int active, working, failed, spare, raid;
int expected_spares;
int devstate[MaxDisks];
- unsigned devid[MaxDisks];
+ dev_t devid[MaxDisks];
int percent;
int parent_dev; /* For subarray, devnum of parent.
* For others, NoMdDev
static int check_array(struct state *st, struct mdstat_ent *mdstat,
int test, struct alert_info *info,
int increments);
-static int add_new_arrays(struct mdstat_ent *mdstat, struct state *statelist,
+static int add_new_arrays(struct mdstat_ent *mdstat, struct state **statelist,
int test, struct alert_info *info);
static void try_spare_migration(struct state *statelist, struct alert_info *info);
static void link_containers_with_subarrays(struct state *list);
info.mailfrom = mailfrom;
info.dosyslog = dosyslog;
- if (daemonise)
- if (make_daemon(pidfile))
- return 1;
+ if (daemonise) {
+ int rv = make_daemon(pidfile);
+ if (rv >= 0)
+ return rv;
+ }
if (share)
if (check_one_sharer(scan))
/* now check if there are any new devices found in mdstat */
if (scan)
- new_found = add_new_arrays(mdstat, statelist, test,
+ new_found = add_new_arrays(mdstat, &statelist, test,
&info);
/* If an array has active < raid && spare == 0 && spare_group != NULL
static int make_daemon(char *pidfile)
{
+ /* Return:
+ * -1 in the forked daemon
+ * 0 in the parent
+ * 1 on error
+ * so a none-negative becomes the exit code.
+ */
int pid = fork();
if (pid > 0) {
if (!pidfile)
dup2(0,1);
dup2(0,2);
setsid();
- return 0;
+ return -1;
}
static int check_one_sharer(int scan)
struct stat buf;
fp = fopen("/var/run/mdadm/autorebuild.pid", "r");
if (fp) {
- fscanf(fp, "%d", &pid);
+ if (fscanf(fp, "%d", &pid) != 1)
+ pid = -1;
sprintf(dir, "/proc/%d", pid);
rv = stat(dir, &buf);
if (rv != -1) {
} else {
fprintf(stderr, Name ": Warning: One"
" autorebuild process already"
- " running.");
+ " running.\n");
}
}
fclose(fp);
}
if (scan) {
- fp = fopen("/var/run/mdadm/autorebuild.pid", "w");
- if (!fp)
- fprintf(stderr, Name ": Cannot create"
- " autorebuild.pid "
- "file\n");
- else {
- pid = getpid();
- fprintf(fp, "%d\n", pid);
- fclose(fp);
+ if (mkdir("/var/run/mdadm", S_IRWXU) < 0 &&
+ errno != EEXIST) {
+ fprintf(stderr, Name ": Can't create "
+ "autorebuild.pid file\n");
+ } else {
+ fp = fopen("/var/run/mdadm/autorebuild.pid", "w");
+ if (!fp)
+ fprintf(stderr, Name ": Cannot create"
+ " autorebuild.pid"
+ "file\n");
+ else {
+ pid = getpid();
+ fprintf(fp, "%d\n", pid);
+ fclose(fp);
+ }
}
}
return 0;
int test, struct alert_info *ainfo,
int increments)
{
+ /* Update the state 'st' to reflect any changes shown in mdstat,
+ * or found by directly examining the array, and return
+ * '1' if the array is degraded, or '0' if it is optimal (or dead).
+ */
struct { int state, major, minor; } info[MaxDisks];
mdu_array_info_t array;
struct mdstat_ent *mse = NULL, *mse2;
))) {
close(fd);
st->err = 0;
- return 0;
+ if ((st->active < st->raid) && st->spare == 0)
+ return 1;
+ else
+ return 0;
}
if (st->utime == 0 && /* new array */
mse->pattern && strchr(mse->pattern, '_') /* degraded */
} else if (mse && mse->pattern && i < (int)strlen(mse->pattern)) {
switch(mse->pattern[i]) {
case 'U': newstate = 6 /* ACTIVE/SYNC */; break;
- case '_': newstate = 0; break;
+ case '_': newstate = 8 /* REMOVED */; break;
}
disc.major = disc.minor = 0;
}
return 0;
}
-static int add_new_arrays(struct mdstat_ent *mdstat, struct state *statelist,
+static int add_new_arrays(struct mdstat_ent *mdstat, struct state **statelist,
int test, struct alert_info *info)
{
struct mdstat_ent *mse;
continue;
}
close(fd);
- st->next = statelist;
+ st->next = *statelist;
st->err = 1;
st->devnum = mse->devnum;
st->percent = -2;
devname2devnum(mse->metadata_version+10);
else
st->parent_dev = NoMdDev;
- statelist = st;
+ *statelist = st;
if (test)
alert("TestMessage", st->devname, NULL, info);
alert("NewArray", st->devname, NULL, info);
return new_found;
}
-static int move_spare(struct state *st2, struct state *st,
- struct alert_info *info)
+static int get_min_spare_size_required(struct state *st, unsigned long long *sizep)
{
- struct mddev_dev devlist;
- char devname[20];
+ int fd;
- /* try to remove and add */
- int fd1 = open(st->devname, O_RDONLY);
- int fd2 = open(st2->devname, O_RDONLY);
- int dev = -1;
- int d;
- if (fd1 < 0 || fd2 < 0) {
- if (fd1>=0) close(fd1);
- if (fd2>=0) close(fd2);
+ if (!st->metadata ||
+ !st->metadata->ss->min_acceptable_spare_size) {
+ *sizep = 0;
return 0;
}
- for (d=st2->raid; d < MaxDisks; d++) {
- if (st2->devid[d] > 0 &&
- st2->devstate[d] == 0) {
- dev = st2->devid[d];
- break;
+
+ fd = open(st->devname, O_RDONLY);
+ if (fd < 0)
+ return 1;
+ if (st->metadata->ss->external)
+ st->metadata->ss->load_container(st->metadata, fd, st->devname);
+ else
+ st->metadata->ss->load_super(st->metadata, fd, st->devname);
+ close(fd);
+ if (!st->metadata->sb)
+ return 1;
+ *sizep = st->metadata->ss->min_acceptable_spare_size(st->metadata);
+ st->metadata->ss->free_super(st->metadata);
+
+ return 0;
+}
+
+static int check_donor(struct state *from, struct state *to)
+{
+ struct state *sub;
+
+ if (from == to)
+ return 0;
+ if (from->parent)
+ /* Cannot move from a member */
+ return 0;
+ if (from->err)
+ return 0;
+ for (sub = from->subarray; sub; sub = sub->subarray)
+ /* If source array has degraded subarrays, don't
+ * remove anything
+ */
+ if (sub->active < sub->raid)
+ return 0;
+ if (from->metadata->ss->external == 0)
+ if (from->active < from->raid)
+ return 0;
+ if (from->spare <= 0)
+ return 0;
+ return 1;
+}
+
+static dev_t choose_spare(struct state *from, struct state *to,
+ struct domainlist *domlist, unsigned long long min_size)
+{
+ int d;
+ dev_t dev = 0;
+
+ for (d = from->raid; !dev && d < MaxDisks; d++) {
+ if (from->devid[d] > 0 &&
+ from->devstate[d] == 0) {
+ struct dev_policy *pol;
+ unsigned long long dev_size;
+
+ if (to->metadata->ss->external &&
+ test_partition_from_id(from->devid[d]))
+ continue;
+
+ if (min_size &&
+ dev_size_from_id(from->devid[d], &dev_size) &&
+ dev_size < min_size)
+ continue;
+
+ pol = devnum_policy(from->devid[d]);
+ if (from->spare_group)
+ pol_add(&pol, pol_domain,
+ from->spare_group, NULL);
+ if (domain_test(domlist, pol, to->metadata->ss->name) == 1)
+ dev = from->devid[d];
+ dev_policy_free(pol);
}
}
- if (dev < 0) {
- close(fd1);
- close(fd2);
+ return dev;
+}
+
+static dev_t container_choose_spare(struct state *from, struct state *to,
+ struct domainlist *domlist,
+ unsigned long long min_size, int active)
+{
+ /* This is similar to choose_spare, but we cannot trust devstate,
+ * so we need to read the metadata instead
+ */
+ struct mdinfo *list;
+ struct supertype *st = from->metadata;
+ int fd = open(from->devname, O_RDONLY);
+ int err;
+ dev_t dev = 0;
+
+ if (fd < 0)
+ return 0;
+ if (!st->ss->getinfo_super_disks) {
+ close(fd);
return 0;
}
-
- devlist.next = NULL;
- devlist.used = 0;
- devlist.re_add = 0;
- devlist.writemostly = 0;
- devlist.devname = devname;
- sprintf(devname, "%d:%d", major(dev), minor(dev));
-
- devlist.disposition = 'r';
- if (Manage_subdevs(st2->devname, fd2, &devlist, -1, 0) == 0) {
- devlist.disposition = 'a';
- if (Manage_subdevs(st->devname, fd1, &devlist, -1, 0) == 0) {
- alert("MoveSpare", st->devname, st2->devname, info);
- close(fd1);
- close(fd2);
+
+ err = st->ss->load_container(st, fd, NULL);
+ close(fd);
+ if (err)
+ return 0;
+
+ if (from == to) {
+ /* We must check if number of active disks has not increased
+ * since ioctl in main loop. mdmon may have added spare
+ * to subarray. If so we do not need to look for more spares
+ * so return non zero value */
+ int active_cnt = 0;
+ struct mdinfo *dp;
+ list = st->ss->getinfo_super_disks(st);
+ if (!list) {
+ st->ss->free_super(st);
+ return 1;
+ }
+ dp = list->devs;
+ while (dp) {
+ if (dp->disk.state & (1<<MD_DISK_SYNC) &&
+ !(dp->disk.state & (1<<MD_DISK_FAULTY)))
+ active_cnt++;
+ dp = dp->next;
+ }
+ sysfs_free(list);
+ if (active < active_cnt) {
+ /* Spare just activated.*/
+ st->ss->free_super(st);
return 1;
}
- else Manage_subdevs(st2->devname, fd2, &devlist, -1, 0);
}
- close(fd1);
- close(fd2);
- return 0;
+
+ /* We only need one spare so full list not needed */
+ list = container_choose_spares(st, min_size, domlist, from->spare_group,
+ to->metadata->ss->name, 1);
+ if (list) {
+ struct mdinfo *disks = list->devs;
+ if (disks)
+ dev = makedev(disks->disk.major, disks->disk.minor);
+ sysfs_free(list);
+ }
+ st->ss->free_super(st);
+ return dev;
}
+
static void try_spare_migration(struct state *statelist, struct alert_info *info)
{
+ struct state *from;
struct state *st;
link_containers_with_subarrays(statelist);
- for (st = statelist; st; st=st->next)
+ for (st = statelist; st; st = st->next)
if (st->active < st->raid &&
- st->spare == 0 &&
- st->spare_group != NULL) {
- struct state *st2;
- for (st2=statelist ; st2 ; st2=st2->next)
- if (st2 != st &&
- st2->spare > 0 &&
- st2->active == st2->raid &&
- st2->spare_group != NULL &&
- strcmp(st->spare_group, st2->spare_group) == 0)
- if (move_spare(st2, st, info))
- break;
+ st->spare == 0 && !st->err) {
+ struct domainlist *domlist = NULL;
+ int d;
+ struct state *to = st;
+ unsigned long long min_size;
+
+ if (to->parent_dev != NoMdDev && !to->parent)
+ /* subarray monitored without parent container
+ * we can't move spares here */
+ continue;
+
+ if (to->parent)
+ /* member of a container */
+ to = to->parent;
+
+ if (get_min_spare_size_required(to, &min_size))
+ continue;
+ if (to->metadata->ss->external) {
+ /* We must make sure there is
+ * no suitable spare in container already.
+ * If there is we don't add more */
+ dev_t devid = container_choose_spare(
+ to, to, NULL, min_size, st->active);
+ if (devid > 0)
+ continue;
+ }
+ for (d = 0; d < MaxDisks; d++)
+ if (to->devid[d])
+ domainlist_add_dev(&domlist,
+ to->devid[d],
+ to->metadata->ss->name);
+ if (to->spare_group)
+ domain_add(&domlist, to->spare_group);
+ /*
+ * No spare migration if the destination
+ * has no domain. Skip this array.
+ */
+ if (!domlist)
+ continue;
+ for (from=statelist ; from ; from=from->next) {
+ dev_t devid;
+ if (!check_donor(from, to))
+ continue;
+ if (from->metadata->ss->external)
+ devid = container_choose_spare(
+ from, to, domlist, min_size, 0);
+ else
+ devid = choose_spare(from, to, domlist,
+ min_size);
+ if (devid > 0
+ && move_spare(from->devname, to->devname, devid)) {
+ alert("MoveSpare", to->devname, from->devname, info);
+ break;
+ }
+ }
+ domain_free(domlist);
}
}
if (is_subarray(&e->metadata_version[9]))
ping_monitor(&e->metadata_version[9]);
else
- ping_monitor(devnum2devname(devnum));
+ ping_monitor_by_id(devnum);
}
free_mdstat(ms);
return rv;