (auto_assem && st == NULL &&
- !conf_test_metadata(tst->ss->name, (pol = devid_policy(stb.st_rdev)),
+ !conf_test_metadata(tst->ss->name,
+ (pol = devid_policy(rdev)),
tst->ss->match_home(tst, c->homehost) == 1)) {
if (report_mismatch)
pr_err("%s has metadata type %s for which auto-assembly is disabled\n",
/* Collect domain information from members only */
if (tmpdev && tmpdev->used == 1) {
if (!pol)
- pol = devid_policy(stb.st_rdev);
+ pol = devid_policy(rdev);
domain_merge(&domains, pol, tst?tst->ss->name:NULL);
}
dev_policy_free(pol);
/* Now reject spares that don't match domains of identified members */
for (tmpdev = devlist; tmpdev; tmpdev = tmpdev->next) {
- struct stat stb;
if (tmpdev->used != 3)
continue;
- if (stat(tmpdev->devname, &stb)< 0) {
- pr_err("fstat failed for %s: %s\n",
- tmpdev->devname, strerror(errno));
+ if (!stat_is_blkdev(tmpdev->devname, &rdev)) {
tmpdev->used = 2;
} else {
- struct dev_policy *pol = devid_policy(stb.st_rdev);
+ struct dev_policy *pol = devid_policy(rdev);
int dt = domain_test(domains, pol, NULL);
if (inargv && dt != 0)
/* take this spare as domains match
struct mddev_dev *tmpdev;
int devcnt = 0;
int nextspare = 0;
-#ifndef MDASSEMBLE
int bitmap_done = 0;
-#endif
int most_recent = -1;
int bestcnt = 0;
int *best = *bestp;
if (tmpdev->used != 1)
continue;
/* looks like a good enough match to update the super block if needed */
-#ifndef MDASSEMBLE
if (c->update) {
/* prepare useful information in info structures */
struct stat stb2;
int err;
fstat(mdfd, &stb2);
- if (strcmp(c->update, "uuid")==0 &&
- !ident->uuid_set) {
- int rfd;
- if ((rfd = open("/dev/urandom", O_RDONLY)) < 0 ||
- read(rfd, ident->uuid, 16) != 16) {
- *(__u32*)(ident->uuid) = random();
- *(__u32*)(ident->uuid+1) = random();
- *(__u32*)(ident->uuid+2) = random();
- *(__u32*)(ident->uuid+3) = random();
- }
- if (rfd >= 0) close(rfd);
+ if (strcmp(c->update, "uuid") == 0 && !ident->uuid_set)
+ random_uuid((__u8 *)ident->uuid);
+
+ if (strcmp(c->update, "ppl") == 0 &&
+ ident->bitmap_fd >= 0) {
+ pr_err("PPL is not compatible with bitmap\n");
+ return -1;
}
+
dfd = dev_open(devname,
tmpdev->disposition == 'I'
? O_RDWR : (O_RDWR|O_EXCL));
} else if (strcmp(c->update, "nodes") == 0) {
tst->nodes = c->nodes;
err = tst->ss->write_bitmap(tst, dfd, NodeNumUpdate);
- } else
+ } else if (strcmp(c->update, "revert-reshape") == 0 &&
+ c->invalid_backup)
+ err = tst->ss->update_super(tst, content,
+ "revert-reshape-nobackup",
+ devname, c->verbose,
+ ident->uuid_set,
+ c->homehost);
+ else
err = tst->ss->update_super(tst, content, c->update,
devname, c->verbose,
ident->uuid_set,
else
bitmap_done = 1;
}
- } else
-#endif
- {
+ } else {
dfd = dev_open(devname,
tmpdev->disposition == 'I'
? O_RDWR : (O_RDWR|O_EXCL));
c->readonly = 1;
}
+ if (content->consistency_policy == CONSISTENCY_POLICY_PPL)
+ clean = 1;
+
rv = set_array_info(mdfd, st, content);
if (rv && !err_ok) {
pr_err("failed to set array info for %s: %s\n",
}
/* First, add the raid disks, but add the chosen one last */
- for (i=0; i<= bestcnt; i++) {
+ for (i = 0; i <= bestcnt; i++) {
int j;
if (i < bestcnt) {
j = best[i];
j = chosen_drive;
if (j >= 0 && !devices[j].included) {
- int dfd = dev_open(devices[j].devname,
- O_RDWR|O_EXCL);
+ int dfd;
+
+ dfd = dev_open(devices[j].devname, O_RDWR|O_EXCL);
if (dfd >= 0) {
remove_partitions(dfd);
close(dfd);
if (rv) {
pr_err("failed to add %s to %s: %s\n",
- devices[j].devname,
- mddev,
+ devices[j].devname, mddev,
strerror(errno));
- if (i < content->array.raid_disks * 2
- || i == bestcnt)
+ if (i < content->array.raid_disks * 2 ||
+ i == bestcnt)
okcnt--;
else
sparecnt--;
- } else if (c->verbose > 0)
+ } else if (c->verbose > 0) {
pr_err("added %s to %s as %d%s%s\n",
devices[j].devname, mddev,
devices[j].i.disk.raid_disk,
devices[j].uptodate?"":
" (possibly out of date)",
- (devices[j].i.disk.state & (1<<MD_DISK_REPLACEMENT))?" replacement":"");
+ (devices[j].i.disk.state &
+ (1<<MD_DISK_REPLACEMENT)) ?
+ " replacement":"");
+ }
} else if (j >= 0) {
if (c->verbose > 0)
pr_err("%s is already in %s as %d\n",
devices[j].devname, mddev,
devices[j].i.disk.raid_disk);
- } else if (c->verbose > 0 && i < content->array.raid_disks*2
- && (i&1) == 0)
+ } else if (c->verbose > 0 &&
+ i < content->array.raid_disks * 2 && (i & 1) == 0)
pr_err("no uptodate device for slot %d of %s\n",
i/2, mddev);
}
if (content->array.level == LEVEL_CONTAINER) {
if (c->verbose >= 0) {
pr_err("Container %s has been assembled with %d drive%s",
- mddev, okcnt+sparecnt+journalcnt,
- okcnt+sparecnt+journalcnt==1?"":"s");
+ mddev, okcnt + sparecnt + journalcnt,
+ okcnt + sparecnt + journalcnt == 1 ? "" : "s");
if (okcnt < (unsigned)content->array.raid_disks)
fprintf(stderr, " (out of %d)",
content->array.raid_disks);
if (st->ss->validate_container) {
struct mdinfo *devices_list;
- struct mdinfo *info_devices = xmalloc(sizeof(struct mdinfo)*(okcnt+sparecnt));
+ struct mdinfo *info_devices;
unsigned int count;
+
devices_list = NULL;
- for (count = 0; count < okcnt+sparecnt; count++) {
+ info_devices = xmalloc(sizeof(struct mdinfo) *
+ (okcnt + sparecnt));
+ for (count = 0; count < okcnt + sparecnt; count++) {
info_devices[count] = devices[count].i;
info_devices[count].next = devices_list;
devices_list = &info_devices[count];
if (c->runstop == 1 ||
(c->runstop <= 0 &&
- ( enough(content->array.level, content->array.raid_disks,
- content->array.layout, clean, avail) &&
- (okcnt + rebuilding_cnt >= req_cnt || start_partial_ok)
- ))) {
+ (enough(content->array.level, content->array.raid_disks,
+ content->array.layout, clean, avail) &&
+ (okcnt + rebuilding_cnt >= req_cnt || start_partial_ok)))) {
/* This array is good-to-go.
* If a reshape is in progress then we might need to
* continue monitoring it. In that case we start
* it read-only and let the grow code make it writable.
*/
int rv;
-#ifndef MDASSEMBLE
+
if (content->reshape_active &&
!(content->reshape_active & RESHAPE_NO_BACKUP) &&
content->delta_disks <= 0) {
c->backup_file, 0,
c->freeze_reshape);
} else if (c->readonly &&
- sysfs_attribute_available(
- content, NULL, "array_state")) {
+ sysfs_attribute_available(content, NULL,
+ "array_state")) {
rv = sysfs_set_str(content, NULL,
"array_state", "readonly");
} else
-#endif
rv = ioctl(mdfd, RUN_ARRAY, NULL);
reopen_mddev(mdfd); /* drop O_EXCL */
if (rv == 0) {
pr_err("%s has been started with %d drive%s",
mddev, okcnt, okcnt==1?"":"s");
if (okcnt < (unsigned)content->array.raid_disks)
- fprintf(stderr, " (out of %d)", content->array.raid_disks);
+ fprintf(stderr, " (out of %d)",
+ content->array.raid_disks);
if (rebuilding_cnt)
- fprintf(stderr, "%s %d rebuilding", sparecnt?",":" and", rebuilding_cnt);
+ fprintf(stderr, "%s %d rebuilding",
+ sparecnt?",":" and",
+ rebuilding_cnt);
if (sparecnt)
- fprintf(stderr, " and %d spare%s", sparecnt, sparecnt==1?"":"s");
+ fprintf(stderr, " and %d spare%s",
+ sparecnt,
+ sparecnt == 1 ? "" : "s");
if (content->journal_clean)
- fprintf(stderr, " and %d journal", journalcnt);
+ fprintf(stderr, " and %d journal",
+ journalcnt);
fprintf(stderr, ".\n");
}
if (content->reshape_active &&
* of the stripe cache - default is 256
*/
int chunk_size = content->array.chunk_size;
+
if (content->reshape_active &&
content->new_chunk > chunk_size)
chunk_size = content->new_chunk;
if (256 < 4 * ((chunk_size+4065)/4096)) {
- struct mdinfo *sra = sysfs_read(mdfd, NULL, 0);
+ struct mdinfo *sra;
+
+ 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);
+ struct mdinfo *sra;
+
+ sra = sysfs_read(mdfd, NULL, 0);
if (sra)
sysfs_set_str(sra, NULL,
"sync_action", "repair");
}
return 0;
}
- pr_err("failed to RUN_ARRAY %s: %s\n",
- mddev, strerror(errno));
+ pr_err("failed to RUN_ARRAY %s: %s\n", mddev, strerror(errno));
if (!enough(content->array.level, content->array.raid_disks,
content->array.layout, 1, avail))
pr_err("Not enough devices to start the array.\n");
else if (!enough(content->array.level,
content->array.raid_disks,
- content->array.layout, clean,
- avail))
+ content->array.layout, clean, avail))
pr_err("Not enough devices to start the array while not clean - consider --force.\n");
return 1;
}
if (c->runstop == -1) {
pr_err("%s assembled from %d drive%s",
- mddev, okcnt, okcnt==1?"":"s");
+ mddev, okcnt, okcnt == 1 ? "" : "s");
if (okcnt != (unsigned)content->array.raid_disks)
- fprintf(stderr, " (out of %d)", content->array.raid_disks);
+ fprintf(stderr, " (out of %d)",
+ content->array.raid_disks);
fprintf(stderr, ", but not started.\n");
return 2;
}
if (c->verbose >= -1) {
- pr_err("%s assembled from %d drive%s", mddev, okcnt, okcnt==1?"":"s");
+ pr_err("%s assembled from %d drive%s",
+ mddev, okcnt, okcnt == 1 ? "" : "s");
if (rebuilding_cnt)
- fprintf(stderr, "%s %d rebuilding", sparecnt?",":" and", rebuilding_cnt);
+ fprintf(stderr, "%s %d rebuilding",
+ sparecnt ? "," : " and", rebuilding_cnt);
if (sparecnt)
- fprintf(stderr, " and %d spare%s", sparecnt, sparecnt==1?"":"s");
+ fprintf(stderr, " and %d spare%s", sparecnt,
+ sparecnt == 1 ? "" : "s");
if (!enough(content->array.level, content->array.raid_disks,
content->array.layout, 1, avail))
fprintf(stderr, " - not enough to start the array.\n");
else if (!enough(content->array.level,
content->array.raid_disks,
- content->array.layout, clean,
- avail))
+ content->array.layout, clean, avail))
fprintf(stderr, " - not enough to start the array while not clean - consider --force.\n");
else {
if (req_cnt == (unsigned)content->array.raid_disks)
- fprintf(stderr, " - need all %d to start it", req_cnt);
+ fprintf(stderr, " - need all %d to start it",
+ req_cnt);
else
fprintf(stderr, " - need %d to start", req_cnt);
fprintf(stderr, " (use --run to insist).\n");
name = strchr(name, ':')+1;
mdfd = create_mddev(mddev, name, ident->autof, trustworthy,
- chosen_name);
+ chosen_name, 0);
}
if (mdfd < 0) {
st->ss->free_super(st);
return 1;
}
mddev = chosen_name;
- if (get_linux_version() < 2004000 ||
- md_get_version(mdfd) < 9000) {
- pr_err("Assemble requires Linux 2.4 or later, and\n"
- " md driver version 0.90.0 or later.\n"
- " Upgrade your kernel or try --build\n");
- close(mdfd);
- return 1;
- }
if (pre_exist == NULL) {
if (mddev_busy(fd2devnm(mdfd))) {
pr_err("%s already active, cannot restart it!\n",
ioctl(mdfd, STOP_ARRAY, NULL);
}
-#ifndef MDASSEMBLE
if (content != &info) {
/* This is a member of a container. Try starting the array. */
int err;
close(mdfd);
return err;
}
-#endif
+
/* Ok, no bad inconsistancy, we can try updating etc */
devices = xcalloc(num_devs, sizeof(*devices));
devmap = xcalloc(num_devs, content->array.raid_disks);
return 1;
}
st->ss->getinfo_super(st, content, NULL);
-#ifndef MDASSEMBLE
- sysfs_init(content, mdfd, NULL);
-#endif
+ if (sysfs_init(content, mdfd, NULL)) {
+ pr_err("Unable to initialize sysfs\n");
+ close(mdfd);
+ free(devices);
+ return 1;
+ }
+
/* after reload context, store journal_clean in context */
content->journal_clean = journal_clean;
for (i=0; i<bestcnt; i++) {
* that was moved aside due to the reshape overwriting live data
* The code of doing this lives in Grow.c
*/
-#ifndef MDASSEMBLE
if (content->reshape_active &&
!(content->reshape_active & RESHAPE_NO_BACKUP)) {
int err = 0;
return err;
}
}
-#endif
/* Almost ready to actually *do* something */
/* First, fill in the map, so that udev can find our name
return rv == 2 ? 0 : rv;
}
-#ifndef MDASSEMBLE
int assemble_container_content(struct supertype *st, int mdfd,
struct mdinfo *content, struct context *c,
char *chosen_name, int *result)
struct mdinfo *dev, *sra, *dev2;
int working = 0, preexist = 0;
int expansion = 0;
- struct map_ent *map = NULL;
int old_raid_disks;
int start_reshape;
- char *avail = NULL;
+ char *avail;
int err;
- sysfs_init(content, mdfd, NULL);
+ if (sysfs_init(content, mdfd, NULL)) {
+ pr_err("Unable to initialize sysfs\n");
+ return 1;
+ }
sra = sysfs_read(mdfd, NULL, GET_VERSION|GET_DEVS);
if (sra == NULL || strcmp(sra->text_version, content->text_version) != 0) {
c->readonly &&
content->text_version[0] == '/')
content->text_version[0] = '-';
- if (sysfs_set_array(content, md_get_version(mdfd)) != 0) {
- if (sra)
- sysfs_free(sra);
+ if (sysfs_set_array(content, 9003) != 0) {
+ sysfs_free(sra);
return 1;
}
}
free(avail);
return 1;/* Nothing new, don't try to start */
}
- map_update(&map, fd2devnm(mdfd),
- content->text_version,
+ map_update(NULL, fd2devnm(mdfd), content->text_version,
content->uuid, chosen_name);
+ if (content->consistency_policy == CONSISTENCY_POLICY_PPL &&
+ st->ss->validate_ppl) {
+ content->array.state |= 1;
+ err = 0;
+
+ for (dev = content->devs; dev; dev = dev->next) {
+ int dfd;
+ char *devpath;
+ int ret;
+
+ ret = st->ss->validate_ppl(st, content, dev);
+ if (ret == 0)
+ continue;
+
+ if (ret < 0) {
+ err = 1;
+ break;
+ }
+
+ if (!c->force) {
+ pr_err("%s contains invalid PPL - consider --force or --update-subarray with --update=no-ppl\n",
+ chosen_name);
+ content->array.state &= ~1;
+ avail[dev->disk.raid_disk] = 0;
+ break;
+ }
+
+ /* have --force - overwrite the invalid ppl */
+ devpath = map_dev(dev->disk.major, dev->disk.minor, 0);
+ dfd = dev_open(devpath, O_RDWR);
+ if (dfd < 0) {
+ pr_err("Failed to open %s\n", devpath);
+ err = 1;
+ break;
+ }
+
+ err = st->ss->write_init_ppl(st, content, dfd);
+ close(dfd);
+
+ if (err)
+ break;
+ }
+
+ if (err) {
+ free(avail);
+ return err;
+ }
+ }
if (enough(content->array.level, content->array.raid_disks,
content->array.layout, content->array.state & 1, avail) == 0) {
return err;
/* FIXME should have an O_EXCL and wait for read-auto */
}
-#endif