drbd: fix refcount error during detach of an already failed disk
authorLars Ellenberg <lars.ellenberg@linbit.com>
Wed, 25 Feb 2015 18:37:28 +0000 (19:37 +0100)
committerJens Axboe <axboe@fb.com>
Wed, 25 Nov 2015 16:22:01 +0000 (09:22 -0700)
commit6434f404b43afa0cfe54fec009760510431ca103
tree3b5adfc83ae2fe16d028d1ec44db50bbe0cf93d1
parent2b479766ee79a821f3cd7fad92fbcf2ff16cacb3
drbd: fix refcount error during detach of an already failed disk

A D_FAILED disk transitions as quickly as possible to
D_DISKLESS. But in the "unresponsive local disk" case,
there remains a time window where a administrative detach command could
find the disk already failed, but some internal meta data IO against the
unresponsive local disk still pending.

In that case, drbd_md_get_buffer() will return NULL.
Don't unconditionally call drbd_md_put_buffer(), or it will cause
refcount imbalance, and prevent any further re-attach on this volume
(until it is deleted and re-created).

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
drivers/block/drbd/drbd_nl.c