Allow routing commands with ?retry_failed=true (#42658)

We respect allocation deciders, including the `MaxRetryAllocationDecider`, when
executing reroute commands. If you specify `?retry_failed=true` then the retry
counter is reset, but today this does not happen until after trying to execute
the reroute commands. This means that if an allocation has repeatedly failed,
but you want to take control and assign a shard to a particular node to work
around the repeated failures, you cannot execute the routing command in the
same call to `POST /_cluster/reroute` as the one that resets the failure
counter.

This commit fixes this by resetting the failure counter first, meaning that you
can now explicitly allocate a repeatedly-failed shard like this:

```
POST /_cluster/reroute?retry_failed=true
{
  "commands": [
    {
      "allocate_replica": {
        "index": "blahblah",
        "shard": 2,
        "node": "node-4"
      }
    }
  ]
}
```

Fixes #39546

server/src/main/java/org/elasticsearch/cluster/routing/allocation/AllocationService.java

@@ -337,16 +337,16 @@ public CommandsResult reroute(final ClusterState clusterState, AllocationCommand
         allocation.debugDecision(true);
         // we ignore disable allocation, because commands are explicit
         allocation.ignoreDisable(true);
-        RoutingExplanations explanations = commands.execute(allocation, explain);
-        // we revert the ignore disable flag, since when rerouting, we want the original setting to take place
-        allocation.ignoreDisable(false);
-        // the assumption is that commands will move / act on shards (or fail through exceptions)
-        // so, there will always be shard "movements", so no need to check on reroute
 
         if (retryFailed) {
             resetFailedAllocationCounter(allocation);
         }
 
+        RoutingExplanations explanations = commands.execute(allocation, explain);
+        // we revert the ignore disable flag, since when rerouting, we want the original setting to take place
+        allocation.ignoreDisable(false);
+        // the assumption is that commands will move / act on shards (or fail through exceptions)
+        // so, there will always be shard "movements", so no need to check on reroute
         reroute(allocation);
         return new CommandsResult(explanations, buildResultAndLogHealthChange(clusterState, allocation, "reroute commands"));
     }

server/src/test/java/org/elasticsearch/cluster/routing/allocation/RetryFailedAllocationTests.java

@@ -0,0 +1,97 @@
+/*
+ * Licensed to Elasticsearch under one or more contributor
+ * license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright
+ * ownership. Elasticsearch licenses this file to you under
+ * the Apache License, Version 2.0 (the "License"); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied.  See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+
+package org.elasticsearch.cluster.routing.allocation;
+
+import org.elasticsearch.ElasticsearchException;
+import org.elasticsearch.Version;
+import org.elasticsearch.cluster.ClusterName;
+import org.elasticsearch.cluster.ClusterState;
+import org.elasticsearch.cluster.ESAllocationTestCase;
+import org.elasticsearch.cluster.metadata.IndexMetaData;
+import org.elasticsearch.cluster.metadata.MetaData;
+import org.elasticsearch.cluster.node.DiscoveryNodes;
+import org.elasticsearch.cluster.routing.RoutingTable;
+import org.elasticsearch.cluster.routing.ShardRouting;
+import org.elasticsearch.cluster.routing.ShardRoutingState;
+import org.elasticsearch.cluster.routing.allocation.command.AllocateReplicaAllocationCommand;
+import org.elasticsearch.cluster.routing.allocation.command.AllocationCommands;
+import org.elasticsearch.cluster.routing.allocation.decider.MaxRetryAllocationDecider;
+import org.elasticsearch.common.settings.Settings;
+
+import java.util.Collections;
+import java.util.List;
+
+import static org.hamcrest.Matchers.equalTo;
+import static org.hamcrest.Matchers.sameInstance;
+
+public class RetryFailedAllocationTests extends ESAllocationTestCase {
+
+    private MockAllocationService strategy;
+    private ClusterState clusterState;
+    private final String INDEX_NAME = "index";
+
+    @Override
+    public void setUp() throws Exception {
+        super.setUp();
+        MetaData metaData = MetaData.builder().put(IndexMetaData.builder(INDEX_NAME)
+            .settings(settings(Version.CURRENT)).numberOfShards(1).numberOfReplicas(1)).build();
+        RoutingTable routingTable = RoutingTable.builder().addAsNew(metaData.index(INDEX_NAME)).build();
+        clusterState = ClusterState.builder(ClusterName.DEFAULT).metaData(metaData).routingTable(routingTable)
+            .nodes(DiscoveryNodes.builder().add(newNode("node1")).add(newNode("node2"))).build();
+        strategy = createAllocationService(Settings.EMPTY);
+    }
+
+    private ShardRouting getPrimary() {
+        return clusterState.getRoutingTable().index(INDEX_NAME).shard(0).primaryShard();
+    }
+
+    private ShardRouting getReplica() {
+        return clusterState.getRoutingTable().index(INDEX_NAME).shard(0).replicaShards().get(0);
+    }
+
+    public void testRetryFailedResetForAllocationCommands() {
+        final int retries = MaxRetryAllocationDecider.SETTING_ALLOCATION_MAX_RETRY.get(Settings.EMPTY);
+        clusterState = strategy.reroute(clusterState, "initial allocation");
+        clusterState = strategy.applyStartedShards(clusterState, Collections.singletonList(getPrimary()));
+
+        // Exhaust all replica allocation attempts with shard failures
+        for (int i = 0; i < retries; i++) {
+            List<FailedShard> failedShards = Collections.singletonList(
+                new FailedShard(getReplica(), "failing-shard::attempt-" + i,
+                    new ElasticsearchException("simulated"), randomBoolean()));
+            clusterState = strategy.applyFailedShards(clusterState, failedShards);
+            clusterState = strategy.reroute(clusterState, "allocation retry attempt-" + i);
+        }
+        assertThat("replica should not be assigned", getReplica().state(), equalTo(ShardRoutingState.UNASSIGNED));
+        assertThat("reroute should be a no-op", strategy.reroute(clusterState, "test"), sameInstance(clusterState));
+
+        // Now allocate replica with retry_failed flag set
+        AllocationService.CommandsResult result = strategy.reroute(clusterState,
+            new AllocationCommands(new AllocateReplicaAllocationCommand(INDEX_NAME, 0,
+                getPrimary().currentNodeId().equals("node1") ? "node2" : "node1")),
+            false, true);
+        clusterState = result.getClusterState();
+
+        assertEquals(ShardRoutingState.INITIALIZING, getReplica().state());
+        clusterState = strategy.applyStartedShards(clusterState, Collections.singletonList(getReplica()));
+        assertEquals(ShardRoutingState.STARTED, getReplica().state());
+        assertFalse(clusterState.getRoutingNodes().hasUnassignedShards());
+    }
+}