diff --git a/CHANGES.txt b/CHANGES.txt
index f21553b823f..0cf6dad1d12 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -816,6 +816,8 @@ Release 0.90.0 - Unreleased
    HBASE-3402  Web UI shows two META regions
    HBASE-3409  Failed server shutdown processing when retrying hlog split
    HBASE-3412  HLogSplitter should handle missing HLogs
+   HBASE-3420  Handling a big rebalance, we can queue multiple instances of
+               a Close event; messes up state
 
 
   IMPROVEMENTS
diff --git a/src/main/java/org/apache/hadoop/hbase/master/AssignmentManager.java b/src/main/java/org/apache/hadoop/hbase/master/AssignmentManager.java
index 675810d8e37..e44b8e0b7c3 100644
--- a/src/main/java/org/apache/hadoop/hbase/master/AssignmentManager.java
+++ b/src/main/java/org/apache/hadoop/hbase/master/AssignmentManager.java
@@ -1567,11 +1567,13 @@ public class AssignmentManager extends ZooKeeperListener {
             // Expired!  Do a retry.
             switch (regionState.getState()) {
               case CLOSED:
-                LOG.info("Region has been CLOSED for too long, " +
-                    "retriggering ClosedRegionHandler");
-                AssignmentManager.this.executorService.submit(
-                    new ClosedRegionHandler(master, AssignmentManager.this,
-                        regionState.getRegion()));
+                LOG.info("Region " + regionInfo.getEncodedName() +
+                  " has been CLOSED for too long, waiting on queued " +
+                  "ClosedRegionHandler to run or server shutdown");
+                // Update our timestamp.
+                synchronized(regionState) {
+                  regionState.update(regionState.getState());
+                }
                 break;
               case OFFLINE:
                 LOG.info("Region has been OFFLINE for too long, " +