honeymoose
/
OpenSearch
mirror of https://github.com/honeymoose/OpenSearch.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Reroute Allocate to force primary allocation when enabled 

Typically, the main reason a reroute allocation command with allow_primary is enabled, is to force create an empty new shard because a shard (and its replicas) were lost. This can't be done today because the shard expects to have a valid index where its allocated, we need to clear its post allocation flag to make sure it is allowed to create a fresh index.
This commit is contained in:
Shay Banon 2013-05-08 12:00:10 +02:00
parent 15c8510e65
commit 8a2e5bbe68
5 changed files with 110 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
src/main/java/org/elasticsearch/cluster/routing/IndexRoutingTable.java
View File

@ -83,6 +83,7 @@ public class IndexRoutingTable implements Iterable<IndexShardRoutingTable> {
/**
* Return the index id
*
* @return id of the index
*/
public String index() {
@ -92,6 +93,7 @@ public class IndexRoutingTable implements Iterable<IndexShardRoutingTable> {
/**
* Return the index id
*
* @return id of the index
*/
public String getIndex() {
@ -100,6 +102,7 @@ public class IndexRoutingTable implements Iterable<IndexShardRoutingTable> {
/**
* creates a new {@link IndexRoutingTable} with all shard versions normalized
*
* @return new {@link IndexRoutingTable}
*/
public IndexRoutingTable normalizeVersions() {
@ -152,8 +155,7 @@ public class IndexRoutingTable implements Iterable<IndexShardRoutingTable> {
* {@link IndexRoutingTable} excluding the nodes with the node ids give as
* the <code>excludedNodes</code> parameter.
*
* @param excludedNodes
* id of nodes that will be excluded
* @param excludedNodes id of nodes that will be excluded
* @return number of distinct nodes this index has at least one shard allocated on
*/
public int numberOfNodesShardsAreAllocatedOn(String... excludedNodes) {
@ -201,6 +203,7 @@ public class IndexRoutingTable implements Iterable<IndexShardRoutingTable> {
/**
* Calculates the number of primary shards in active state in routing table
*
* @return number of active primary shards
*/
public int primaryShardsActive() {
@ -237,6 +240,7 @@ public class IndexRoutingTable implements Iterable<IndexShardRoutingTable> {
/**
* Returns a {@link List} of shards that match one of the states listed in {@link ShardRoutingState states}
*
* @param states a set of {@link ShardRoutingState states}
* @return a {@link List} of shards that match one of the given {@link ShardRoutingState states}
*/
@ -308,9 +312,9 @@ public class IndexRoutingTable implements Iterable<IndexShardRoutingTable> {
/**
* Reads an {@link IndexRoutingTable} from an {@link StreamInput}
*
* @param in {@link StreamInput} to read the {@link IndexRoutingTable} from
* @return {@link IndexRoutingTable} read
*
* @throws IOException if something happens during read
*/
public static IndexRoutingTable readFrom(StreamInput in) throws IOException {
@ -327,6 +331,7 @@ public class IndexRoutingTable implements Iterable<IndexShardRoutingTable> {
/**
* Writes an {@link IndexRoutingTable} to a {@link StreamOutput}.
*
* @param index {@link IndexRoutingTable} to write
* @param out {@link StreamOutput} to write to
* @throws IOException if something happens during write
@ -421,6 +426,16 @@ public class IndexRoutingTable implements Iterable<IndexShardRoutingTable> {
return this;
}
/**
* Clears the post allocation flag for the specified shard
*/
public Builder clearPostAllocationFlag(ShardId shardId) {
assert this.index.equals(shardId.index().name());
IndexShardRoutingTable indexShard = shards.get(shardId.id());
shards.put(indexShard.shardId().id(), new IndexShardRoutingTable(indexShard.shardId(), indexShard.shards(), false));
return this;
}
/**
* Adds a new shard routing (makes a copy of it), with reference data used from the index shard routing table
* if it needs to be created.

24
src/main/java/org/elasticsearch/cluster/routing/RoutingNodes.java
View File

@ -19,12 +19,15 @@
package org.elasticsearch.cluster.routing;
import com.google.common.collect.ImmutableSet;
import com.google.common.collect.Sets;
import gnu.trove.map.hash.TObjectIntHashMap;
import org.elasticsearch.cluster.ClusterState;
import org.elasticsearch.cluster.block.ClusterBlocks;
import org.elasticsearch.cluster.metadata.IndexMetaData;
import org.elasticsearch.cluster.metadata.MetaData;
import org.elasticsearch.cluster.node.DiscoveryNode;
import org.elasticsearch.index.shard.ShardId;
import java.util.*;
@ -49,6 +52,8 @@ public class RoutingNodes implements Iterable<RoutingNode> {
private final List<MutableShardRouting> ignoredUnassigned = newArrayList();
private Set<ShardId> clearPostAllocationFlag;
private final Map<String, TObjectIntHashMap<String>> nodesPerAttributeNames = new HashMap<String, TObjectIntHashMap<String>>();
public RoutingNodes(ClusterState clusterState) {
@ -159,6 +164,25 @@ public class RoutingNodes implements Iterable<RoutingNode> {
return nodesToShards();
}
/**
* Clears the post allocation flag for the provided shard id. NOTE: this should be used cautiously
* since it will lead to data loss of the primary shard is not allocated, as it will allocate
* the primary shard on a node and *not* expect it to have an existing valid index there.
*/
public void addClearPostAllocationFlag(ShardId shardId) {
if (clearPostAllocationFlag == null) {
clearPostAllocationFlag = Sets.newHashSet();
}
clearPostAllocationFlag.add(shardId);
}
public Iterable<ShardId> getShardsToClearPostAllocationFlag() {
if (clearPostAllocationFlag == null) {
return ImmutableSet.of();
}
return clearPostAllocationFlag;
}
public RoutingNode node(String nodeId) {
return nodesToShards.get(nodeId);
}

11
src/main/java/org/elasticsearch/cluster/routing/RoutingTable.java
View File

@ -26,6 +26,7 @@ import org.elasticsearch.cluster.metadata.MetaData;
import org.elasticsearch.common.io.stream.StreamInput;
import org.elasticsearch.common.io.stream.StreamOutput;
import org.elasticsearch.index.Index;
import org.elasticsearch.index.shard.ShardId;
import org.elasticsearch.indices.IndexMissingException;
import java.io.IOException;
@ -39,6 +40,7 @@ import static com.google.common.collect.Maps.newHashMap;
/**
* Represents a global cluster-wide routing table for all indices including the
* version of the current routing state.
*
* @see IndexRoutingTable
*/
public class RoutingTable implements Iterable<IndexRoutingTable> {
@ -57,6 +59,7 @@ public class RoutingTable implements Iterable<IndexRoutingTable> {
/**
* Returns the version of the {@link RoutingTable}.
*
* @return version of the {@link RoutingTable}
*/
public long version() {
@ -305,6 +308,14 @@ public class RoutingTable implements Iterable<IndexRoutingTable> {
IndexShardRoutingTable refData = routingNodes.routingTable().index(shardRoutingEntry.index()).shard(shardRoutingEntry.id());
indexBuilder.addShard(refData, shardRoutingEntry);
}
for (ShardId shardId : routingNodes.getShardsToClearPostAllocationFlag()) {
IndexRoutingTable.Builder indexRoutingBuilder = indexRoutingTableBuilders.get(shardId.index().name());
if (indexRoutingBuilder != null) {
indexRoutingBuilder.clearPostAllocationFlag(shardId);
}
}
for (IndexRoutingTable.Builder indexBuilder : indexRoutingTableBuilders.values()) {
add(indexBuilder);
}

8
src/main/java/org/elasticsearch/cluster/routing/allocation/command/AllocateAllocationCommand.java
View File

@ -135,6 +135,7 @@ public class AllocateAllocationCommand implements AllocationCommand {
/**
* Get the shards id
*
* @return id of the shard
*/
public ShardId shardId() {
@ -143,6 +144,7 @@ public class AllocateAllocationCommand implements AllocationCommand {
/**
* Get the id of the Node
*
* @return id of the Node
*/
public String node() {
@ -151,6 +153,7 @@ public class AllocateAllocationCommand implements AllocationCommand {
/**
* Determine if primary allocation is allowed
*
* @return <code>true</code> if primary allocation is allowed. Otherwise <code>false</code>
*/
public boolean allowPrimary() {
@ -191,6 +194,11 @@ public class AllocateAllocationCommand implements AllocationCommand {
}
it.remove();
routingNode.add(shardRouting);
if (shardRouting.primary()) {
// we need to clear the post allocation flag, since its an explicit allocation of the primary shard
// and we want to force allocate it (and create a new index for it)
allocation.routingNodes().addClearPostAllocationFlag(shardRouting.shardId());
}
break;
}
}

36
src/test/java/org/elasticsearch/test/integration/cluster/allocation/ClusterRerouteTests.java
View File

@ -20,14 +20,17 @@
package org.elasticsearch.test.integration.cluster.allocation;
import org.elasticsearch.action.admin.cluster.health.ClusterHealthResponse;
import org.elasticsearch.action.admin.cluster.health.ClusterHealthStatus;
import org.elasticsearch.cluster.ClusterState;
import org.elasticsearch.cluster.routing.ShardRoutingState;
import org.elasticsearch.cluster.routing.allocation.command.AllocateAllocationCommand;
import org.elasticsearch.cluster.routing.allocation.command.MoveAllocationCommand;
import org.elasticsearch.common.Priority;
import org.elasticsearch.common.io.FileSystemUtils;
import org.elasticsearch.common.logging.ESLogger;
import org.elasticsearch.common.logging.Loggers;
import org.elasticsearch.common.settings.Settings;
import org.elasticsearch.env.NodeEnvironment;
import org.elasticsearch.gateway.Gateway;
import org.elasticsearch.index.shard.ShardId;
import org.elasticsearch.node.internal.InternalNode;
@ -35,6 +38,8 @@ import org.elasticsearch.test.integration.AbstractNodesTests;
import org.testng.annotations.AfterMethod;
import org.testng.annotations.Test;
import java.io.File;
import static org.elasticsearch.common.settings.ImmutableSettings.settingsBuilder;
import static org.hamcrest.MatcherAssert.assertThat;
import static org.hamcrest.Matchers.equalTo;
@ -161,5 +166,36 @@ public class ClusterRerouteTests extends AbstractNodesTests {
state = client("node1").admin().cluster().prepareState().execute().actionGet().getState();
assertThat(state.routingNodes().unassigned().size(), equalTo(1));
assertThat(state.routingNodes().node(state.nodes().resolveNode("node1").id()).shards().get(0).state(), equalTo(ShardRoutingState.STARTED));
client("node1").prepareIndex("test", "type", "1").setSource("field", "value").setRefresh(true).execute().actionGet();
logger.info("--> closing all nodes");
File shardLocation = ((InternalNode) node("node1")).injector().getInstance(NodeEnvironment.class).shardLocations(new ShardId("test", 0))[0];
closeAllNodes();
logger.info("--> deleting the shard data");
FileSystemUtils.deleteRecursively(shardLocation);
logger.info("--> starting the first node back, will not allocate the shard since it has no data, but the index will be there");
startNode("node1", commonSettings);
// wait a bit for the cluster to realize that the shard is not there...
// TODO can we get around this? the cluster is RED, so what do we wait for?
Thread.sleep(300);
assertThat(client("node1").admin().cluster().prepareHealth().execute().actionGet().getStatus(), equalTo(ClusterHealthStatus.RED));
logger.info("--> explicitly allocate primary");
state = client("node1").admin().cluster().prepareReroute()
.add(new AllocateAllocationCommand(new ShardId("test", 0), "node1", true))
.execute().actionGet().getState();
assertThat(state.routingNodes().unassigned().size(), equalTo(1));
assertThat(state.routingNodes().node(state.nodes().resolveNode("node1").id()).shards().get(0).state(), equalTo(ShardRoutingState.INITIALIZING));
healthResponse = client("node1").admin().cluster().prepareHealth().setWaitForEvents(Priority.LANGUID).setWaitForYellowStatus().execute().actionGet();
assertThat(healthResponse.isTimedOut(), equalTo(false));
logger.info("--> get the state, verify shard 1 primary allocated");
state = client("node1").admin().cluster().prepareState().execute().actionGet().getState();
assertThat(state.routingNodes().unassigned().size(), equalTo(1));
assertThat(state.routingNodes().node(state.nodes().resolveNode("node1").id()).shards().get(0).state(), equalTo(ShardRoutingState.STARTED));
}
}