ctdb-recoverd: Set recovery lock handle at start of attempt

This allows the attempt to be cancelled if an election is lost and an
unlock is done before the attempt is completed.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=13617

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Tue Sep 18 02:18:30 CEST 2018 on sn-devel-144

(cherry picked from commit 486022ef8f43251258f255ffa15f1a01bc6aa2b7)

Autobuild-User(v4-8-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-8-test): Thu Sep 20 14:05:59 CEST 2018 on sn-devel-144

diff --git a/ctdb/server/ctdb_recoverd.c b/ctdb/server/ctdb_recoverd.c
index 5e49e8842d997efb620b71c6f3aa16e03237d5bf..62e4c4655da1f79b4fa6cba041089a4ec6b10eae 100644 (file)
--- a/ctdb/server/ctdb_recoverd.c
+++ b/ctdb/server/ctdb_recoverd.c
@@ -954,17 +954,18 @@ static bool ctdb_recovery_lock(struct ctdb_recoverd *rec)
                return false;
        }
 
+       rec->recovery_lock_handle = s;
+       s->h = h;
+
        while (! s->done) {
                tevent_loop_once(ctdb->ev);
        }
 
        if (! s->locked) {
-               talloc_free(s);
+               TALLOC_FREE(rec->recovery_lock_handle);
                return false;
        }
 
-       rec->recovery_lock_handle = s;
-       s->h = h;
        ctdb_ctrl_report_recd_lock_latency(ctdb,
                                           CONTROL_TIMEOUT(),
                                           s->latency);