Commit c0ffeb64 authored by Zheng Qixing's avatar Zheng Qixing Committed by Yu Kuai
Browse files

md: allow removing faulty rdev during resync



During RAID resync, faulty rdev cannot be removed and will result in
"Device or resource busy" error when attempting hot removal.

Reproduction steps:
  mdadm -Cv /dev/md0 -l1 -n3 -e1.2 /dev/sd{b..d}
  mdadm /dev/md0 -f /dev/sdb
  mdadm /dev/md0 -r /dev/sdb
  -> mdadm: hot remove failed for /dev/sdb: Device or resource busy

After commit 4b10a3bc ("md: ensure resync is prioritized over
recovery"), when a device becomes faulty during resync, the
md_choose_sync_action() function returns early without calling
remove_and_add_spares(), preventing faulty device removal.

This patch extracts a helper function remove_spares() to support
removing faulty devices during RAID resync operations.

Fixes: 4b10a3bc ("md: ensure resync is prioritized over recovery")
Signed-off-by: default avatarZheng Qixing <zhengqixing@huawei.com>
Reviewed-by: default avatarLi Nan <linan122@huawei.com>
Link: https://lore.kernel.org/linux-raid/20250707075412.150301-1-zhengqixing@huaweicloud.com


Signed-off-by: default avatarYu Kuai <yukuai3@huawei.com>
parent 3ec8db61
Loading
Loading
Loading
Loading
+17 −7
Original line number Diff line number Diff line
@@ -9459,17 +9459,11 @@ static bool md_spares_need_change(struct mddev *mddev)
	return false;
}

static int remove_and_add_spares(struct mddev *mddev,
				 struct md_rdev *this)
static int remove_spares(struct mddev *mddev, struct md_rdev *this)
{
	struct md_rdev *rdev;
	int spares = 0;
	int removed = 0;

	if (this && test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
		/* Mustn't remove devices when resync thread is running */
		return 0;

	rdev_for_each(rdev, mddev) {
		if ((this == NULL || rdev == this) && rdev_removeable(rdev) &&
		    !mddev->pers->hot_remove_disk(mddev, rdev)) {
@@ -9483,6 +9477,21 @@ static int remove_and_add_spares(struct mddev *mddev,
	if (removed && mddev->kobj.sd)
		sysfs_notify_dirent_safe(mddev->sysfs_degraded);

	return removed;
}

static int remove_and_add_spares(struct mddev *mddev,
				 struct md_rdev *this)
{
	struct md_rdev *rdev;
	int spares = 0;
	int removed = 0;

	if (this && test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
		/* Mustn't remove devices when resync thread is running */
		return 0;

	removed = remove_spares(mddev, this);
	if (this && removed)
		goto no_add;

@@ -9525,6 +9534,7 @@ static bool md_choose_sync_action(struct mddev *mddev, int *spares)

	/* Check if resync is in progress. */
	if (mddev->recovery_cp < MaxSector) {
		remove_spares(mddev, NULL);
		set_bit(MD_RECOVERY_SYNC, &mddev->recovery);
		clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
		return true;