Enable failure detector in tests

Disabled for raptor because currently they fail with failure-detector
enabled and jmx tests because they rely on having a consistent number of
nodes.

presto-hive/src/test/java/com/facebook/presto/hive/TestHivePushdownFilterQueries.java

@@ -103,6 +103,9 @@ private static QueryRunner createQueryRunner()
     {
         DistributedQueryRunner queryRunner = HiveQueryRunner.createQueryRunner(getTables(),
                 ImmutableMap.of("experimental.pushdown-subfields-enabled", "true"),
+                // TODO: enable failure detector.  Currently this test has a ton of major GC activity on travis,
+                //  and the failure detector may make the test run longer
+                ImmutableMap.of("failure-detector.enabled", "false"),
                 "sql-standard",
                 ImmutableMap.of("hive.pushdown-filter-enabled", "true"),
                 Optional.empty());
@@ -173,8 +176,8 @@ public void testLegacyUnnest()
 
         assertQuery(legacyUnnest, "SELECT orderkey, date.day FROM lineitem_ex CROSS JOIN UNNEST(dates) t(date)",
                 "SELECT orderkey, day(shipdate) FROM lineitem WHERE orderkey % 31 <> 0 UNION ALL " +
-                "SELECT orderkey, day(commitdate) FROM lineitem WHERE orderkey % 31 <> 0 UNION ALL " +
-                "SELECT orderkey, day(receiptdate) FROM lineitem WHERE orderkey % 31 <> 0");
+                        "SELECT orderkey, day(commitdate) FROM lineitem WHERE orderkey % 31 <> 0 UNION ALL " +
+                        "SELECT orderkey, day(receiptdate) FROM lineitem WHERE orderkey % 31 <> 0");
     }
 
     @Test

presto-hive/src/test/java/com/facebook/presto/hive/TestHiveRecoverableGroupedExecution.java

@@ -94,8 +94,9 @@ public void setUp()
                         // this effectively disables the retries
                         "query.remote-task.max-error-duration", "1s",
                         // allow 2 out of 4 tasks to fail
-                        "max-failed-task-percentage", "0.6"),
-
+                        "max-failed-task-percentage", "0.6",
+                        // turn off the failure detector since with the shortened timeouts, it becomes too prone to failure
+                        "failure-detector.enabled", "false"),
                 Optional.empty());
         executor = listeningDecorator(newCachedThreadPool());
     }

presto-jmx/src/test/java/com/facebook/presto/connector/jmx/JmxQueryRunner.java

@@ -15,6 +15,7 @@
 
 import com.facebook.presto.Session;
 import com.facebook.presto.tests.DistributedQueryRunner;
+import com.google.common.collect.ImmutableMap;
 
 import static com.facebook.airlift.testing.Closeables.closeAllSuppress;
 import static com.facebook.presto.connector.jmx.JmxMetadata.JMX_SCHEMA_NAME;
@@ -30,7 +31,11 @@ public static DistributedQueryRunner createJmxQueryRunner()
     {
         DistributedQueryRunner queryRunner = null;
         try {
-            queryRunner = new DistributedQueryRunner(createSession(), 3);
+            queryRunner = DistributedQueryRunner.builder(createSession())
+                    .setNodeCount(3)
+                    // disable failure-detector to prevent flaky tests since the jmx tests rely on the number of nodes being consistent
+                    .setCoordinatorProperties(ImmutableMap.of("failure-detector.enabled", "false"))
+                    .build();
 
             queryRunner.installPlugin(new JmxPlugin());
             queryRunner.createCatalog("jmx", "jmx");

presto-main/src/main/java/com/facebook/presto/server/testing/TestingPrestoServer.java

@@ -232,11 +232,6 @@ public TestingPrestoServer(
                 .put("task.max-worker-threads", "4")
                 .put("exchange.client-threads", "4");
 
-        if (coordinator) {
-            // TODO: enable failure detector
-            serverProperties.put("failure-detector.enabled", "false");
-        }
-
         ImmutableList.Builder<Module> modules = ImmutableList.<Module>builder()
                 .add(new TestingNodeModule(Optional.ofNullable(environment)))
                 .add(new TestingHttpServerModule(parseInt(coordinator ? coordinatorPort : "0")))

presto-raptor/src/test/java/com/facebook/presto/raptor/RaptorQueryRunner.java

@@ -57,7 +57,12 @@ public static DistributedQueryRunner createRaptorQueryRunner(
             Map<String, String> extraRaptorProperties)
             throws Exception
     {
-        DistributedQueryRunner queryRunner = new DistributedQueryRunner(createSession("tpch"), 2, extraProperties);
+        DistributedQueryRunner queryRunner = DistributedQueryRunner.builder(createSession("tpch"))
+                .setNodeCount(2)
+                // TODO: enable failure-detector for raptor. Currently enabling the failure detector causes failures.
+                .setCoordinatorProperties(ImmutableMap.of("failure-detector.enabled", "false"))
+                .setExtraProperties(extraProperties)
+                .build();
 
         queryRunner.installPlugin(new TpchPlugin());
         queryRunner.createCatalog("tpch", "tpch");

presto-tests/src/test/java/com/facebook/presto/tests/TestMinWorkerRequirement.java

@@ -21,6 +21,7 @@
 import static org.testng.Assert.fail;
 
 // run single threaded to avoid creating multiple query runners at once
+// failure detector is disabled in these tests to prevent flakiness since the tests assert a specific number of workers are present
 @Test(singleThreaded = true)
 public class TestMinWorkerRequirement
 {
@@ -29,7 +30,10 @@ public void testInsufficientInitialWorkerNodes()
             throws Exception
     {
         try (DistributedQueryRunner queryRunner = TpchQueryRunnerBuilder.builder()
-                .setSingleCoordinatorProperty("query-manager.initialization-required-workers", "5")
+                .setCoordinatorProperties(ImmutableMap.<String, String>builder()
+                        .put("query-manager.initialization-required-workers", "5")
+                        .put("failure-detector.enabled", "false")
+                        .build())
                 .setNodeCount(4)
                 .build()) {
             queryRunner.execute("SELECT 1");
@@ -43,7 +47,10 @@ public void testInsufficientInitialWorkerNodesWithCoordinatorExcluded()
     {
         try (DistributedQueryRunner queryRunner = TpchQueryRunnerBuilder.builder()
                 .setSingleExtraProperty("node-scheduler.include-coordinator", "false")
-                .setSingleCoordinatorProperty("query-manager.initialization-required-workers", "4")
+                .setCoordinatorProperties(ImmutableMap.<String, String>builder()
+                        .put("query-manager.initialization-required-workers", "4")
+                        .put("failure-detector.enabled", "false")
+                        .build())
                 .setNodeCount(4)
                 .build()) {
             queryRunner.execute("SELECT 1");
@@ -56,7 +63,10 @@ public void testSufficientInitialWorkerNodes()
             throws Exception
     {
         try (DistributedQueryRunner queryRunner = TpchQueryRunnerBuilder.builder()
-                .setSingleCoordinatorProperty("query-manager.initialization-required-workers", "4")
+                .setCoordinatorProperties(ImmutableMap.<String, String>builder()
+                        .put("query-manager.initialization-required-workers", "4")
+                        .put("failure-detector.enabled", "false")
+                        .build())
                 .setNodeCount(4)
                 .build()) {
             queryRunner.execute("SELECT 1");
@@ -77,6 +87,7 @@ public void testInitializationTimeout()
                 .setCoordinatorProperties(ImmutableMap.<String, String>builder()
                         .put("query-manager.initialization-required-workers", "5")
                         .put("query-manager.initialization-timeout", "1ns")
+                        .put("failure-detector.enabled", "false")
                         .build())
                 .setNodeCount(4)
                 .build()) {
@@ -93,6 +104,7 @@ public void testInsufficientWorkerNodes()
                 .setCoordinatorProperties(ImmutableMap.<String, String>builder()
                         .put("query-manager.required-workers", "5")
                         .put("query-manager.required-workers-max-wait", "1ns")
+                        .put("failure-detector.enabled", "false")
                         .build())
                 .setNodeCount(4)
                 .build()) {
@@ -110,6 +122,7 @@ public void testInsufficientWorkerNodesWithCoordinatorExcluded()
                         .put("node-scheduler.include-coordinator", "false")
                         .put("query-manager.required-workers", "4")
                         .put("query-manager.required-workers-max-wait", "1ns")
+                        .put("failure-detector.enabled", "false")
                         .build())
                 .setNodeCount(4)
                 .build()) {
@@ -126,6 +139,7 @@ public void testInsufficientWorkerNodesAfterDrop()
                 .setCoordinatorProperties(ImmutableMap.<String, String>builder()
                         .put("query-manager.required-workers", "4")
                         .put("query-manager.required-workers-max-wait", "1ns")
+                        .put("failure-detector.enabled", "false")
                         .build())
                 .setNodeCount(4)
                 .build()) {