Rev 535: make sure we start the freeze process quickly on all nodes when we are going to do recovery - this prevents serialisation of freeze, which can take a long time in http://samba.org/~tridge/ctdb

[tridge at samba.org]

------------------------------------------------------------
revno: 535
revision-id: tridge at samba.org-20070611130323-4jb1shtg78r4wsed
parent: tridge at samba.org-20070611122723-0n15ehpi0fa9na90
committer: Andrew Tridgell <tridge at samba.org>
branch nick: tridge
timestamp: Mon 2007-06-11 23:03:23 +1000
message:
  make sure we start the freeze process quickly on all nodes when we are going to do recovery - this prevents serialisation of freeze, which can take a long time
modified:
  server/ctdb_recoverd.c         recoverd.c-20070503213540-bvxuyd9jm1f7ig90-1
=== modified file 'server/ctdb_recoverd.c'
--- a/server/ctdb_recoverd.c	2007-06-11 11:37:09 +0000
+++ b/server/ctdb_recoverd.c	2007-06-11 13:03:23 +0000
@@ -125,6 +125,11 @@
 {
 	int j, ret;
 
+	/* start the freeze process immediately on all nodes */
+	ctdb_control(ctdb, CTDB_BROADCAST_CONNECTED, 0, 
+		     CTDB_CONTROL_FREEZE, CTDB_CTRL_FLAG_NOREPLY, tdb_null, 
+		     NULL, NULL, NULL, NULL, NULL);
+
 	/* set recovery mode to active on all nodes */
 	for (j=0; j<nodemap->num; j++) {
 		/* dont change it for nodes that are unavailable */