Merge remote-tracking branch 'origin/main' into fix_transport_service…

…_on_closed_executor
henningandersen · May 31, 2024 · d89ecb6 · d89ecb6
2 parents 7747639 + 3f23c15
commit d89ecb6
Show file tree

Hide file tree

Showing 39 changed files with 296 additions and 333 deletions.
diff --git a/docs/changelog/109240.yaml b/docs/changelog/109240.yaml
@@ -0,0 +1,5 @@
+pr: 109240
+summary: Fix trappy timeout in allocation explain API
+area: Allocation
+type: bug
+issues: []
diff --git a/docs/changelog/109241.yaml b/docs/changelog/109241.yaml
@@ -0,0 +1,5 @@
+pr: 109241
+summary: Fix misc trappy allocation API timeouts
+area: Allocation
+type: bug
+issues: []
diff --git a/docs/reference/data-streams/tsds-index-settings.asciidoc b/docs/reference/data-streams/tsds-index-settings.asciidoc
@@ -60,5 +60,5 @@ information, refer to <<dimension-based-routing>>.
 `index.mapping.dimension_fields.limit`::
 (<<dynamic-index-settings,Dynamic>>, integer)
 Maximum number of <<time-series-dimension,time series dimensions>> for the
-index. Defaults to `21`.
+index. Defaults to `32768`.
 // end::dimensions-limit[]
diff --git a/.../src/yamlRestTest/resources/rest-api-spec/test/data_stream/200_rollover_failure_store.yml b/.../src/yamlRestTest/resources/rest-api-spec/test/data_stream/200_rollover_failure_store.yml
@@ -218,7 +218,7 @@ teardown:
 
   - do:
       allowed_warnings:
-        - "index template [generic_logs_template] has index patterns [logs-*] matching patterns from existing older templates [global] with patterns (global => [*]); this template [generic_logs_template] will take precedence during new index creation"
+        - "index template [my-template] has index patterns [data-*] matching patterns from existing older templates [global] with patterns (global => [*]); this template [my-template] will take precedence during new index creation"
       indices.put_index_template:
         name: my-template
         body:

diff --git a/muted-tests.yml b/muted-tests.yml
@@ -35,10 +35,6 @@ tests:
 - class: "org.elasticsearch.upgrades.MlTrainedModelsUpgradeIT"
   issue: "https://github.com/elastic/elasticsearch/issues/108993"
   method: "testTrainedModelInference"
-- class: "org.elasticsearch.datastreams.DataStreamsClientYamlTestSuiteIT"
-  issue: "https://github.com/elastic/elasticsearch/issues/109154"
-  method: "test {p0=data_stream/200_rollover_failure_store/Lazily roll over a data\
-    \ stream's failure store after an ingest failure}"
 - class: "org.elasticsearch.xpack.security.CoreWithSecurityClientYamlTestSuiteIT"
   issue: "https://github.com/elastic/elasticsearch/issues/109188"
   method: "test {yaml=search/180_locale_dependent_mapping/Test Index and Search locale\

diff --git a/rest-api-spec/src/main/resources/rest-api-spec/api/_internal.delete_desired_balance.json b/rest-api-spec/src/main/resources/rest-api-spec/api/_internal.delete_desired_balance.json
@@ -18,6 +18,12 @@
           ]
         }
       ]
+    },
+    "params": {
+      "master_timeout":{
+        "type":"time",
+        "description":"Timeout for connection to master node"
+      }
     }
   }
 }
diff --git a/rest-api-spec/src/main/resources/rest-api-spec/api/_internal.get_desired_balance.json b/rest-api-spec/src/main/resources/rest-api-spec/api/_internal.get_desired_balance.json
@@ -18,6 +18,12 @@
           ]
         }
       ]
+    },
+    "params": {
+      "master_timeout":{
+        "type":"time",
+        "description":"Timeout for connection to master node"
+      }
     }
   }
 }
diff --git a/rest-api-spec/src/main/resources/rest-api-spec/api/cluster.allocation_explain.json b/rest-api-spec/src/main/resources/rest-api-spec/api/cluster.allocation_explain.json
@@ -22,6 +22,10 @@
       ]
     },
     "params":{
+      "master_timeout":{
+        "type":"time",
+        "description":"Timeout for connection to master node"
+      },
       "include_yes_decisions":{
         "type":"boolean",
         "description":"Return 'YES' decisions in explanation (default: false)"

diff --git a/...st/java/org/elasticsearch/action/admin/cluster/allocation/ClusterAllocationExplainIT.java b/...st/java/org/elasticsearch/action/admin/cluster/allocation/ClusterAllocationExplainIT.java
@@ -158,12 +158,7 @@ public void testUnassignedReplicaDelayedAllocation() throws Exception {
         // wait till we have passed any pending shard data fetching
         assertEquals(
             AllocationDecision.ALLOCATION_DELAYED,
-            clusterAdmin().prepareAllocationExplain()
-                .setIndex("idx")
-                .setShard(0)
-                .setPrimary(false)
-                .get()
-                .getExplanation()
+            ClusterAllocationExplanationUtils.getClusterAllocationExplanation(client(), "idx", 0, false)
                 .getShardAllocationDecision()
                 .getAllocateDecision()
                 .getAllocationDecision()
@@ -1076,12 +1071,12 @@ public void testCannotAllocateStaleReplicaExplanation() throws Exception {
 
         // wait until the system has fetched shard data and we know there is no valid shard copy
         assertBusy(() -> {
-            ClusterAllocationExplanation explanation = clusterAdmin().prepareAllocationExplain()
-                .setIndex("idx")
-                .setShard(0)
-                .setPrimary(true)
-                .get()
-                .getExplanation();
+            ClusterAllocationExplanation explanation = ClusterAllocationExplanationUtils.getClusterAllocationExplanation(
+                client(),
+                "idx",
+                0,
+                true
+            );
             assertTrue(explanation.getShardAllocationDecision().getAllocateDecision().isDecisionTaken());
             assertEquals(
                 AllocationDecision.NO_VALID_SHARD_COPY,
@@ -1223,19 +1218,11 @@ private ClusterAllocationExplanation runExplain(boolean primary, boolean include
         return runExplain(primary, null, includeYesDecisions, includeDiskInfo);
     }
 
-    private ClusterAllocationExplanation runExplain(boolean primary, String nodeId, boolean includeYesDecisions, boolean includeDiskInfo)
-        throws Exception {
-
-        ClusterAllocationExplanation explanation = admin().cluster()
-            .prepareAllocationExplain()
-            .setIndex("idx")
-            .setShard(0)
-            .setPrimary(primary)
-            .setIncludeYesDecisions(includeYesDecisions)
-            .setIncludeDiskInfo(includeDiskInfo)
-            .setCurrentNode(nodeId)
-            .get()
-            .getExplanation();
+    private ClusterAllocationExplanation runExplain(boolean primary, String nodeId, boolean includeYesDecisions, boolean includeDiskInfo) {
+        final var request = new ClusterAllocationExplainRequest(TEST_REQUEST_TIMEOUT, "idx", 0, primary, nodeId);
+        request.includeYesDecisions(includeYesDecisions);
+        request.includeDiskInfo(includeDiskInfo);
+        final var explanation = safeGet(client().execute(TransportClusterAllocationExplainAction.TYPE, request)).getExplanation();
         if (logger.isDebugEnabled()) {
             logger.debug("--> explain json output: \n{}", Strings.toString(explanation, true, true));
         }

diff --git a/...org/elasticsearch/action/admin/cluster/allocation/TransportGetDesiredBalanceActionIT.java b/...org/elasticsearch/action/admin/cluster/allocation/TransportGetDesiredBalanceActionIT.java
@@ -40,8 +40,9 @@ public void testDesiredBalanceOnMultiNodeCluster() throws Exception {
         var clusterHealthResponse = clusterAdmin().health(new ClusterHealthRequest().waitForStatus(ClusterHealthStatus.GREEN)).get();
         assertEquals(RestStatus.OK, clusterHealthResponse.status());
 
-        DesiredBalanceResponse desiredBalanceResponse = client().execute(TransportGetDesiredBalanceAction.TYPE, new DesiredBalanceRequest())
-            .get();
+        final var desiredBalanceResponse = safeGet(
+            client().execute(TransportGetDesiredBalanceAction.TYPE, new DesiredBalanceRequest(TEST_REQUEST_TIMEOUT))
+        );
 
         assertEquals(1, desiredBalanceResponse.getRoutingTable().size());
         Map<Integer, DesiredBalanceResponse.DesiredShards> shardsMap = desiredBalanceResponse.getRoutingTable().get(index);
@@ -75,8 +76,9 @@ public void testDesiredBalanceWithUnassignedShards() throws Exception {
         var clusterHealthResponse = clusterAdmin().health(new ClusterHealthRequest(index).waitForStatus(ClusterHealthStatus.YELLOW)).get();
         assertEquals(RestStatus.OK, clusterHealthResponse.status());
 
-        DesiredBalanceResponse desiredBalanceResponse = client().execute(TransportGetDesiredBalanceAction.TYPE, new DesiredBalanceRequest())
-            .get();
+        final var desiredBalanceResponse = safeGet(
+            client().execute(TransportGetDesiredBalanceAction.TYPE, new DesiredBalanceRequest(TEST_REQUEST_TIMEOUT))
+        );
 
         assertEquals(1, desiredBalanceResponse.getRoutingTable().size());
         Map<Integer, DesiredBalanceResponse.DesiredShards> shardsMap = desiredBalanceResponse.getRoutingTable().get(index);

diff --git a/server/src/internalClusterTest/java/org/elasticsearch/cluster/PrevalidateShardPathIT.java b/server/src/internalClusterTest/java/org/elasticsearch/cluster/PrevalidateShardPathIT.java
@@ -9,8 +9,7 @@
 package org.elasticsearch.cluster;
 
 import org.apache.lucene.tests.util.LuceneTestCase;
-import org.elasticsearch.action.admin.cluster.allocation.ClusterAllocationExplainRequest;
-import org.elasticsearch.action.admin.cluster.allocation.TransportClusterAllocationExplainAction;
+import org.elasticsearch.action.admin.cluster.allocation.ClusterAllocationExplanationUtils;
 import org.elasticsearch.action.admin.cluster.node.shutdown.NodePrevalidateShardPathResponse;
 import org.elasticsearch.action.admin.cluster.node.shutdown.PrevalidateShardPathRequest;
 import org.elasticsearch.action.admin.cluster.node.shutdown.PrevalidateShardPathResponse;
@@ -90,13 +89,12 @@ public void testCheckShards() throws Exception {
                         .filter(s -> node2ShardIds.contains(s.shardId()))
                         .filter(s -> s.currentNodeId().equals(node2Id))
                         .toList()) {
-                        var explanation = client().execute(
-                            TransportClusterAllocationExplainAction.TYPE,
-                            new ClusterAllocationExplainRequest().setIndex(node2Shard.getIndexName())
-                                .setCurrentNode(node2Shard.currentNodeId())
-                                .setShard(node2Shard.id())
-                                .setPrimary(node2Shard.primary())
-                        ).get();
+                        var explanation = ClusterAllocationExplanationUtils.getClusterAllocationExplanation(
+                            client(),
+                            node2Shard.getIndexName(),
+                            node2Shard.id(),
+                            node2Shard.primary()
+                        );
                         logger.info(
                             "Shard: {} is still located on relocation source node: {}. Allocation explanation: {}",
                             node2Shard.shardId(),

diff --git a/server/src/internalClusterTest/java/org/elasticsearch/cluster/routing/AllocationIdIT.java b/server/src/internalClusterTest/java/org/elasticsearch/cluster/routing/AllocationIdIT.java
@@ -8,14 +8,13 @@
 
 package org.elasticsearch.cluster.routing;
 
-import org.elasticsearch.action.admin.cluster.allocation.ClusterAllocationExplanation;
+import org.elasticsearch.action.admin.cluster.allocation.ClusterAllocationExplanationUtils;
 import org.elasticsearch.action.admin.cluster.health.ClusterHealthRequest;
 import org.elasticsearch.action.admin.indices.stats.ShardStats;
 import org.elasticsearch.action.index.IndexRequestBuilder;
 import org.elasticsearch.cluster.ClusterState;
 import org.elasticsearch.cluster.health.ClusterHealthStatus;
 import org.elasticsearch.cluster.routing.allocation.AllocationDecision;
-import org.elasticsearch.cluster.routing.allocation.ShardAllocationDecision;
 import org.elasticsearch.cluster.routing.allocation.command.AllocateStalePrimaryAllocationCommand;
 import org.elasticsearch.common.settings.Settings;
 import org.elasticsearch.index.IndexService;
@@ -199,14 +198,12 @@ private void putFakeCorruptionMarker(IndexSettings indexSettings, ShardId shardI
 
     private void checkNoValidShardCopy(String indexName, ShardId shardId) throws Exception {
         assertBusy(() -> {
-            final ClusterAllocationExplanation explanation = clusterAdmin().prepareAllocationExplain()
-                .setIndex(indexName)
-                .setShard(shardId.id())
-                .setPrimary(true)
-                .get()
-                .getExplanation();
-
-            final ShardAllocationDecision shardAllocationDecision = explanation.getShardAllocationDecision();
+            final var shardAllocationDecision = ClusterAllocationExplanationUtils.getClusterAllocationExplanation(
+                client(),
+                indexName,
+                shardId.id(),
+                true
+            ).getShardAllocationDecision();
             assertThat(shardAllocationDecision.isDecisionTaken(), equalTo(true));
             assertThat(
                 shardAllocationDecision.getAllocateDecision().getAllocationDecision(),

diff --git a/...rnalClusterTest/java/org/elasticsearch/index/shard/RemoveCorruptedShardDataCommandIT.java b/...rnalClusterTest/java/org/elasticsearch/index/shard/RemoveCorruptedShardDataCommandIT.java
@@ -19,7 +19,6 @@
 import org.apache.lucene.store.LockObtainFailedException;
 import org.apache.lucene.store.NativeFSLockFactory;
 import org.elasticsearch.ExceptionsHelper;
-import org.elasticsearch.action.admin.cluster.allocation.ClusterAllocationExplanation;
 import org.elasticsearch.action.admin.cluster.node.stats.NodesStatsResponse;
 import org.elasticsearch.action.admin.indices.flush.FlushRequest;
 import org.elasticsearch.action.admin.indices.recovery.RecoveryResponse;
@@ -37,7 +36,6 @@
 import org.elasticsearch.cluster.routing.ShardRoutingState;
 import org.elasticsearch.cluster.routing.UnassignedInfo;
 import org.elasticsearch.cluster.routing.allocation.AllocationDecision;
-import org.elasticsearch.cluster.routing.allocation.ShardAllocationDecision;
 import org.elasticsearch.cluster.routing.allocation.command.AllocateStalePrimaryAllocationCommand;
 import org.elasticsearch.common.settings.Settings;
 import org.elasticsearch.common.unit.ByteSizeUnit;
@@ -78,6 +76,7 @@
 import java.util.stream.Collectors;
 import java.util.stream.StreamSupport;
 
+import static org.elasticsearch.action.admin.cluster.allocation.ClusterAllocationExplanationUtils.getClusterAllocationExplanation;
 import static org.elasticsearch.common.util.CollectionUtils.iterableAsArrayList;
 import static org.elasticsearch.index.query.QueryBuilders.matchAllQuery;
 import static org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertAcked;
@@ -173,14 +172,7 @@ public Settings onNodeStopped(String nodeName) throws Exception {
 
         // shard should be failed due to a corrupted index
         assertBusy(() -> {
-            final ClusterAllocationExplanation explanation = clusterAdmin().prepareAllocationExplain()
-                .setIndex(indexName)
-                .setShard(0)
-                .setPrimary(true)
-                .get()
-                .getExplanation();
-
-            final ShardAllocationDecision shardAllocationDecision = explanation.getShardAllocationDecision();
+            final var shardAllocationDecision = getClusterAllocationExplanation(client(), indexName, 0, true).getShardAllocationDecision();
             assertThat(shardAllocationDecision.isDecisionTaken(), equalTo(true));
             assertThat(
                 shardAllocationDecision.getAllocateDecision().getAllocationDecision(),
@@ -219,14 +211,7 @@ public Settings onNodeStopped(String nodeName) throws Exception {
 
         // there is only _stale_ primary (due to new allocation id)
         assertBusy(() -> {
-            final ClusterAllocationExplanation explanation = clusterAdmin().prepareAllocationExplain()
-                .setIndex(indexName)
-                .setShard(0)
-                .setPrimary(true)
-                .get()
-                .getExplanation();
-
-            final ShardAllocationDecision shardAllocationDecision = explanation.getShardAllocationDecision();
+            final var shardAllocationDecision = getClusterAllocationExplanation(client(), indexName, 0, true).getShardAllocationDecision();
             assertThat(shardAllocationDecision.isDecisionTaken(), equalTo(true));
             assertThat(
                 shardAllocationDecision.getAllocateDecision().getAllocationDecision(),
@@ -237,13 +222,7 @@ public Settings onNodeStopped(String nodeName) throws Exception {
         clusterAdmin().prepareReroute().add(new AllocateStalePrimaryAllocationCommand(indexName, 0, nodeId, true)).get();
 
         assertBusy(() -> {
-            final ClusterAllocationExplanation explanation = clusterAdmin().prepareAllocationExplain()
-                .setIndex(indexName)
-                .setShard(0)
-                .setPrimary(true)
-                .get()
-                .getExplanation();
-
+            final var explanation = getClusterAllocationExplanation(client(), indexName, 0, true);
             assertThat(explanation.getCurrentNode(), notNullValue());
             assertThat(explanation.getShardState(), equalTo(ShardRoutingState.STARTED));
         });
@@ -331,13 +310,7 @@ public Settings onNodeStopped(String nodeName) throws Exception {
 
         // all shards should be failed due to a corrupted translog
         assertBusy(() -> {
-            final UnassignedInfo unassignedInfo = clusterAdmin().prepareAllocationExplain()
-                .setIndex(indexName)
-                .setShard(0)
-                .setPrimary(true)
-                .get()
-                .getExplanation()
-                .getUnassignedInfo();
+            final UnassignedInfo unassignedInfo = getClusterAllocationExplanation(client(), indexName, 0, true).getUnassignedInfo();
             assertThat(unassignedInfo.getReason(), equalTo(UnassignedInfo.Reason.ALLOCATION_FAILED));
             assertThat(ExceptionsHelper.unwrap(unassignedInfo.getFailure(), TranslogCorruptedException.class), not(nullValue()));
         });
@@ -392,14 +365,7 @@ public Settings onNodeStopped(String nodeName) throws Exception {
 
         // there is only _stale_ primary (due to new allocation id)
         assertBusy(() -> {
-            final ClusterAllocationExplanation explanation = clusterAdmin().prepareAllocationExplain()
-                .setIndex(indexName)
-                .setShard(0)
-                .setPrimary(true)
-                .get()
-                .getExplanation();
-
-            final ShardAllocationDecision shardAllocationDecision = explanation.getShardAllocationDecision();
+            final var shardAllocationDecision = getClusterAllocationExplanation(client(), indexName, 0, true).getShardAllocationDecision();
             assertThat(shardAllocationDecision.isDecisionTaken(), equalTo(true));
             assertThat(
                 shardAllocationDecision.getAllocateDecision().getAllocationDecision(),
@@ -410,13 +376,7 @@ public Settings onNodeStopped(String nodeName) throws Exception {
         clusterAdmin().prepareReroute().add(new AllocateStalePrimaryAllocationCommand(indexName, 0, primaryNodeId, true)).get();
 
         assertBusy(() -> {
-            final ClusterAllocationExplanation explanation = clusterAdmin().prepareAllocationExplain()
-                .setIndex(indexName)
-                .setShard(0)
-                .setPrimary(true)
-                .get()
-                .getExplanation();
-
+            final var explanation = getClusterAllocationExplanation(client(), indexName, 0, true);
             assertThat(explanation.getCurrentNode(), notNullValue());
             assertThat(explanation.getShardState(), equalTo(ShardRoutingState.STARTED));
         });

diff --git a/server/src/internalClusterTest/java/org/elasticsearch/index/store/CorruptedTranslogIT.java b/server/src/internalClusterTest/java/org/elasticsearch/index/store/CorruptedTranslogIT.java
@@ -9,7 +9,6 @@
 package org.elasticsearch.index.store;
 
 import org.elasticsearch.ExceptionsHelper;
-import org.elasticsearch.action.admin.cluster.allocation.ClusterAllocationExplainResponse;
 import org.elasticsearch.action.index.IndexRequestBuilder;
 import org.elasticsearch.action.search.SearchPhaseExecutionException;
 import org.elasticsearch.cluster.routing.UnassignedInfo;
@@ -32,6 +31,7 @@
 import java.util.Arrays;
 import java.util.Collection;
 
+import static org.elasticsearch.action.admin.cluster.allocation.ClusterAllocationExplanationUtils.getClusterAllocationExplanation;
 import static org.elasticsearch.index.query.QueryBuilders.matchAllQuery;
 import static org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertAcked;
 import static org.hamcrest.Matchers.containsString;
@@ -79,13 +79,9 @@ public void onAllNodesStopped() throws Exception {
 
         assertBusy(() -> {
             // assertBusy since the shard starts out unassigned with reason CLUSTER_RECOVERED, then it's assigned, and then it fails.
-            final ClusterAllocationExplainResponse allocationExplainResponse = clusterAdmin().prepareAllocationExplain()
-                .setIndex("test")
-                .setShard(0)
-                .setPrimary(true)
-                .get();
-            final String description = Strings.toString(allocationExplainResponse.getExplanation());
-            final UnassignedInfo unassignedInfo = allocationExplainResponse.getExplanation().getUnassignedInfo();
+            final var allocationExplainResponse = getClusterAllocationExplanation(client(), "test", 0, true);
+            final var description = Strings.toString(allocationExplainResponse);
+            final var unassignedInfo = allocationExplainResponse.getUnassignedInfo();
             assertThat(description, unassignedInfo, not(nullValue()));
             assertThat(description, unassignedInfo.getReason(), equalTo(UnassignedInfo.Reason.ALLOCATION_FAILED));
             var failure = unassignedInfo.getFailure();