From: Ronnie Sahlberg <sahlberg@ronnie>
Date: Sun, 15 Jul 2007 22:36:09 +0000 (+1000)
Subject: add a check if start_node is beyond the end of the nodemap and reset it
X-Git-Tag: tevent-0.9.20~348^2~2450^2
X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=4d1f3acc9413bb2fbfbaa967499ba4e67de1fc04;p=thirdparty%2Fsamba.git

add a check if start_node is beyond the end of the nodemap and reset it
back to 0 if it is to prevent an infinite loop.

this could happen if in the future we add a mechanism to add/remove
nodes to a cluster at runtime

(This used to be ctdb commit 217e80a468713fec86ccb0608460e3401046bb98)
---

diff --git a/ctdb/server/ctdb_takeover.c b/ctdb/server/ctdb_takeover.c
index 56589d90ea5..76f87691e08 100644
--- a/ctdb/server/ctdb_takeover.c
+++ b/ctdb/server/ctdb_takeover.c
@@ -457,6 +457,14 @@ static void ctdb_takeover_find_node(struct ctdb_context *ctdb, struct ctdb_node_
 	static int start_node=0;
 	int j;
 
+	/* If we add facilities to add/remove nodes to a cluster at runtime
+	   we must make sure that start_node is suddently not beyond the
+	   end of the nodelist
+	*/
+	if (start_node >= nodemap->num) {
+		start_node = 0;
+	}
+
 	j=start_node;
 	while (1) {
 		if (!(nodemap->nodes[j].flags & mask_flags) &&