From 67ac6011db5d2b0c853d573ff474b25c85dfb644 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 13 Aug 2009 10:06:24 +1000 Subject: md/raid5: allow new reshape modes to be restarted in the middle. md/raid5 doesn't allow a reshape to restart if it involves writing over the same part of disk that it would be reading from. This happens at the beginning of a reshape that increases the number of devices, at the end of a reshape that decreases the number of devices, and continuously for a reshape that does not change the number of devices. The current code is correct for the "increase number of devices" case as the critical section at the start is handled by userspace performing a backup. It does not work for reducing the number of devices, or the no-change case. For 'reducing', we need to invert the test. For no-change we cannot really be sure things will be safe, so simply require the array to be read-only, which is how the user-space code which carefully starts such arrays works. Signed-off-by: NeilBrown --- drivers/md/raid5.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) (limited to 'drivers/md/raid5.c') diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 2b521ee67df..b8a22a2205c 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -4509,7 +4509,26 @@ static int run(mddev_t *mddev) (old_disks-max_degraded)); /* here_old is the first stripe that we might need to read * from */ - if (here_new >= here_old) { + if (mddev->delta_disks == 0) { + /* We cannot be sure it is safe to start an in-place + * reshape. It is only safe if user-space if monitoring + * and taking constant backups. + * mdadm always starts a situation like this in + * readonly mode so it can take control before + * allowing any writes. So just check for that. + */ + if ((here_new * mddev->new_chunk_sectors != + here_old * mddev->chunk_sectors) || + mddev->ro == 0) { + printk(KERN_ERR "raid5: in-place reshape must be started" + " in read-only mode - aborting\n"); + return -EINVAL; + } + } else if (mddev->delta_disks < 0 + ? (here_new * mddev->new_chunk_sectors <= + here_old * mddev->chunk_sectors) + : (here_new * mddev->new_chunk_sectors >= + here_old * mddev->chunk_sectors)) { /* Reading from the same stripe as writing to - bad */ printk(KERN_ERR "raid5: reshape_position too early for " "auto-recovery - aborting.\n"); -- cgit v1.2.3 From a639755cf885e437b2fe4168d35157fa90d530ab Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 13 Aug 2009 10:13:00 +1000 Subject: md/raid5: make sure a reshape restarts at the correct address. This "if" don't allow for the possibility that the number of devices doesn't change, and so sector_nr isn't set correctly in that case. So change '>' to '>='. Signed-off-by: NeilBrown --- drivers/md/raid5.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/md/raid5.c') diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index b8a22a2205c..94a74cb5ccc 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3785,7 +3785,7 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped conf->reshape_progress < raid5_size(mddev, 0, 0)) { sector_nr = raid5_size(mddev, 0, 0) - conf->reshape_progress; - } else if (mddev->delta_disks > 0 && + } else if (mddev->delta_disks >= 0 && conf->reshape_progress > 0) sector_nr = conf->reshape_progress; sector_div(sector_nr, new_data_disks); -- cgit v1.2.3 From 1a67dde0abba36421a1257d01ba9de2f6d1c160a Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 13 Aug 2009 10:41:49 +1000 Subject: md/raid5: Properly remove excess drives after shrinking a raid5/6 We were removing the drives, from the array, but not removing symlinks from /sys/.... and not marking the device as having been removed. Signed-off-by: NeilBrown --- drivers/md/raid5.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'drivers/md/raid5.c') diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 94a74cb5ccc..b8a2c5dc67b 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -5097,8 +5097,15 @@ static void raid5_finish_reshape(mddev_t *mddev) mddev->degraded--; for (d = conf->raid_disks ; d < conf->raid_disks - mddev->delta_disks; - d++) - raid5_remove_disk(mddev, d); + d++) { + mdk_rdev_t *rdev = conf->disks[d].rdev; + if (rdev && raid5_remove_disk(mddev, d) == 0) { + char nm[20]; + sprintf(nm, "rd%d", rdev->raid_disk); + sysfs_remove_link(&mddev->kobj, nm); + rdev->raid_disk = -1; + } + } } mddev->layout = conf->algorithm; mddev->chunk_sectors = conf->chunk_sectors; -- cgit v1.2.3