ceph: allow connection to be reopened by fault callback

Fix the messenger to allow a ceph_con_open() during the fault callback.
Previously the work wasn't getting queued on the connection because the
fault path avoids requeued work (normally spurious).  Loop on reopening by
checking for the OPENING state bit.

This fixes OSD reconnects when a TCP connection drops.

Signed-off-by: Sage Weil <sage@newdream.net>

fs/ceph/messenger.c

@@ -1808,7 +1808,7 @@ done:
 	clear_bit(BUSY, &con->state);
 	dout("con->state=%lu\n", con->state);
 	if (test_bit(QUEUED, &con->state)) {
-		if (!backoff) {
+		if (!backoff || test_bit(OPENING, &con->state)) {
 			dout("con_work %p QUEUED reset, looping\n", con);
 			goto more;
 		}