Merge pull request #10308 from thalesmg/test-increase-peer-timeout-v50

test(peer): increase init and startup timeout for peer nodes

.ci/docker-compose-file/cassandra/cassandra.yaml

@@ -469,8 +469,8 @@ concurrent_materialized_view_writes: 32
 # accepting writes when the limit is exceeded until a flush completes,
 # and will trigger a flush based on memtable_cleanup_threshold
 # If omitted, Cassandra will set both to 1/4 the size of the heap.
-# memtable_heap_space_in_mb: 2048
-# memtable_offheap_space_in_mb: 2048
+memtable_heap_space_in_mb: 2048
+memtable_offheap_space_in_mb: 2048
 
 # memtable_cleanup_threshold is deprecated. The default calculation
 # is the only reasonable choice. See the comments on  memtable_flush_writers

.ci/docker-compose-file/docker-compose-cassandra.yaml

@@ -12,6 +12,8 @@ services:
     environment:
       CASSANDRA_BROADCAST_ADDRESS: "1.2.3.4"
       CASSANDRA_RPC_ADDRESS: "0.0.0.0"
+      HEAP_NEWSIZE: "128M"
+      MAX_HEAP_SIZE: "2048M"
     volumes:
       - ./certs:/certs
     #ports:

apps/emqx/test/emqx_common_test_helpers.erl

@@ -660,6 +660,7 @@ start_slave(Name, Opts) when is_list(Opts) ->
 start_slave(Name, Opts) when is_map(Opts) ->
     SlaveMod = maps:get(peer_mod, Opts, ct_slave),
     Node = node_name(Name),
+    put_peer_mod(Node, SlaveMod),
     DoStart =
         fun() ->
             case SlaveMod of
@@ -669,8 +670,8 @@ start_slave(Name, Opts) when is_map(Opts) ->
                         [
                             {kill_if_fail, true},
                             {monitor_master, true},
-                            {init_timeout, 10000},
-                            {startup_timeout, 10000},
+                            {init_timeout, 20_000},
+                            {startup_timeout, 20_000},
                             {erl_flags, erl_flags()}
                         ]
                     );
@@ -687,7 +688,6 @@ start_slave(Name, Opts) when is_map(Opts) ->
             throw(Other)
     end,
     pong = net_adm:ping(Node),
-    put_peer_mod(Node, SlaveMod),
     setup_node(Node, Opts),
     ok = snabbkaffe:forward_trace(Node),
     Node.

apps/emqx/test/emqx_test_janitor.erl

@@ -65,7 +65,7 @@ terminate(_Reason, #{callbacks := Callbacks}) ->
 handle_call({push, Callback}, _From, State = #{callbacks := Callbacks}) ->
     {reply, ok, State#{callbacks := [Callback | Callbacks]}};
 handle_call(terminate, _From, State = #{callbacks := Callbacks}) ->
-    lists:foreach(fun(Fun) -> Fun() end, Callbacks),
+    lists:foreach(fun(Fun) -> catch Fun() end, Callbacks),
     {stop, normal, ok, State};
 handle_call(_Req, _From, State) ->
     {reply, error, State}.

lib-ee/emqx_ee_bridge/test/emqx_bridge_impl_kafka_consumer_SUITE.erl

@@ -388,7 +388,9 @@ end_per_testcase(_Testcase, Config) ->
                 maps:values(ProducersMapping)
             ),
             ok = wolff:stop_and_delete_supervised_client(KafkaProducerClientId),
-            emqx_common_test_helpers:call_janitor(),
+            %% in CI, apparently this needs more time since the
+            %% machines struggle with all the containers running...
+            emqx_common_test_helpers:call_janitor(60_000),
             ok = snabbkaffe:stop(),
             ok
     end.
@@ -1664,7 +1666,7 @@ t_cluster_group(Config) ->
                  || {Name, Opts} <- Cluster
                 ],
             on_exit(fun() ->
-                lists:foreach(
+                emqx_misc:pmap(
                     fun(N) ->
                         ct:pal("stopping ~p", [N]),
                         ok = emqx_common_test_helpers:stop_slave(N)
@@ -1875,7 +1877,7 @@ t_cluster_node_down(Config) ->
                     Cluster
                 ),
             on_exit(fun() ->
-                lists:foreach(
+                emqx_misc:pmap(
                     fun(N) ->
                         ct:pal("stopping ~p", [N]),
                         ok = emqx_common_test_helpers:stop_slave(N)
@@ -1894,11 +1896,15 @@ t_cluster_node_down(Config) ->
             {ok, _} = snabbkaffe:receive_events(SRef0),
             lists:foreach(
                 fun(N) ->
+                    ?retry(
+                        _Sleep1 = 100,
+                        _Attempts1 = 50,
                         ?assertMatch(
                             {ok, _},
                             erpc:call(N, emqx_bridge, lookup, [BridgeId]),
                             #{node => N}
                         )
+                    )
                 end,
                 Nodes
             ),