drbd: add some more explicit drbd_md_sync

It sometimes may take a while for the after state change work to be
scheduled, which does drbd_md_sync. At convenient places, we should do
explicit drbd_md_sync to have the new state information on disk as soon
as possible.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
This commit is contained in:
Lars Ellenberg 2010-10-14 13:37:40 +02:00 committed by Philipp Reisner
parent 9d282875d8
commit 856c50c7b6

View File

@ -3804,6 +3804,9 @@ static void drbdd(struct drbd_conf *mdev)
err_out:
drbd_force_state(mdev, NS(conn, C_PROTOCOL_ERROR));
}
/* If we leave here, we probably want to update at least the
* "Connected" indicator on stable storage. Do so explicitly here. */
drbd_md_sync(mdev);
}
void drbd_flush_workqueue(struct drbd_conf *mdev)
@ -4685,10 +4688,12 @@ int drbd_asender(struct drbd_thread *thi)
if (0) {
reconnect:
drbd_force_state(mdev, NS(conn, C_NETWORK_FAILURE));
drbd_md_sync(mdev);
}
if (0) {
disconnect:
drbd_force_state(mdev, NS(conn, C_DISCONNECTING));
drbd_md_sync(mdev);
}
clear_bit(SIGNAL_ASENDER, &mdev->flags);