drbd: fix "stalled" empty resync

With sync-after dependencies, given "lucky" timing of pause/unpause
events, and the end of an empty (0 bits set) resync was sometimes not
detected on the SyncTarget, leading to a "stalled" SyncSource state.

Fixed this by expecting not only "Inconsistent -> UpToDate" but also
"Consistent -> UpToDate" transitions for the peer disk state
to end a resync.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>

diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index aa6c52c80e2ecaf6b99af47e3a704fa313c74c40..1a48e02b83bc73709efd460eaefd251eee0aa626 100644 (file)
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -3169,9 +3169,14 @@ static int receive_state(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned
        os = ns = mdev->state;
        spin_unlock_irq(&mdev->req_lock);
 
-       /* peer says his disk is uptodate, while we think it is inconsistent,
-        * and this happens while we think we have a sync going on. */
-       if (os.pdsk == D_INCONSISTENT && real_peer_disk == D_UP_TO_DATE &&
+       /* If this is the "end of sync" confirmation, usually the peer disk
+        * transitions from D_INCONSISTENT to D_UP_TO_DATE. For empty (0 bits
+        * set) resync started in PausedSyncT, or if the timing of pause-/
+        * unpause-sync events has been "just right", the peer disk may
+        * transition from D_CONSISTENT to D_UP_TO_DATE as well.
+        */
+       if ((os.pdsk == D_INCONSISTENT || os.pdsk == D_CONSISTENT) &&
+           real_peer_disk == D_UP_TO_DATE &&
            os.conn > C_CONNECTED && os.disk == D_UP_TO_DATE) {
                /* If we are (becoming) SyncSource, but peer is still in sync
                 * preparation, ignore its uptodate-ness to avoid flapping, it