DM RAID: Fix inability to handle failed devices during start-up

The code fails to create a RAID array if any of the superblocks cannot be read.
However, if the userspace program passing in the device-mapper table spots the
failure and passes in the sentinel values ('- -') for the failed array position,
then everything is handled properly. This functional gap should not exist.

This patch makes it so that if a superblock cannot be read, the array position's
fields are initialized as though '- -' were passed in via the CTR table. That
is, the device is failed and the position should not be used; but if there is
sufficient redundancy, the array should still be activated.

Signed-off-by: Jonathan Brassow <jbrassow@redhat.com>

Index: linux-upstream/drivers/md/dm-raid.c
================================================== =================
--- linux-upstream.orig/drivers/md/dm-raid.c
+++ linux-upstream/drivers/md/dm-raid.c
@@ -603,7 +603,9 @@ static int read_disk_sb(struct md_rdev *
return 0;

if (!sync_page_io(rdev, 0, size, rdev->sb_page, READ, 1)) {
- DMERR("Failed to read device superblock");
+ DMERR("Failed to read superblock of device at position %d",
+ rdev->raid_disk);
+ set_bit(Faulty, &rdev->flags);
return -EINVAL;
}

@@ -854,9 +856,25 @@ static int super_validate(struct mddev *
static int analyse_superblocks(struct dm_target *ti, struct raid_set *rs)
{
int ret;
+ unsigned redundancy = 0;
+ struct raid_dev *dev;
struct md_rdev *rdev, *freshest, *tmp;
struct mddev *mddev = &rs->md;

+ switch (rs->raid_type->level) {
+ case 1:
+ redundancy = rs->md.raid_disks - 1;
+ break;
+ case 4:
+ case 5:
+ case 6:
+ redundancy = rs->raid_type->parity_devs;
+ break;
+ default:
+ ti->error = "Unknown RAID type";
+ return -EINVAL;
+ }
+
freshest = NULL;
rdev_for_each(rdev, tmp, mddev) {
if (!rdev->meta_bdev)
@@ -871,6 +889,34 @@ static int analyse_superblocks(struct dm
case 0:
break;
default:
+ dev = container_of(rdev, struct raid_dev, rdev);
+ if (redundancy--) {
+ if (dev->meta_dev)
+ dm_put_device(ti, dev->meta_dev);
+ dev->meta_dev = NULL;
+ rdev->meta_bdev = NULL;
+
+ if (rdev->sb_page)
+ put_page(rdev->sb_page);
+ rdev->sb_page = NULL;
+
+ rdev->sb_loaded = 0;
+
+ /*
+ * We might be able to salvage the data device
+ * even though the meta device has failed. For
+ * now, we behave as though '- -' had been
+ * passed in for this device via the CTR table.
+ */
+ if (dev->data_dev)
+ dm_put_device(ti, dev->data_dev);
+ dev->data_dev = NULL;
+ rdev->bdev = NULL;
+
+ list_del(&rdev->same_set);
+
+ continue;
+ }
ti->error = "Failed to load superblock";
return ret;
}


--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel