keep track of which nodes are up and don't cleanup after the same node twice

1 files changed, 20 insertions, 12 deletions

diff --git a/src/rabbit_node_monitor.erl b/src/rabbit_node_monitor.erl
index 1f30a2fc..9583cc6b 100644
--- a/src/rabbit_node_monitor.erl
+++ b/src/rabbit_node_monitor.erl
@@ -27,6 +27,8 @@
 -define(SERVER, ?MODULE).
 -define(RABBIT_UP_RPC_TIMEOUT, 2000).
 
+-record(state, {rabbits}).
+
 %%----------------------------------------------------------------------------
 
 -ifdef(use_specs).
@@ -61,27 +63,28 @@ notify_cluster() ->
 
 init([]) ->
     ok = net_kernel:monitor_nodes(true),
-    {ok, no_state}.
+    {ok, #state{rabbits = sets:new()}}.
 
 handle_call(_Request, _From, State) ->
     {noreply, State}.
 
-handle_cast({rabbit_running_on, Node}, State) ->
-    rabbit_log:info("node ~p up~n", [Node]),
+handle_cast({rabbit_running_on, Node}, State = #state{rabbits = Rabbits}) ->
+    rabbit_log:info("rabbit on ~p up~n", [Node]),
     erlang:monitor(process, {rabbit, Node}),
     ok = rabbit_alarm:on_node_up(Node),
-    {noreply, State};
+    {noreply, State#state{rabbits = sets:add_element(Node, Rabbits)}};
 handle_cast(_Msg, State) ->
     {noreply, State}.
 
 handle_info({nodedown, Node}, State) ->
     rabbit_log:info("node ~p down~n", [Node]),
-    ok = handle_dead_rabbit(Node),
-    {noreply, State};
+    {noreply, handle_dead_rabbit(Node, State)};
+handle_info({nodeup, Node}, State = #state{rabbits = Nodes}) ->
+    rabbit_log:info("node ~p up~n", [Node]),
+    {noreply, State#state{rabbits = sets:add_element(Node, Nodes)}};
 handle_info({'DOWN', _MRef, process, {rabbit, Node}, _Reason}, State) ->
     rabbit_log:info("node ~p lost 'rabbit'~n", [Node]),
-    ok = handle_dead_rabbit(Node),
-    {noreply, State};
+    {noreply, handle_dead_rabbit(Node, State)};
 handle_info(_Info, State) ->
     {noreply, State}.
 
@@ -96,7 +99,12 @@ code_change(_OldVsn, State, _Extra) ->
 %% TODO: This may turn out to be a performance hog when there are lots
 %% of nodes.  We really only need to execute some of these statements
 %% on *one* node, rather than all of them.
-handle_dead_rabbit(Node) ->
-    ok = rabbit_networking:on_node_down(Node),
-    ok = rabbit_amqqueue:on_node_down(Node),
-    ok = rabbit_alarm:on_node_down(Node).
+handle_dead_rabbit(Node, State = #state{rabbits = Rabbits}) ->
+    case sets:is_element(Node, Rabbits) of
+        true  -> rabbit_log:info("handling dead rabbit ~p~n", [Node]),
+                 ok = rabbit_networking:on_node_down(Node),
+                 ok = rabbit_amqqueue:on_node_down(Node),
+                 ok = rabbit_alarm:on_node_down(Node),
+                 State#state{rabbits = sets:del_element(Node, Rabbits)};
+        false -> State
+    end.