drbd: Ensure that the peer was not rebootet in the meantime before resending TL

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_nl.c b/drivers/block/drbd/drbd_nl.c
index d764f3c..921793c 100644 (file)
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -208,9 +208,6 @@ enum drbd_disk_state drbd_try_outdate_peer(struct drbd_conf *mdev)
                return mdev->state.pdsk;
        }
 
-       if (fp == FP_STONITH)
-               _drbd_request_state(mdev, NS(susp, 1), CS_WAIT_COMPLETE);
-
        r = drbd_khelper(mdev, "fence-peer");
 
        switch ((r>>8) & 0xff) {

diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 101ad18..3a8131a 100644 (file)
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -3265,7 +3265,18 @@ static int receive_state(struct drbd_conf *mdev, struct p_header *h)
        ns.peer_isp = (peer_state.aftr_isp | peer_state.user_isp);
        if ((nconn == C_CONNECTED || nconn == C_WF_BITMAP_S) && ns.disk == D_NEGOTIATING)
                ns.disk = mdev->new_state_tmp.disk;
-
+       if (ns.pdsk == D_CONSISTENT && ns.susp && nconn == C_CONNECTED && oconn < C_CONNECTED &&
+           test_bit(NEW_CUR_UUID, &mdev->flags)) {
+               /* Do not allow tl_restart(resend) for a rebooted peer. We can only allow this
+                  for temporal network outages! */
+               spin_unlock_irq(&mdev->req_lock);
+               dev_err(DEV, "Aborting Connect, can not thaw IO with an only Consistent peer\n");
+               tl_clear(mdev);
+               drbd_uuid_new_current(mdev);
+               clear_bit(NEW_CUR_UUID, &mdev->flags);
+               drbd_force_state(mdev, NS2(conn, C_PROTOCOL_ERROR, susp, 0));
+               return FALSE;
+       }
        rv = _drbd_set_state(mdev, ns, CS_VERBOSE | CS_HARD, NULL);
        ns = mdev->state;
        spin_unlock_irq(&mdev->req_lock);

diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index fbe0278..76b6682 100644 (file)
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -662,7 +662,7 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what,
                /* If RQ_NET_OK is already set, we got a P_WRITE_ACK or P_RECV_ACK
                   before the connection loss (B&C only); only P_BARRIER_ACK was missing.
                   Trowing them out of the TL here by pretending we got a BARRIER_ACK
-                  TODO: Either resync them, or ensure peer was not rebooted. */
+                  We ensure that the peer was not rebooted */
                if (!(req->rq_state & RQ_NET_OK)) {
                        if (req->w.cb) {
                                drbd_queue_work(&mdev->data.work, &req->w);