more informative and useful error messages

2 files changed, 10 insertions, 4 deletions

diff --git a/src/rabbit_node_monitor.erl b/src/rabbit_node_monitor.erl
index c5c70708..0d3423c9 100644
--- a/src/rabbit_node_monitor.erl
+++ b/src/rabbit_node_monitor.erl
@@ -90,8 +90,10 @@ running_nodes_file_name() ->
     filename:join(rabbit_mnesia:dir(), "nodes_running_at_shutdown").
 
 prepare_cluster_status_files() ->
+    CorruptFiles = fun () -> throw({error, corrupt_cluster_status_files}) end,
     RunningNodes1 = case try_read_file(running_nodes_file_name()) of
                         {ok, [Nodes]} when is_list(Nodes) -> Nodes;
+                        {ok, _      }                     -> CorruptFiles();
                         non_existant                      -> []
                     end,
     {AllNodes1, WantDiscNode} =
@@ -100,6 +102,8 @@ prepare_cluster_status_files() ->
                 {AllNodes, lists:member(node(), DiscNodes0)};
             {ok, [AllNodes0]} when is_list(AllNodes0) ->
                 {AllNodes0, legacy_should_be_disc_node(AllNodes0)};
+            {ok, _} ->
+                CorruptFiles();
             non_existant ->
                 {[], true}
         end,
diff --git a/src/rabbit_upgrade.erl b/src/rabbit_upgrade.erl
index c43ce236..3fbfeed0 100644
--- a/src/rabbit_upgrade.erl
+++ b/src/rabbit_upgrade.erl
@@ -152,12 +152,14 @@ upgrade_mode(AllNodes) ->
                 {true, []}  ->
                     primary;
                 {true, _}  ->
-                    %% TODO: Here I'm assuming that the various cluster status
-                    %% files are consistent with each other, I think I can
-                    %% provide a solution if they're not...
+                    Filename = rabbit_node_monitor:running_nodes_filename(),
                     die("Cluster upgrade needed but other disc nodes shut "
                         "down after this one.~nPlease first start the last "
-                        "disc node to shut down.~n", []);
+                        "disc node to shut down.~n~nNote: if several disc "
+                        "nodes were shut down simultaneously they may "
+                        "all~nshow this message. In which case, remove "
+                        "the lock file on one of them and~nstart that node. "
+                        "The lock file on this node is:~n~n ~s ", [Filename]);
                 {false, _} ->
                     die("Cluster upgrade needed but this is a ram node.~n"
                         "Please first start the last disc node to shut down.",