Manage: support --fail set-X and --remove set-X

A RAID10 array can have 'sets' of devices which are reported by
--detail.
They can now be collectively failed or removed.

Signed-off-by: NeilBrown <neilb@suse.de>

diff --git a/Manage.c b/Manage.c
index e37f4155d32147bfd87d9725750c9a2ab40af080..bdf2eaf3ea2b0da570e739ec4da47fea623ada2f 100644 (file)
--- a/Manage.c
+++ b/Manage.c
@@ -452,6 +452,40 @@ static void add_detached(struct mddev_dev *dv, int fd, char disp)
        }
 }
 
+static void add_set(struct mddev_dev *dv, int fd, char set_char)
+{
+       mdu_array_info_t array;
+       mdu_disk_info_t disk;
+       int remaining_disks;
+       int copies, set;
+       int i;
+
+       if (ioctl(fd, GET_ARRAY_INFO, &array) != 0)
+               return;
+       if (array.level != 10)
+               return;
+       copies = ((array.layout & 0xff) *
+                 ((array.layout >> 8) & 0xff));
+       if (array.raid_disks % copies)
+               return;
+
+       remaining_disks = array.nr_disks;
+       for (i = 0; i < MAX_DISKS && remaining_disks > 0; i++) {
+               char buf[40];
+               disk.number = i;
+               if (ioctl(fd, GET_DISK_INFO, &disk) != 0)
+                       continue;
+               if (disk.major == 0 && disk.minor == 0)
+                       continue;
+               remaining_disks--;
+               set = disk.raid_disk % copies;
+               if (set_char != set + 'A')
+                       continue;
+               sprintf(buf, "%d:%d", disk.major, disk.minor);
+               dv = add_one(dv, buf, dv->disposition);
+       }
+}
+
 int attempt_re_add(int fd, int tfd, struct mddev_dev *dv,
                   struct supertype *dev_st, struct supertype *tst,
                   unsigned long rdev,
@@ -1195,6 +1229,35 @@ int Manage_subdevs(char *devname, int fd,
                        continue;
                }
 
+               if (strncmp(dv->devname, "set-", 4) == 0 &&
+                   strlen(dv->devname) == 5) {
+                       int copies;
+
+                       if (dv->disposition != 'r' &&
+                           dv->disposition != 'f') {
+                               pr_err("'%s' only meaningful with -r or -f\n",
+                                      dv->devname);
+                               goto abort;
+                       }
+                       if (array.level != 10) {
+                               pr_err("'%s' only meaningful with RAID10 arrays\n",
+                                      dv->devname);
+                               goto abort;
+                       }
+                       copies = ((array.layout & 0xff) *
+                                 ((array.layout >> 8) & 0xff));
+                       if (array.raid_disks % copies != 0 ||
+                           dv->devname[4] < 'A' ||
+                           dv->devname[4] >= 'A' + copies ||
+                           copies > 26) {
+                               pr_err("'%s' not meaningful with this array\n",
+                                      dv->devname);
+                               goto abort;
+                       }
+                       add_set(dv, fd, dv->devname[4]);
+                       continue;
+               }
+
                if (strchr(dv->devname, '/') == NULL &&
                    strchr(dv->devname, ':') == NULL &&
                    strlen(dv->devname) < 50) {

diff --git a/mdadm.8.in b/mdadm.8.in
index 39034704a5aa56cdd20916ec163755041ea31637..ef937816d1293dac7f855ebe53031e467d19897e 100644 (file)
--- a/mdadm.8.in
+++ b/mdadm.8.in
@@ -1273,30 +1273,50 @@ resolved.
 .TP
 .BR \-r ", " \-\-remove
 remove listed devices.  They must not be active.  i.e. they should
-be failed or spare devices.  As well as the name of a device file
+be failed or spare devices.
+
+As well as the name of a device file
 (e.g.
 .BR /dev/sda1 )
 the words
-.B failed
-and
+.BR failed ,
 .B detached
+and names like
+.B set-A
 can be given to
 .BR \-\-remove .
 The first causes all failed device to be removed.  The second causes
 any device which is no longer connected to the system (i.e an 'open'
 returns
 .BR ENXIO )
-to be removed.  This will only succeed for devices that are spares or
-have already been marked as failed.
+to be removed.
+The third will remove a set as describe below under
+.BR \-\-fail .
 
 .TP
 .BR \-f ", " \-\-fail
 Mark listed devices as faulty.
 As well as the name of a device file, the word
 .B detached
-can be given.  This will cause any device that has been detached from
+or a set name like
+.B set\-A
+can be given.  The former will cause any device that has been detached from
 the system to be marked as failed.  It can then be removed.
 
+For RAID10 arrays where the number of copies evenly divides the number
+of devices, the devices can be conceptually divided into sets where
+each set contains a single complete copy of the data on the array.
+Sometimes a RAID10 array will be configured so that these sets are on
+separate controllers.  In this case all the devices in one set can be
+failed by giving a name like
+.B set\-A
+or
+.B set\-B
+to
+.BR \-\-fail .
+The appropriate set names are reported by
+.BR \-\-detail .
+
 .TP
 .BR \-\-set\-faulty
 same as