Commit 90382ed9 authored by Goldwyn Rodrigues's avatar Goldwyn Rodrigues Committed by NeilBrown
Browse files

Fix read-balancing during node failure 



During a node failure, We need to suspend read balancing so that the
reads are directed to the first device and stale data is not read.
Suspending writes is not required because these would be recorded and
synced eventually.

A new flag MD_CLUSTER_SUSPEND_READ_BALANCING is set in recover_prep().
area_resyncing() will respond true for the entire devices if this
flag is set and the request type is READ. The flag is cleared
in recover_done().

Signed-off-by: default avatarGoldwyn Rodrigues <rgoldwyn@suse.com>
Reported-By: default avatarDavid Teigland <teigland@redhat.com>
Signed-off-by: default avatarNeilBrown <neilb@suse.com>
parent 33e38ac6

drivers/md/md-cluster.c

+11 −1
Original line number Original line Diff line number Diff line
@@ -44,6 +44,7 @@ struct resync_info { 




/* md_cluster_info flags */

/* md_cluster_info flags */

#define		MD_CLUSTER_WAITING_FOR_NEWDISK		1

#define		MD_CLUSTER_WAITING_FOR_NEWDISK		1

#define		MD_CLUSTER_SUSPEND_READ_BALANCING	2









struct md_cluster_info {

struct md_cluster_info {
@@ -275,6 +276,9 @@ clear_bit: 




static void recover_prep(void *arg)

static void recover_prep(void *arg)

{

{

	struct mddev *mddev = arg;

	struct md_cluster_info *cinfo = mddev->cluster_info;

	set_bit(MD_CLUSTER_SUSPEND_READ_BALANCING, &cinfo->state);

}

}





static void recover_slot(void *arg, struct dlm_slot *slot)

static void recover_slot(void *arg, struct dlm_slot *slot)
@@ -307,6 +311,7 @@ static void recover_done(void *arg, struct dlm_slot *slots, 




	cinfo->slot_number = our_slot;

	cinfo->slot_number = our_slot;

	complete(&cinfo->completion);

	complete(&cinfo->completion);

	clear_bit(MD_CLUSTER_SUSPEND_READ_BALANCING, &cinfo->state);

}

}





static const struct dlm_lockspace_ops md_ls_ops = {

static const struct dlm_lockspace_ops md_ls_ops = {
@@ -816,12 +821,17 @@ static void resync_finish(struct mddev *mddev) 
	resync_send(mddev, RESYNCING, 0, 0);

	resync_send(mddev, RESYNCING, 0, 0);

}

}





static int area_resyncing(struct mddev *mddev, sector_t lo, sector_t hi)

static int area_resyncing(struct mddev *mddev, int direction,

		sector_t lo, sector_t hi)

{

{

	struct md_cluster_info *cinfo = mddev->cluster_info;

	struct md_cluster_info *cinfo = mddev->cluster_info;

	int ret = 0;

	int ret = 0;

	struct suspend_info *s;

	struct suspend_info *s;





	if ((direction == READ) &&

		test_bit(MD_CLUSTER_SUSPEND_READ_BALANCING, &cinfo->state))

		return 1;



	spin_lock_irq(&cinfo->suspend_lock);

	spin_lock_irq(&cinfo->suspend_lock);

	if (list_empty(&cinfo->suspend_list))

	if (list_empty(&cinfo->suspend_list))

		goto out;

		goto out;

drivers/md/md-cluster.h

+1 −1
Original line number Original line Diff line number Diff line
@@ -18,7 +18,7 @@ struct md_cluster_operations { 
	int (*metadata_update_start)(struct mddev *mddev);

	int (*metadata_update_start)(struct mddev *mddev);

	int (*metadata_update_finish)(struct mddev *mddev);

	int (*metadata_update_finish)(struct mddev *mddev);

	int (*metadata_update_cancel)(struct mddev *mddev);

	int (*metadata_update_cancel)(struct mddev *mddev);

	int (*area_resyncing)(struct mddev *mddev, sector_t lo, sector_t hi);

	int (*area_resyncing)(struct mddev *mddev, int direction, sector_t lo, sector_t hi);

	int (*add_new_disk_start)(struct mddev *mddev, struct md_rdev *rdev);

	int (*add_new_disk_start)(struct mddev *mddev, struct md_rdev *rdev);

	int (*add_new_disk_finish)(struct mddev *mddev);

	int (*add_new_disk_finish)(struct mddev *mddev);

	int (*new_disk_ack)(struct mddev *mddev, bool ack);

	int (*new_disk_ack)(struct mddev *mddev, bool ack);

drivers/md/raid1.c

+4 −3
Original line number Original line Diff line number Diff line
@@ -541,7 +541,7 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect 




	if ((conf->mddev->recovery_cp < this_sector + sectors) ||

	if ((conf->mddev->recovery_cp < this_sector + sectors) ||

	    (mddev_is_clustered(conf->mddev) &&

	    (mddev_is_clustered(conf->mddev) &&

	    md_cluster_ops->area_resyncing(conf->mddev, this_sector,

	    md_cluster_ops->area_resyncing(conf->mddev, READ, this_sector,

		    this_sector + sectors)))

		    this_sector + sectors)))

		choose_first = 1;

		choose_first = 1;

	else

	else
@@ -1111,7 +1111,8 @@ static void make_request(struct mddev *mddev, struct bio * bio) 
	    ((bio_end_sector(bio) > mddev->suspend_lo &&

	    ((bio_end_sector(bio) > mddev->suspend_lo &&

	    bio->bi_iter.bi_sector < mddev->suspend_hi) ||

	    bio->bi_iter.bi_sector < mddev->suspend_hi) ||

	    (mddev_is_clustered(mddev) &&

	    (mddev_is_clustered(mddev) &&

	     md_cluster_ops->area_resyncing(mddev, bio->bi_iter.bi_sector, bio_end_sector(bio))))) {

	     md_cluster_ops->area_resyncing(mddev, WRITE,

		     bio->bi_iter.bi_sector, bio_end_sector(bio))))) {

		/* As the suspend_* range is controlled by

		/* As the suspend_* range is controlled by

		 * userspace, we want an interruptible

		 * userspace, we want an interruptible

		 * wait.

		 * wait.
@@ -1124,7 +1125,7 @@ static void make_request(struct mddev *mddev, struct bio * bio) 
			if (bio_end_sector(bio) <= mddev->suspend_lo ||

			if (bio_end_sector(bio) <= mddev->suspend_lo ||

			    bio->bi_iter.bi_sector >= mddev->suspend_hi ||

			    bio->bi_iter.bi_sector >= mddev->suspend_hi ||

			    (mddev_is_clustered(mddev) &&

			    (mddev_is_clustered(mddev) &&

			     !md_cluster_ops->area_resyncing(mddev,

			     !md_cluster_ops->area_resyncing(mddev, WRITE,

				     bio->bi_iter.bi_sector, bio_end_sector(bio))))

				     bio->bi_iter.bi_sector, bio_end_sector(bio))))

				break;

				break;

			schedule();

			schedule();

CRA Git | Maintained and supported by SUSTech CRA and CCSE