Merge remote-tracking branch 'upstream/main' into fix-logsdb-settings-provider-mapping-filters

Author: jordan-powers

docs/changelog/135966.yaml

@@ -0,0 +1,6 @@
+pr: 135966
+summary: Ensure queued `AbstractRunnables` are notified when executor stops
+area: Machine Learning
+type: bug
+issues:
+ - 134651

rest-api-spec/src/main/resources/rest-api-spec/api/indices.get_sample_stats.json

@@ -0,0 +1,31 @@
+{
+  "indices.get_sample_stats": {
+    "documentation": {
+      "url": "https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-indices-get-sample",
+      "description": "Get stats about a random sample of ingested data"
+    },
+    "stability": "experimental",
+    "visibility": "public",
+    "headers": {
+      "accept": [
+        "application/json"
+      ]
+    },
+    "url": {
+      "paths": [
+        {
+          "path": "/{index}/_sample/stats",
+          "methods": [
+            "GET"
+          ],
+          "parts": {
+            "index": {
+              "type": "string",
+              "description": "The name of a data stream or index"
+            }
+          }
+        }
+      ]
+    }
+  }
+}

rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/indices.get_sample_stats/10_basic.yml

@@ -0,0 +1,19 @@
+---
+"Test get sample stats for index with no sample config":
+  - requires:
+      cluster_features: [ "random_sampling" ]
+      reason: requires feature 'random_sampling' to get random samples
+
+  - do:
+      indices.get_sample_stats:
+        index: non_existent
+      catch: missing
+
+  - do:
+      indices.create:
+        index: no_config
+
+  - do:
+      indices.get_sample_stats:
+        index: no_config
+      catch: missing

server/src/internalClusterTest/java/org/elasticsearch/action/admin/indices/sampling/GetSampleActionIT.java

@@ -16,13 +16,15 @@
 import org.elasticsearch.cluster.metadata.ProjectMetadata;
 import org.elasticsearch.cluster.service.ClusterService;
 import org.elasticsearch.common.Priority;
+import org.elasticsearch.core.TimeValue;
 import org.elasticsearch.ingest.SamplingService;
 import org.elasticsearch.test.ESIntegTestCase;
 
 import java.util.List;
 import java.util.Map;
 
 import static org.hamcrest.Matchers.equalTo;
+import static org.hamcrest.Matchers.greaterThan;
 
 public class GetSampleActionIT extends ESIntegTestCase {
 
@@ -34,7 +36,8 @@ public void testGetSample() throws Exception {
         createIndex(indexName);
         // the index exists but there is no sampling configuration for it, so getting its sample will throw an exception:
         assertGetSampleThrowsResourceNotFoundException(indexName);
-        addSamplingConfig(indexName);
+        final int maxSamples = 30;
+        addSamplingConfig(indexName, maxSamples);
         // There is now a sampling configuration, but no data has been ingested:
         assertEmptySample(indexName);
         int docsToIndex = randomIntBetween(1, 20);
@@ -49,6 +52,29 @@ public void testGetSample() throws Exception {
         for (int i = 0; i < docsToIndex; i++) {
             assertRawDocument(sample.get(i), indexName);
         }
+
+        GetSampleStatsAction.Request statsRequest = new GetSampleStatsAction.Request(indexName);
+        GetSampleStatsAction.Response statsResponse = client().execute(GetSampleStatsAction.INSTANCE, statsRequest).actionGet();
+        SamplingService.SampleStats stats = statsResponse.getSampleStats();
+        assertThat(stats.getSamples(), equalTo((long) docsToIndex));
+        assertThat(stats.getPotentialSamples(), equalTo((long) docsToIndex));
+        assertThat(stats.getTimeSampling(), greaterThan(TimeValue.ZERO));
+        assertThat(stats.getSamplesRejectedForMaxSamplesExceeded(), equalTo(0L));
+        assertThat(stats.getSamplesRejectedForRate(), equalTo(0L));
+        assertThat(stats.getSamplesRejectedForCondition(), equalTo(0L));
+        assertThat(stats.getSamplesRejectedForCondition(), equalTo(0L));
+
+        final int samplesOverMax = randomIntBetween(1, 5);
+        for (int i = docsToIndex; i < maxSamples + samplesOverMax; i++) {
+            indexDoc(indexName, randomIdentifier(), randomAlphanumericOfLength(10), randomAlphanumericOfLength(10));
+        }
+        statsRequest = new GetSampleStatsAction.Request(indexName);
+        statsResponse = client().execute(GetSampleStatsAction.INSTANCE, statsRequest).actionGet();
+        stats = statsResponse.getSampleStats();
+        assertThat(stats.getSamples(), equalTo((long) maxSamples));
+        assertThat(stats.getPotentialSamples(), equalTo((long) maxSamples + samplesOverMax));
+        assertThat(stats.getSamplesRejectedForMaxSamplesExceeded(), equalTo((long) samplesOverMax));
+
     }
 
     private void assertRawDocument(SamplingService.RawDocument rawDocument, String indexName) {
@@ -68,7 +94,7 @@ private void assertGetSampleThrowsResourceNotFoundException(String indexName) {
     }
 
     @SuppressWarnings("deprecation")
-    private void addSamplingConfig(String indexName) throws Exception {
+    private void addSamplingConfig(String indexName, int maxSamples) throws Exception {
         /*
          * Note: The following code writes a sampling config directly to the cluster state. It can be replaced with a call to the action
          * that does this once that action exists.
@@ -81,7 +107,7 @@ public ClusterState execute(ClusterState currentState) throws Exception {
                     currentState.metadata().getProject(ProjectId.DEFAULT)
                 );
                 SamplingMetadata samplingMetadata = new SamplingMetadata(
-                    Map.of(indexName, new SamplingConfiguration(1.0d, 100, null, null, null))
+                    Map.of(indexName, new SamplingConfiguration(1.0d, maxSamples, null, null, null))
                 );
                 projectMetadataBuilder.putCustom(SamplingMetadata.TYPE, samplingMetadata);
                 ClusterState newState = new ClusterState.Builder(currentState).putProjectMetadata(projectMetadataBuilder).build();

server/src/main/java/org/elasticsearch/action/ActionModule.java

@@ -130,8 +130,11 @@
 import org.elasticsearch.action.admin.indices.rollover.RolloverAction;
 import org.elasticsearch.action.admin.indices.rollover.TransportRolloverAction;
 import org.elasticsearch.action.admin.indices.sampling.GetSampleAction;
+import org.elasticsearch.action.admin.indices.sampling.GetSampleStatsAction;
 import org.elasticsearch.action.admin.indices.sampling.RestGetSampleAction;
+import org.elasticsearch.action.admin.indices.sampling.RestGetSampleStatsAction;
 import org.elasticsearch.action.admin.indices.sampling.TransportGetSampleAction;
+import org.elasticsearch.action.admin.indices.sampling.TransportGetSampleStatsAction;
 import org.elasticsearch.action.admin.indices.segments.IndicesSegmentsAction;
 import org.elasticsearch.action.admin.indices.segments.TransportIndicesSegmentsAction;
 import org.elasticsearch.action.admin.indices.settings.get.GetSettingsAction;
@@ -821,6 +824,7 @@ public <Request extends ActionRequest, Response extends ActionResponse> void reg
 
         if (RANDOM_SAMPLING_FEATURE_FLAG) {
             actions.register(GetSampleAction.INSTANCE, TransportGetSampleAction.class);
+            actions.register(GetSampleStatsAction.INSTANCE, TransportGetSampleStatsAction.class);
         }
 
         return unmodifiableMap(actions.getRegistry());
@@ -1053,6 +1057,7 @@ public void initRestHandlers(Supplier<DiscoveryNodes> nodesInCluster, Predicate<
 
         if (RANDOM_SAMPLING_FEATURE_FLAG) {
             registerHandler.accept(new RestGetSampleAction());
+            registerHandler.accept(new RestGetSampleStatsAction());
         }
     }