From d774b19fe6a1626cca3c92b125e56a22771bba5a Mon Sep 17 00:00:00 2001
From: Mark Reynolds <mreynolds@redhat.com>
Date: May 19 2015 14:46:32 +0000
Subject: Ticket 48158 - cleanAllRUV task limit not being enforced correctly


Bug Description:  The task limit check was only counting tasks that had
                  already begun the cleaning process.  If the task was waiting
                  for a replica to be online before it starting the cleaning
                  process, then task count was not incremented, and limit
                  check was essentially bypassed.

Fix Description:  Use the "pre-cleaned" rid list to determine the task count.
                  The pre-cleaned list is updated as soon a task request
                  comes in - not once the cleaning starts.

https://fedorahosted.org/389/ticket/48158

Reviwed by: nhosoi(Thanks!)

---

diff --git a/ldap/servers/plugins/replication/repl5_replica_config.c b/ldap/servers/plugins/replication/repl5_replica_config.c
index 717a765..6d6c733 100644
--- a/ldap/servers/plugins/replication/repl5_replica_config.c
+++ b/ldap/servers/plugins/replication/repl5_replica_config.c
@@ -3459,7 +3459,7 @@ get_cleanruv_task_count()
 
    slapi_rwlock_wrlock(rid_lock);
    for(i = 0; i < CLEANRIDSIZ; i++){
-       if(cleaned_rids[i] != 0){
+       if(pre_cleaned_rids[i] != 0){
            count++;
        }
    }