Change ClusterState and PendingClusterTasksResponse's toString() to their prettyPrint format (#21245)
The current XContent output is much harder to read than the prettyPrint format. This commit folds prettyPrint into toString and removes it.
This commit is contained in:
parent
2187b65809
commit
0daf483587
|
@ -25,7 +25,6 @@ import org.elasticsearch.common.io.stream.StreamInput;
|
|||
import org.elasticsearch.common.io.stream.StreamOutput;
|
||||
import org.elasticsearch.common.xcontent.ToXContent;
|
||||
import org.elasticsearch.common.xcontent.XContentBuilder;
|
||||
import org.elasticsearch.common.xcontent.XContentFactory;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.util.ArrayList;
|
||||
|
@ -59,7 +58,8 @@ public class PendingClusterTasksResponse extends ActionResponse implements Itera
|
|||
return pendingTasks.iterator();
|
||||
}
|
||||
|
||||
public String prettyPrint() {
|
||||
@Override
|
||||
public String toString() {
|
||||
StringBuilder sb = new StringBuilder();
|
||||
sb.append("tasks: (").append(pendingTasks.size()).append("):\n");
|
||||
for (PendingClusterTask pendingClusterTask : this) {
|
||||
|
@ -68,19 +68,6 @@ public class PendingClusterTasksResponse extends ActionResponse implements Itera
|
|||
return sb.toString();
|
||||
}
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
try {
|
||||
XContentBuilder builder = XContentFactory.jsonBuilder().prettyPrint();
|
||||
builder.startObject();
|
||||
toXContent(builder, EMPTY_PARAMS);
|
||||
builder.endObject();
|
||||
return builder.string();
|
||||
} catch (IOException e) {
|
||||
return "{ \"error\" : \"" + e.getMessage() + "\"}";
|
||||
}
|
||||
}
|
||||
|
||||
@Override
|
||||
public XContentBuilder toXContent(XContentBuilder builder, Params params) throws IOException {
|
||||
builder.startArray(Fields.TASKS);
|
||||
|
|
|
@ -274,15 +274,16 @@ public class ClusterState implements ToXContent, Diffable<ClusterState> {
|
|||
return routingNodes;
|
||||
}
|
||||
|
||||
public String prettyPrint() {
|
||||
@Override
|
||||
public String toString() {
|
||||
StringBuilder sb = new StringBuilder();
|
||||
sb.append("cluster uuid: ").append(metaData.clusterUUID()).append("\n");
|
||||
sb.append("version: ").append(version).append("\n");
|
||||
sb.append("state uuid: ").append(stateUUID).append("\n");
|
||||
sb.append("from_diff: ").append(wasReadFromDiff).append("\n");
|
||||
sb.append("meta data version: ").append(metaData.version()).append("\n");
|
||||
final String TAB = " ";
|
||||
for (IndexMetaData indexMetaData : metaData) {
|
||||
final String TAB = " ";
|
||||
sb.append(TAB).append(indexMetaData.getIndex());
|
||||
sb.append(": v[").append(indexMetaData.getVersion()).append("]\n");
|
||||
for (int shard = 0; shard < indexMetaData.getNumberOfShards(); shard++) {
|
||||
|
@ -291,24 +292,19 @@ public class ClusterState implements ToXContent, Diffable<ClusterState> {
|
|||
sb.append("isa_ids ").append(indexMetaData.inSyncAllocationIds(shard)).append("\n");
|
||||
}
|
||||
}
|
||||
sb.append(blocks().prettyPrint());
|
||||
sb.append(nodes().prettyPrint());
|
||||
sb.append(routingTable().prettyPrint());
|
||||
sb.append(getRoutingNodes().prettyPrint());
|
||||
return sb.toString();
|
||||
}
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
try {
|
||||
XContentBuilder builder = XContentFactory.jsonBuilder().prettyPrint();
|
||||
builder.startObject();
|
||||
toXContent(builder, EMPTY_PARAMS);
|
||||
builder.endObject();
|
||||
return builder.string();
|
||||
} catch (IOException e) {
|
||||
return "{ \"error\" : \"" + e.getMessage() + "\"}";
|
||||
sb.append(blocks());
|
||||
sb.append(nodes());
|
||||
sb.append(routingTable());
|
||||
sb.append(getRoutingNodes());
|
||||
if (customs.isEmpty() == false) {
|
||||
sb.append("customs:\n");
|
||||
for (ObjectObjectCursor<String, Custom> cursor : customs) {
|
||||
final String type = cursor.key;
|
||||
final Custom custom = cursor.value;
|
||||
sb.append(TAB).append(type).append(": ").append(custom);
|
||||
}
|
||||
}
|
||||
return sb.toString();
|
||||
}
|
||||
|
||||
/**
|
||||
|
|
|
@ -164,7 +164,7 @@ public class ShardStateAction extends AbstractComponent {
|
|||
@Override
|
||||
public void onNewClusterState(ClusterState state) {
|
||||
if (logger.isTraceEnabled()) {
|
||||
logger.trace("new cluster state [{}] after waiting for master election to fail shard entry [{}]", state.prettyPrint(), shardEntry);
|
||||
logger.trace("new cluster state [{}] after waiting for master election to fail shard entry [{}]", state, shardEntry);
|
||||
}
|
||||
sendShardAction(actionName, observer, shardEntry, listener);
|
||||
}
|
||||
|
|
|
@ -199,7 +199,8 @@ public class ClusterBlocks extends AbstractDiffable<ClusterBlocks> {
|
|||
return new ClusterBlockException(unmodifiableSet(blocks.collect(toSet())));
|
||||
}
|
||||
|
||||
public String prettyPrint() {
|
||||
@Override
|
||||
public String toString() {
|
||||
if (global.isEmpty() && indices().isEmpty()) {
|
||||
return "";
|
||||
}
|
||||
|
|
|
@ -26,7 +26,6 @@ import org.elasticsearch.Version;
|
|||
import org.elasticsearch.cluster.AbstractDiffable;
|
||||
import org.elasticsearch.common.Booleans;
|
||||
import org.elasticsearch.common.Nullable;
|
||||
import org.elasticsearch.common.Strings;
|
||||
import org.elasticsearch.common.collect.ImmutableOpenMap;
|
||||
import org.elasticsearch.common.io.stream.StreamInput;
|
||||
import org.elasticsearch.common.io.stream.StreamOutput;
|
||||
|
@ -398,14 +397,6 @@ public class DiscoveryNodes extends AbstractDiffable<DiscoveryNodes> implements
|
|||
|
||||
@Override
|
||||
public String toString() {
|
||||
StringBuilder sb = new StringBuilder();
|
||||
sb.append("{");
|
||||
sb.append(Strings.collectionToDelimitedString(this, ","));
|
||||
sb.append("}");
|
||||
return sb.toString();
|
||||
}
|
||||
|
||||
public String prettyPrint() {
|
||||
StringBuilder sb = new StringBuilder();
|
||||
sb.append("nodes: \n");
|
||||
for (DiscoveryNode node : this) {
|
||||
|
|
|
@ -391,7 +391,8 @@ public class RoutingNodes implements Iterable<RoutingNode> {
|
|||
return shards;
|
||||
}
|
||||
|
||||
public String prettyPrint() {
|
||||
@Override
|
||||
public String toString() {
|
||||
StringBuilder sb = new StringBuilder("routing_nodes:\n");
|
||||
for (RoutingNode routingNode : this) {
|
||||
sb.append(routingNode.prettyPrint());
|
||||
|
|
|
@ -109,7 +109,7 @@ public class RoutingService extends AbstractLifecycleComponent {
|
|||
rerouting.set(false);
|
||||
ClusterState state = clusterService.state();
|
||||
if (logger.isTraceEnabled()) {
|
||||
logger.error((Supplier<?>) () -> new ParameterizedMessage("unexpected failure during [{}], current state:\n{}", source, state.prettyPrint()), e);
|
||||
logger.error((Supplier<?>) () -> new ParameterizedMessage("unexpected failure during [{}], current state:\n{}", source, state), e);
|
||||
} else {
|
||||
logger.error((Supplier<?>) () -> new ParameterizedMessage("unexpected failure during [{}], current state version [{}]", source, state.version()), e);
|
||||
}
|
||||
|
@ -118,7 +118,7 @@ public class RoutingService extends AbstractLifecycleComponent {
|
|||
} catch (Exception e) {
|
||||
rerouting.set(false);
|
||||
ClusterState state = clusterService.state();
|
||||
logger.warn((Supplier<?>) () -> new ParameterizedMessage("failed to reroute routing table, current state:\n{}", state.prettyPrint()), e);
|
||||
logger.warn((Supplier<?>) () -> new ParameterizedMessage("failed to reroute routing table, current state:\n{}", state), e);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
|
|
@ -613,7 +613,8 @@ public class RoutingTable implements Iterable<IndexRoutingTable>, Diffable<Routi
|
|||
}
|
||||
}
|
||||
|
||||
public String prettyPrint() {
|
||||
@Override
|
||||
public String toString() {
|
||||
StringBuilder sb = new StringBuilder("routing_table (version ").append(version).append("):\n");
|
||||
for (ObjectObjectCursor<String, IndexRoutingTable> entry : indicesRouting) {
|
||||
sb.append(entry.value.prettyPrint()).append('\n');
|
||||
|
|
|
@ -233,7 +233,7 @@ public class IndexMetaDataUpdater extends RoutingChangesObserver.AbstractRouting
|
|||
Set<String> oldInSyncAllocations = oldIndexMetaData.inSyncAllocationIds(shardNumber);
|
||||
Set<String> idsToRemove = shardEntry.getValue().stream().map(e -> e.getAllocationId()).collect(Collectors.toSet());
|
||||
assert idsToRemove.stream().allMatch(id -> oldRoutingTable.getByAllocationId(shardEntry.getKey(), id) == null) :
|
||||
"removing stale ids: " + idsToRemove + ", some of which have still a routing entry: " + oldRoutingTable.prettyPrint();
|
||||
"removing stale ids: " + idsToRemove + ", some of which have still a routing entry: " + oldRoutingTable;
|
||||
Set<String> remainingInSyncAllocations = Sets.difference(oldInSyncAllocations, idsToRemove);
|
||||
assert remainingInSyncAllocations.isEmpty() == false : "Set of in-sync ids cannot become empty for shard " +
|
||||
shardEntry.getKey() + " (before: " + oldInSyncAllocations + ", ids to remove: " + idsToRemove + ")";
|
||||
|
|
|
@ -562,9 +562,9 @@ public class ClusterService extends AbstractLifecycleComponent {
|
|||
executionTime,
|
||||
previousClusterState.version(),
|
||||
tasksSummary,
|
||||
previousClusterState.nodes().prettyPrint(),
|
||||
previousClusterState.routingTable().prettyPrint(),
|
||||
previousClusterState.getRoutingNodes().prettyPrint()),
|
||||
previousClusterState.nodes(),
|
||||
previousClusterState.routingTable(),
|
||||
previousClusterState.getRoutingNodes()),
|
||||
e);
|
||||
}
|
||||
warnAboutSlowTaskIfNeeded(executionTime, tasksSummary);
|
||||
|
@ -656,7 +656,7 @@ public class ClusterService extends AbstractLifecycleComponent {
|
|||
newClusterState.status(ClusterState.ClusterStateStatus.BEING_APPLIED);
|
||||
|
||||
if (logger.isTraceEnabled()) {
|
||||
logger.trace("cluster state updated, source [{}]\n{}", tasksSummary, newClusterState.prettyPrint());
|
||||
logger.trace("cluster state updated, source [{}]\n{}", tasksSummary, newClusterState);
|
||||
} else if (logger.isDebugEnabled()) {
|
||||
logger.debug("cluster state updated, version [{}], source [{}]", newClusterState.version(), tasksSummary);
|
||||
}
|
||||
|
@ -759,7 +759,7 @@ public class ClusterService extends AbstractLifecycleComponent {
|
|||
TimeValue executionTime = TimeValue.timeValueMillis(Math.max(0, TimeValue.nsecToMSec(currentTimeInNanos() - startTimeNS)));
|
||||
final long version = newClusterState.version();
|
||||
final String stateUUID = newClusterState.stateUUID();
|
||||
final String prettyPrint = newClusterState.prettyPrint();
|
||||
final String fullState = newClusterState.toString();
|
||||
logger.warn(
|
||||
(Supplier<?>) () -> new ParameterizedMessage(
|
||||
"failed to apply updated cluster state in [{}]:\nversion [{}], uuid [{}], source [{}]\n{}",
|
||||
|
@ -767,7 +767,7 @@ public class ClusterService extends AbstractLifecycleComponent {
|
|||
version,
|
||||
stateUUID,
|
||||
tasksSummary,
|
||||
prettyPrint),
|
||||
fullState),
|
||||
e);
|
||||
// TODO: do we want to call updateTask.onFailure here?
|
||||
}
|
||||
|
@ -826,9 +826,7 @@ public class ClusterService extends AbstractLifecycleComponent {
|
|||
(Supplier<?>) () -> new ParameterizedMessage(
|
||||
"exception thrown by listener while notifying of cluster state processed from [{}], old cluster state:\n" +
|
||||
"{}\nnew cluster state:\n{}",
|
||||
source,
|
||||
oldState.prettyPrint(),
|
||||
newState.prettyPrint()),
|
||||
source, oldState, newState),
|
||||
e);
|
||||
}
|
||||
}
|
||||
|
|
|
@ -464,7 +464,7 @@ public class NodeJoinController extends AbstractComponent {
|
|||
}
|
||||
|
||||
private ClusterState.Builder becomeMasterAndTrimConflictingNodes(ClusterState currentState, List<DiscoveryNode> joiningNodes) {
|
||||
assert currentState.nodes().getMasterNodeId() == null : currentState.prettyPrint();
|
||||
assert currentState.nodes().getMasterNodeId() == null : currentState;
|
||||
DiscoveryNodes.Builder nodesBuilder = DiscoveryNodes.builder(currentState.nodes());
|
||||
nodesBuilder.masterNodeId(currentState.nodes().getLocalNodeId());
|
||||
ClusterBlocks clusterBlocks = ClusterBlocks.builder().blocks(currentState.blocks())
|
||||
|
|
|
@ -26,7 +26,6 @@ import org.elasticsearch.action.UnavailableShardsException;
|
|||
import org.elasticsearch.action.admin.indices.flush.FlushRequest;
|
||||
import org.elasticsearch.action.admin.indices.flush.FlushResponse;
|
||||
import org.elasticsearch.action.admin.indices.flush.TransportFlushAction;
|
||||
import org.elasticsearch.action.support.ActionFilter;
|
||||
import org.elasticsearch.action.support.ActionFilters;
|
||||
import org.elasticsearch.action.support.broadcast.BroadcastRequest;
|
||||
import org.elasticsearch.action.support.broadcast.BroadcastResponse;
|
||||
|
@ -101,7 +100,7 @@ public class BroadcastReplicationTests extends ESTestCase {
|
|||
transportService.start();
|
||||
transportService.acceptIncomingRequests();
|
||||
broadcastReplicationAction = new TestBroadcastReplicationAction(Settings.EMPTY, threadPool, clusterService, transportService,
|
||||
new ActionFilters(new HashSet<ActionFilter>()), new IndexNameExpressionResolver(Settings.EMPTY), null);
|
||||
new ActionFilters(new HashSet<>()), new IndexNameExpressionResolver(Settings.EMPTY), null);
|
||||
}
|
||||
|
||||
@After
|
||||
|
@ -120,7 +119,7 @@ public class BroadcastReplicationTests extends ESTestCase {
|
|||
final String index = "test";
|
||||
setState(clusterService, state(index, randomBoolean(),
|
||||
randomBoolean() ? ShardRoutingState.INITIALIZING : ShardRoutingState.UNASSIGNED, ShardRoutingState.UNASSIGNED));
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state());
|
||||
Future<BroadcastResponse> response = (broadcastReplicationAction.execute(new DummyBroadcastRequest().indices(index)));
|
||||
for (Tuple<ShardId, ActionListener<ReplicationResponse>> shardRequests : broadcastReplicationAction.capturedShardRequests) {
|
||||
if (randomBoolean()) {
|
||||
|
@ -139,11 +138,11 @@ public class BroadcastReplicationTests extends ESTestCase {
|
|||
final String index = "test";
|
||||
setState(clusterService, state(index, randomBoolean(),
|
||||
ShardRoutingState.STARTED));
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state());
|
||||
Future<BroadcastResponse> response = (broadcastReplicationAction.execute(new DummyBroadcastRequest().indices(index)));
|
||||
for (Tuple<ShardId, ActionListener<ReplicationResponse>> shardRequests : broadcastReplicationAction.capturedShardRequests) {
|
||||
ReplicationResponse replicationResponse = new ReplicationResponse();
|
||||
replicationResponse.setShardInfo(new ReplicationResponse.ShardInfo(1, 1, new ReplicationResponse.ShardInfo.Failure[0]));
|
||||
replicationResponse.setShardInfo(new ReplicationResponse.ShardInfo(1, 1));
|
||||
shardRequests.v2().onResponse(replicationResponse);
|
||||
}
|
||||
logger.info("total shards: {}, ", response.get().getTotalShards());
|
||||
|
@ -154,7 +153,7 @@ public class BroadcastReplicationTests extends ESTestCase {
|
|||
final String index = "test";
|
||||
int numShards = 1 + randomInt(3);
|
||||
setState(clusterService, stateWithAssignedPrimariesAndOneReplica(index, numShards));
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state());
|
||||
Future<BroadcastResponse> response = (broadcastReplicationAction.execute(new DummyBroadcastRequest().indices(index)));
|
||||
int succeeded = 0;
|
||||
int failed = 0;
|
||||
|
@ -184,7 +183,7 @@ public class BroadcastReplicationTests extends ESTestCase {
|
|||
|
||||
public void testNoShards() throws InterruptedException, ExecutionException, IOException {
|
||||
setState(clusterService, stateWithNoShard());
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state());
|
||||
BroadcastResponse response = executeAndAssertImmediateResponse(broadcastReplicationAction, new DummyBroadcastRequest());
|
||||
assertBroadcastResponse(0, 0, 0, response, null);
|
||||
}
|
||||
|
@ -194,7 +193,7 @@ public class BroadcastReplicationTests extends ESTestCase {
|
|||
final ShardId shardId = new ShardId(index, "_na_", 0);
|
||||
ClusterState clusterState = state(index, randomBoolean(),
|
||||
randomBoolean() ? ShardRoutingState.INITIALIZING : ShardRoutingState.UNASSIGNED, ShardRoutingState.UNASSIGNED);
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state());
|
||||
List<ShardId> shards = broadcastReplicationAction.shards(new DummyBroadcastRequest().indices(shardId.getIndexName()), clusterState);
|
||||
assertThat(shards.size(), equalTo(1));
|
||||
assertThat(shards.get(0), equalTo(shardId));
|
||||
|
|
|
@ -256,7 +256,7 @@ public class ReplicationOperationTests extends ESTestCase {
|
|||
final ClusterState initialState,
|
||||
final ClusterState changedState) throws Exception {
|
||||
AtomicReference<ClusterState> state = new AtomicReference<>(initialState);
|
||||
logger.debug("--> using initial state:\n{}", state.get().prettyPrint());
|
||||
logger.debug("--> using initial state:\n{}", state.get());
|
||||
final long primaryTerm = initialState.getMetaData().index(shardId.getIndexName()).primaryTerm(shardId.id());
|
||||
final ShardRouting primaryShard = state.get().routingTable().shardRoutingTable(shardId).primaryShard();
|
||||
final TestPrimary primary = new TestPrimary(primaryShard, primaryTerm) {
|
||||
|
@ -264,7 +264,7 @@ public class ReplicationOperationTests extends ESTestCase {
|
|||
public Result perform(Request request) throws Exception {
|
||||
Result result = super.perform(request);
|
||||
state.set(changedState);
|
||||
logger.debug("--> state after primary operation:\n{}", state.get().prettyPrint());
|
||||
logger.debug("--> state after primary operation:\n{}", state.get());
|
||||
return result;
|
||||
}
|
||||
};
|
||||
|
@ -303,8 +303,7 @@ public class ReplicationOperationTests extends ESTestCase {
|
|||
logger.debug("using active shard count of [{}], assigned shards [{}], total shards [{}]." +
|
||||
" expecting op to [{}]. using state: \n{}",
|
||||
request.waitForActiveShards(), 1 + assignedReplicas, 1 + assignedReplicas + unassignedReplicas,
|
||||
passesActiveShardCheck ? "succeed" : "retry",
|
||||
state.prettyPrint());
|
||||
passesActiveShardCheck ? "succeed" : "retry", state);
|
||||
final long primaryTerm = state.metaData().index(index).primaryTerm(shardId.id());
|
||||
final IndexShardRoutingTable shardRoutingTable = state.routingTable().index(index).shard(shardId.id());
|
||||
PlainActionFuture<TestPrimary.Result> listener = new PlainActionFuture<>();
|
||||
|
|
|
@ -77,7 +77,7 @@ import org.junit.Before;
|
|||
import org.junit.BeforeClass;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.util.Arrays;
|
||||
import java.util.Collections;
|
||||
import java.util.HashSet;
|
||||
import java.util.List;
|
||||
import java.util.Locale;
|
||||
|
@ -227,7 +227,7 @@ public class TransportReplicationActionTests extends ESTestCase {
|
|||
randomBoolean() ? ShardRoutingState.INITIALIZING : ShardRoutingState.UNASSIGNED));
|
||||
ReplicationTask task = maybeTask();
|
||||
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state());
|
||||
|
||||
Request request = new Request(shardId).timeout("1ms");
|
||||
PlainActionFuture<Response> listener = new PlainActionFuture<>();
|
||||
|
@ -246,7 +246,7 @@ public class TransportReplicationActionTests extends ESTestCase {
|
|||
assertTrue(request.isRetrySet.get());
|
||||
|
||||
setState(clusterService, state(index, true, ShardRoutingState.STARTED));
|
||||
logger.debug("--> primary assigned state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> primary assigned state:\n{}", clusterService.state());
|
||||
|
||||
final IndexShardRoutingTable shardRoutingTable = clusterService.state().routingTable().index(index).shard(shardId.id());
|
||||
final String primaryNodeId = shardRoutingTable.primaryShard().currentNodeId();
|
||||
|
@ -275,7 +275,7 @@ public class TransportReplicationActionTests extends ESTestCase {
|
|||
String relocationTargetNode = state.getRoutingTable().shardRoutingTable(shardId).primaryShard().relocatingNodeId();
|
||||
state = ClusterState.builder(state).nodes(DiscoveryNodes.builder(state.nodes()).localNodeId(relocationTargetNode)).build();
|
||||
setState(clusterService, state);
|
||||
logger.debug("--> relocation ongoing state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> relocation ongoing state:\n{}", clusterService.state());
|
||||
|
||||
Request request = new Request(shardId).timeout("1ms").routedBasedOnClusterVersion(clusterService.state().version() + 1);
|
||||
PlainActionFuture<Response> listener = new PlainActionFuture<>();
|
||||
|
@ -295,10 +295,10 @@ public class TransportReplicationActionTests extends ESTestCase {
|
|||
ShardRouting relocationTarget = clusterService.state().getRoutingTable().shardRoutingTable(shardId)
|
||||
.shardsWithState(ShardRoutingState.INITIALIZING).get(0);
|
||||
AllocationService allocationService = ESAllocationTestCase.createAllocationService();
|
||||
ClusterState updatedState = allocationService.applyStartedShards(state, Arrays.asList(relocationTarget));
|
||||
ClusterState updatedState = allocationService.applyStartedShards(state, Collections.singletonList(relocationTarget));
|
||||
|
||||
setState(clusterService, updatedState);
|
||||
logger.debug("--> relocation complete state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> relocation complete state:\n{}", clusterService.state());
|
||||
|
||||
IndexShardRoutingTable shardRoutingTable = clusterService.state().routingTable().index(index).shard(shardId.id());
|
||||
final String primaryNodeId = shardRoutingTable.primaryShard().currentNodeId();
|
||||
|
@ -315,7 +315,7 @@ public class TransportReplicationActionTests extends ESTestCase {
|
|||
// no replicas in oder to skip the replication part
|
||||
setState(clusterService, state(index, true,
|
||||
randomBoolean() ? ShardRoutingState.INITIALIZING : ShardRoutingState.UNASSIGNED));
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state());
|
||||
Request request = new Request(new ShardId("unknown_index", "_na_", 0)).timeout("1ms");
|
||||
PlainActionFuture<Response> listener = new PlainActionFuture<>();
|
||||
ReplicationTask task = maybeTask();
|
||||
|
@ -339,7 +339,7 @@ public class TransportReplicationActionTests extends ESTestCase {
|
|||
final ShardId shardId = new ShardId(index, "_na_", 0);
|
||||
// no replicas in order to skip the replication part
|
||||
setState(clusterService, stateWithActivePrimary(index, true, randomInt(3)));
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state());
|
||||
Request request = new Request(shardId);
|
||||
boolean timeout = randomBoolean();
|
||||
if (timeout) {
|
||||
|
@ -396,7 +396,7 @@ public class TransportReplicationActionTests extends ESTestCase {
|
|||
ReplicationTask task = maybeTask();
|
||||
|
||||
setState(clusterService, stateWithActivePrimary(index, randomBoolean(), 3));
|
||||
logger.debug("using state: \n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("using state: \n{}", clusterService.state());
|
||||
|
||||
final IndexShardRoutingTable shardRoutingTable = clusterService.state().routingTable().index(index).shard(shardId.id());
|
||||
final String primaryNodeId = shardRoutingTable.primaryShard().currentNodeId();
|
||||
|
@ -545,7 +545,7 @@ public class TransportReplicationActionTests extends ESTestCase {
|
|||
final String index = "test";
|
||||
final ShardId shardId = new ShardId(index, "_na_", 0);
|
||||
ClusterState state = stateWithActivePrimary(index, true, 1 + randomInt(3), randomInt(2));
|
||||
logger.info("using state: {}", state.prettyPrint());
|
||||
logger.info("using state: {}", state);
|
||||
setState(clusterService, state);
|
||||
|
||||
// check that at unknown node fails
|
||||
|
@ -651,7 +651,7 @@ public class TransportReplicationActionTests extends ESTestCase {
|
|||
// no replica, we only want to test on primary
|
||||
final ClusterState state = state(index, true, ShardRoutingState.STARTED);
|
||||
setState(clusterService, state);
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state().prettyPrint());
|
||||
logger.debug("--> using initial state:\n{}", clusterService.state());
|
||||
final ShardRouting primaryShard = state.routingTable().shardRoutingTable(shardId).primaryShard();
|
||||
Request request = new Request(shardId);
|
||||
PlainActionFuture<Response> listener = new PlainActionFuture<>();
|
||||
|
|
|
@ -219,7 +219,7 @@ public class NoMasterNodeIT extends ESIntegTestCase {
|
|||
ensureSearchable("test1", "test2");
|
||||
|
||||
ClusterStateResponse clusterState = client().admin().cluster().prepareState().get();
|
||||
logger.info("Cluster state:\n{}", clusterState.getState().prettyPrint());
|
||||
logger.info("Cluster state:\n{}", clusterState.getState());
|
||||
|
||||
internalCluster().stopRandomDataNode();
|
||||
assertTrue(awaitBusy(() -> {
|
||||
|
|
|
@ -277,9 +277,9 @@ public class ClusterStateHealthTests extends ESTestCase {
|
|||
// if the inactive primaries are due solely to recovery (not failed allocation or previously being allocated)
|
||||
// then cluster health is YELLOW, otherwise RED
|
||||
if (primaryInactiveDueToRecovery(indexName, clusterState)) {
|
||||
assertThat("clusterState is:\n" + clusterState.prettyPrint(), health.getStatus(), equalTo(ClusterHealthStatus.YELLOW));
|
||||
assertThat("clusterState is:\n" + clusterState, health.getStatus(), equalTo(ClusterHealthStatus.YELLOW));
|
||||
} else {
|
||||
assertThat("clusterState is:\n" + clusterState.prettyPrint(), health.getStatus(), equalTo(ClusterHealthStatus.RED));
|
||||
assertThat("clusterState is:\n" + clusterState, health.getStatus(), equalTo(ClusterHealthStatus.RED));
|
||||
}
|
||||
}
|
||||
}
|
||||
|
|
|
@ -135,8 +135,8 @@ public class DiscoveryNodesTests extends ESTestCase {
|
|||
|
||||
final DiscoveryNodes discoNodesA = builderA.build();
|
||||
final DiscoveryNodes discoNodesB = builderB.build();
|
||||
logger.info("nodes A: {}", discoNodesA.prettyPrint());
|
||||
logger.info("nodes B: {}", discoNodesB.prettyPrint());
|
||||
logger.info("nodes A: {}", discoNodesA);
|
||||
logger.info("nodes B: {}", discoNodesB);
|
||||
|
||||
DiscoveryNodes.Delta delta = discoNodesB.delta(discoNodesA);
|
||||
|
||||
|
|
|
@ -152,7 +152,9 @@ public class PrimaryAllocationIT extends ESIntegTestCase {
|
|||
client().admin().cluster().prepareReroute().add(new AllocateStalePrimaryAllocationCommand("test", 0, dataNodeWithNoShardCopy, true)).get();
|
||||
|
||||
logger.info("--> wait until shard is failed and becomes unassigned again");
|
||||
assertBusy(() -> assertTrue(client().admin().cluster().prepareState().get().getState().prettyPrint(), client().admin().cluster().prepareState().get().getState().getRoutingTable().index("test").allPrimaryShardsUnassigned()));
|
||||
assertBusy(() ->
|
||||
assertTrue(client().admin().cluster().prepareState().get().getState().toString(),
|
||||
client().admin().cluster().prepareState().get().getState().getRoutingTable().index("test").allPrimaryShardsUnassigned()));
|
||||
assertThat(client().admin().cluster().prepareState().get().getState().getRoutingTable().index("test").getShards().get(0).primaryShard().unassignedInfo().getReason(), equalTo(UnassignedInfo.Reason.ALLOCATION_FAILED));
|
||||
}
|
||||
|
||||
|
|
|
@ -306,7 +306,7 @@ public class UnassignedInfoTests extends ESAllocationTestCase {
|
|||
clusterState = ClusterState.builder(clusterState).nodes(DiscoveryNodes.builder(clusterState.nodes()).remove("node2")).build();
|
||||
// make sure both replicas are marked as delayed (i.e. not reallocated)
|
||||
clusterState = allocation.deassociateDeadNodes(clusterState, true, "reroute");
|
||||
assertThat(clusterState.prettyPrint(), UnassignedInfo.getNumberOfDelayedUnassigned(clusterState), equalTo(2));
|
||||
assertThat(clusterState.toString(), UnassignedInfo.getNumberOfDelayedUnassigned(clusterState), equalTo(2));
|
||||
}
|
||||
|
||||
public void testFindNextDelayedAllocation() {
|
||||
|
|
|
@ -89,7 +89,7 @@ public class AddIncrementallyTests extends ESAllocationTestCase {
|
|||
clusterState = addNodes(clusterState, service, 1, nodeOffset++);
|
||||
assertNumIndexShardsPerNode(clusterState, Matchers.lessThanOrEqualTo(2));
|
||||
assertThat(clusterState.getRoutingNodes().unassigned().size(), equalTo(0));
|
||||
logger.debug("ClusterState: {}", clusterState.getRoutingNodes().prettyPrint());
|
||||
logger.debug("ClusterState: {}", clusterState.getRoutingNodes());
|
||||
}
|
||||
|
||||
public void testMinimalRelocations() {
|
||||
|
@ -150,7 +150,7 @@ public class AddIncrementallyTests extends ESAllocationTestCase {
|
|||
newState = service.applyStartedShards(clusterState, routingNodes.shardsWithState(INITIALIZING));
|
||||
assertThat(newState, equalTo(clusterState));
|
||||
assertNumIndexShardsPerNode(clusterState, equalTo(2));
|
||||
logger.debug("ClusterState: {}", clusterState.getRoutingNodes().prettyPrint());
|
||||
logger.debug("ClusterState: {}", clusterState.getRoutingNodes());
|
||||
}
|
||||
|
||||
public void testMinimalRelocationsNoLimit() {
|
||||
|
@ -212,7 +212,7 @@ public class AddIncrementallyTests extends ESAllocationTestCase {
|
|||
newState = service.applyStartedShards(clusterState, routingNodes.shardsWithState(INITIALIZING));
|
||||
assertThat(newState, equalTo(clusterState));
|
||||
assertNumIndexShardsPerNode(clusterState, equalTo(2));
|
||||
logger.debug("ClusterState: {}", clusterState.getRoutingNodes().prettyPrint());
|
||||
logger.debug("ClusterState: {}", clusterState.getRoutingNodes());
|
||||
}
|
||||
|
||||
|
||||
|
|
|
@ -366,7 +366,7 @@ public class NodeVersionAllocationDeciderTests extends ESAllocationTestCase {
|
|||
}
|
||||
|
||||
private ClusterState stabilize(ClusterState clusterState, AllocationService service) {
|
||||
logger.trace("RoutingNodes: {}", clusterState.getRoutingNodes().prettyPrint());
|
||||
logger.trace("RoutingNodes: {}", clusterState.getRoutingNodes());
|
||||
|
||||
clusterState = service.deassociateDeadNodes(clusterState, true, "reroute");
|
||||
RoutingNodes routingNodes = clusterState.getRoutingNodes();
|
||||
|
@ -375,7 +375,7 @@ public class NodeVersionAllocationDeciderTests extends ESAllocationTestCase {
|
|||
logger.info("complete rebalancing");
|
||||
boolean changed;
|
||||
do {
|
||||
logger.trace("RoutingNodes: {}", clusterState.getRoutingNodes().prettyPrint());
|
||||
logger.trace("RoutingNodes: {}", clusterState.getRoutingNodes());
|
||||
ClusterState newState = service.applyStartedShards(clusterState, routingNodes.shardsWithState(INITIALIZING));
|
||||
changed = newState.equals(clusterState) == false;
|
||||
clusterState = newState;
|
||||
|
@ -386,7 +386,7 @@ public class NodeVersionAllocationDeciderTests extends ESAllocationTestCase {
|
|||
}
|
||||
|
||||
private void assertRecoveryNodeVersions(RoutingNodes routingNodes) {
|
||||
logger.trace("RoutingNodes: {}", routingNodes.prettyPrint());
|
||||
logger.trace("RoutingNodes: {}", routingNodes);
|
||||
|
||||
List<ShardRouting> mutableShardRoutings = routingNodes.shardsWithState(ShardRoutingState.RELOCATING);
|
||||
for (ShardRouting r : mutableShardRoutings) {
|
||||
|
|
|
@ -159,7 +159,7 @@ public class RandomAllocationDeciderTests extends ESAllocationTestCase {
|
|||
|
||||
} while (clusterState.getRoutingNodes().shardsWithState(ShardRoutingState.INITIALIZING).size() != 0 ||
|
||||
clusterState.getRoutingNodes().shardsWithState(ShardRoutingState.UNASSIGNED).size() != 0 && iterations < 200);
|
||||
logger.info("Done Balancing after [{}] iterations. State:\n{}", iterations, clusterState.prettyPrint());
|
||||
logger.info("Done Balancing after [{}] iterations. State:\n{}", iterations, clusterState);
|
||||
// we stop after 200 iterations if it didn't stabelize by then something is likely to be wrong
|
||||
assertThat("max num iteration exceeded", iterations, Matchers.lessThan(200));
|
||||
assertThat(clusterState.getRoutingNodes().shardsWithState(ShardRoutingState.INITIALIZING).size(), equalTo(0));
|
||||
|
|
|
@ -70,15 +70,14 @@ public class StartedShardsRoutingTests extends ESAllocationTestCase {
|
|||
logger.info("--> test starting of shard");
|
||||
|
||||
ClusterState newState = allocation.applyStartedShards(state, Arrays.asList(initShard));
|
||||
assertThat("failed to start " + initShard + "\ncurrent routing table:" + newState.routingTable().prettyPrint(),
|
||||
newState, not(equalTo(state)));
|
||||
assertTrue(initShard + "isn't started \ncurrent routing table:" + newState.routingTable().prettyPrint(),
|
||||
assertThat("failed to start " + initShard + "\ncurrent routing table:" + newState.routingTable(), newState, not(equalTo(state)));
|
||||
assertTrue(initShard + "isn't started \ncurrent routing table:" + newState.routingTable(),
|
||||
newState.routingTable().index("test").shard(initShard.id()).allShardsStarted());
|
||||
state = newState;
|
||||
|
||||
logger.info("--> testing starting of relocating shards");
|
||||
newState = allocation.applyStartedShards(state, Arrays.asList(relocatingShard.getTargetRelocatingShard()));
|
||||
assertThat("failed to start " + relocatingShard + "\ncurrent routing table:" + newState.routingTable().prettyPrint(),
|
||||
assertThat("failed to start " + relocatingShard + "\ncurrent routing table:" + newState.routingTable(),
|
||||
newState, not(equalTo(state)));
|
||||
ShardRouting shardRouting = newState.routingTable().index("test").shard(relocatingShard.id()).getShards().get(0);
|
||||
assertThat(shardRouting.state(), equalTo(ShardRoutingState.STARTED));
|
||||
|
|
|
@ -22,6 +22,7 @@ package org.elasticsearch.cluster.serialization;
|
|||
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;
|
||||
|
@ -30,7 +31,6 @@ import org.elasticsearch.cluster.routing.allocation.AllocationService;
|
|||
import org.elasticsearch.common.io.stream.BytesStreamOutput;
|
||||
import org.elasticsearch.common.io.stream.StreamInput;
|
||||
import org.elasticsearch.common.settings.Settings;
|
||||
import org.elasticsearch.cluster.ESAllocationTestCase;
|
||||
|
||||
import static org.hamcrest.Matchers.equalTo;
|
||||
|
||||
|
@ -56,7 +56,7 @@ public class ClusterSerializationTests extends ESAllocationTestCase {
|
|||
|
||||
assertThat(serializedClusterState.getClusterName().value(), equalTo(clusterState.getClusterName().value()));
|
||||
|
||||
assertThat(serializedClusterState.routingTable().prettyPrint(), equalTo(clusterState.routingTable().prettyPrint()));
|
||||
assertThat(serializedClusterState.routingTable().toString(), equalTo(clusterState.routingTable().toString()));
|
||||
}
|
||||
|
||||
public void testRoutingTableSerialization() throws Exception {
|
||||
|
@ -81,7 +81,7 @@ public class ClusterSerializationTests extends ESAllocationTestCase {
|
|||
StreamInput inStream = outStream.bytes().streamInput();
|
||||
RoutingTable target = RoutingTable.Builder.readFrom(inStream);
|
||||
|
||||
assertThat(target.prettyPrint(), equalTo(source.prettyPrint()));
|
||||
assertThat(target.toString(), equalTo(source.toString()));
|
||||
}
|
||||
|
||||
}
|
||||
|
|
|
@ -21,6 +21,7 @@ package org.elasticsearch.cluster.serialization;
|
|||
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.IndexTemplateMetaData;
|
||||
import org.elasticsearch.cluster.metadata.MetaData;
|
||||
|
@ -28,8 +29,8 @@ import org.elasticsearch.cluster.node.DiscoveryNode;
|
|||
import org.elasticsearch.cluster.node.DiscoveryNodes;
|
||||
import org.elasticsearch.cluster.routing.RoutingTable;
|
||||
import org.elasticsearch.cluster.routing.allocation.AllocationService;
|
||||
import org.elasticsearch.common.Strings;
|
||||
import org.elasticsearch.common.settings.Settings;
|
||||
import org.elasticsearch.cluster.ESAllocationTestCase;
|
||||
|
||||
import static java.util.Collections.emptyMap;
|
||||
import static java.util.Collections.emptySet;
|
||||
|
@ -55,7 +56,7 @@ public class ClusterStateToStringTests extends ESAllocationTestCase {
|
|||
AllocationService strategy = createAllocationService();
|
||||
clusterState = ClusterState.builder(clusterState).routingTable(strategy.reroute(clusterState, "reroute").routingTable()).build();
|
||||
|
||||
String clusterStateString = clusterState.toString();
|
||||
String clusterStateString = Strings.toString(clusterState, true);
|
||||
assertNotNull(clusterStateString);
|
||||
|
||||
assertThat(clusterStateString, containsString("test_idx"));
|
||||
|
|
|
@ -381,7 +381,7 @@ public class DiscoveryWithServiceDisruptionsIT extends ESIntegTestCase {
|
|||
}
|
||||
if (!success) {
|
||||
fail("node [" + node + "] has no master or has blocks, despite of being on the right side of the partition. State dump:\n"
|
||||
+ nodeState.prettyPrint());
|
||||
+ nodeState);
|
||||
}
|
||||
}
|
||||
|
||||
|
@ -468,13 +468,13 @@ public class DiscoveryWithServiceDisruptionsIT extends ESIntegTestCase {
|
|||
assertEquals("unequal node count", state.nodes().getSize(), nodeState.nodes().getSize());
|
||||
assertEquals("different masters ", state.nodes().getMasterNodeId(), nodeState.nodes().getMasterNodeId());
|
||||
assertEquals("different meta data version", state.metaData().version(), nodeState.metaData().version());
|
||||
if (!state.routingTable().prettyPrint().equals(nodeState.routingTable().prettyPrint())) {
|
||||
if (!state.routingTable().toString().equals(nodeState.routingTable().toString())) {
|
||||
fail("different routing");
|
||||
}
|
||||
} catch (AssertionError t) {
|
||||
fail("failed comparing cluster state: " + t.getMessage() + "\n" +
|
||||
"--- cluster state of node [" + nodes.get(0) + "]: ---\n" + state.prettyPrint() +
|
||||
"\n--- cluster state [" + node + "]: ---\n" + nodeState.prettyPrint());
|
||||
"--- cluster state of node [" + nodes.get(0) + "]: ---\n" + state +
|
||||
"\n--- cluster state [" + node + "]: ---\n" + nodeState);
|
||||
}
|
||||
|
||||
}
|
||||
|
@ -1267,7 +1267,7 @@ public class DiscoveryWithServiceDisruptionsIT extends ESIntegTestCase {
|
|||
|
||||
final ClusterState state = client().admin().cluster().prepareState().get().getState();
|
||||
if (state.metaData().hasIndex("test") == false) {
|
||||
fail("index 'test' was lost. current cluster state: " + state.prettyPrint());
|
||||
fail("index 'test' was lost. current cluster state: " + state);
|
||||
}
|
||||
|
||||
}
|
||||
|
@ -1368,7 +1368,7 @@ public class DiscoveryWithServiceDisruptionsIT extends ESIntegTestCase {
|
|||
assertBusy(() -> {
|
||||
for (String node : nodes) {
|
||||
ClusterState state = getNodeClusterState(node);
|
||||
String failMsgSuffix = "cluster_state:\n" + state.prettyPrint();
|
||||
String failMsgSuffix = "cluster_state:\n" + state;
|
||||
assertThat("wrong node count on [" + node + "]. " + failMsgSuffix, state.nodes().getSize(), equalTo(nodes.size()));
|
||||
String otherMasterNodeName = state.nodes().getMasterNode() != null ? state.nodes().getMasterNode().getName() : null;
|
||||
assertThat("wrong master on node [" + node + "]. " + failMsgSuffix, otherMasterNodeName, equalTo(masterNode));
|
||||
|
|
|
@ -66,7 +66,7 @@ public class ZenUnicastDiscoveryIT extends ESIntegTestCase {
|
|||
internalCluster().startNodesAsync(currentNumNodes - unicastHostOrdinals.length).get();
|
||||
|
||||
if (client().admin().cluster().prepareHealth().setWaitForNodes("" + currentNumNodes).get().isTimedOut()) {
|
||||
logger.info("cluster forming timed out, cluster state:\n{}", client().admin().cluster().prepareState().get().getState().prettyPrint());
|
||||
logger.info("cluster forming timed out, cluster state:\n{}", client().admin().cluster().prepareState().get().getState());
|
||||
fail("timed out waiting for cluster to form with [" + currentNumNodes + "] nodes");
|
||||
}
|
||||
}
|
||||
|
|
|
@ -676,10 +676,10 @@ public class NodeJoinControllerTests extends ESTestCase {
|
|||
|
||||
protected void assertNodesInCurrentState(List<DiscoveryNode> expectedNodes) {
|
||||
final ClusterState state = clusterService.state();
|
||||
logger.info("assert for [{}] in:\n{}", expectedNodes, state.prettyPrint());
|
||||
logger.info("assert for [{}] in:\n{}", expectedNodes, state);
|
||||
DiscoveryNodes discoveryNodes = state.nodes();
|
||||
for (DiscoveryNode node : expectedNodes) {
|
||||
assertThat("missing " + node + "\n" + discoveryNodes.prettyPrint(), discoveryNodes.get(node.getId()), equalTo(node));
|
||||
assertThat("missing " + node + "\n" + discoveryNodes, discoveryNodes.get(node.getId()), equalTo(node));
|
||||
}
|
||||
assertThat(discoveryNodes.getSize(), equalTo(expectedNodes.size()));
|
||||
}
|
||||
|
|
|
@ -41,8 +41,6 @@ import org.elasticsearch.common.settings.ClusterSettings;
|
|||
import org.elasticsearch.common.settings.Settings;
|
||||
import org.elasticsearch.discovery.Discovery;
|
||||
import org.elasticsearch.discovery.DiscoverySettings;
|
||||
import org.elasticsearch.discovery.zen.DiscoveryNodesProvider;
|
||||
import org.elasticsearch.discovery.zen.PublishClusterStateAction;
|
||||
import org.elasticsearch.env.NodeEnvironment;
|
||||
import org.elasticsearch.node.Node;
|
||||
import org.elasticsearch.test.ESTestCase;
|
||||
|
@ -852,8 +850,8 @@ public class PublishClusterStateActionTests extends ESTestCase {
|
|||
|
||||
void assertSameState(ClusterState actual, ClusterState expected) {
|
||||
assertThat(actual, notNullValue());
|
||||
final String reason = "\n--> actual ClusterState: " + actual.prettyPrint() + "\n" +
|
||||
"--> expected ClusterState:" + expected.prettyPrint();
|
||||
final String reason = "\n--> actual ClusterState: " + actual + "\n" +
|
||||
"--> expected ClusterState:" + expected;
|
||||
assertThat("unequal UUIDs" + reason, actual.stateUUID(), equalTo(expected.stateUUID()));
|
||||
assertThat("unequal versions" + reason, actual.version(), equalTo(expected.version()));
|
||||
}
|
||||
|
|
|
@ -60,8 +60,8 @@ import org.elasticsearch.index.shard.IndexEventListener;
|
|||
import org.elasticsearch.index.shard.IndexShard;
|
||||
import org.elasticsearch.index.shard.IndexShardState;
|
||||
import org.elasticsearch.index.shard.ShardId;
|
||||
import org.elasticsearch.indices.recovery.RecoveryFileChunkRequest;
|
||||
import org.elasticsearch.indices.recovery.PeerRecoveryTargetService;
|
||||
import org.elasticsearch.indices.recovery.RecoveryFileChunkRequest;
|
||||
import org.elasticsearch.monitor.fs.FsInfo;
|
||||
import org.elasticsearch.node.Node;
|
||||
import org.elasticsearch.plugins.Plugin;
|
||||
|
@ -178,7 +178,8 @@ public class CorruptedFileIT extends ESIntegTestCase {
|
|||
.timeout("5m") // sometimes due to cluster rebalacing and random settings default timeout is just not enough.
|
||||
.waitForNoRelocatingShards(true)).actionGet();
|
||||
if (health.isTimedOut()) {
|
||||
logger.info("cluster state:\n{}\n{}", client().admin().cluster().prepareState().get().getState().prettyPrint(), client().admin().cluster().preparePendingClusterTasks().get().prettyPrint());
|
||||
logger.info("cluster state:\n{}\n{}",
|
||||
client().admin().cluster().prepareState().get().getState(), client().admin().cluster().preparePendingClusterTasks().get());
|
||||
assertThat("timed out waiting for green state", health.isTimedOut(), equalTo(false));
|
||||
}
|
||||
assertThat(health.getStatus(), equalTo(ClusterHealthStatus.GREEN));
|
||||
|
@ -284,7 +285,8 @@ public class CorruptedFileIT extends ESIntegTestCase {
|
|||
.health(Requests.clusterHealthRequest("test")).get();
|
||||
if (response.getStatus() != ClusterHealthStatus.RED) {
|
||||
logger.info("Cluster turned red in busy loop: {}", didClusterTurnRed);
|
||||
logger.info("cluster state:\n{}\n{}", client().admin().cluster().prepareState().get().getState().prettyPrint(), client().admin().cluster().preparePendingClusterTasks().get().prettyPrint());
|
||||
logger.info("cluster state:\n{}\n{}",
|
||||
client().admin().cluster().prepareState().get().getState(), client().admin().cluster().preparePendingClusterTasks().get());
|
||||
}
|
||||
assertThat(response.getStatus(), is(ClusterHealthStatus.RED));
|
||||
ClusterState state = client().admin().cluster().prepareState().get().getState();
|
||||
|
@ -445,7 +447,8 @@ public class CorruptedFileIT extends ESIntegTestCase {
|
|||
ClusterHealthResponse actionGet = client().admin().cluster()
|
||||
.health(Requests.clusterHealthRequest("test").waitForGreenStatus()).actionGet();
|
||||
if (actionGet.isTimedOut()) {
|
||||
logger.info("ensureGreen timed out, cluster state:\n{}\n{}", client().admin().cluster().prepareState().get().getState().prettyPrint(), client().admin().cluster().preparePendingClusterTasks().get().prettyPrint());
|
||||
logger.info("ensureGreen timed out, cluster state:\n{}\n{}",
|
||||
client().admin().cluster().prepareState().get().getState(), client().admin().cluster().preparePendingClusterTasks().get());
|
||||
assertThat("timed out waiting for green state", actionGet.isTimedOut(), equalTo(false));
|
||||
}
|
||||
// we are green so primaries got not corrupted.
|
||||
|
|
|
@ -91,7 +91,7 @@ public class IndicesClusterStateServiceRandomUpdatesTests extends AbstractIndice
|
|||
} catch (AssertionError error) {
|
||||
ClusterState finalState = state;
|
||||
logger.error((org.apache.logging.log4j.util.Supplier<?>) () ->
|
||||
new ParameterizedMessage("failed to random change state. last good state: \n{}", finalState.prettyPrint()), error);
|
||||
new ParameterizedMessage("failed to random change state. last good state: \n{}", finalState), error);
|
||||
throw error;
|
||||
}
|
||||
}
|
||||
|
@ -107,7 +107,7 @@ public class IndicesClusterStateServiceRandomUpdatesTests extends AbstractIndice
|
|||
} catch (AssertionError error) {
|
||||
logger.error((org.apache.logging.log4j.util.Supplier<?>) () -> new ParameterizedMessage(
|
||||
"failed to apply change on [{}].\n *** Previous state ***\n{}\n *** New state ***\n{}",
|
||||
node, event.previousState().prettyPrint(), event.state().prettyPrint()), error);
|
||||
node, event.previousState(), event.state()), error);
|
||||
throw error;
|
||||
}
|
||||
|
||||
|
@ -117,7 +117,7 @@ public class IndicesClusterStateServiceRandomUpdatesTests extends AbstractIndice
|
|||
}
|
||||
|
||||
// TODO: check if we can go to green by starting all shards and finishing all iterations
|
||||
logger.info("Final cluster state: {}", state.prettyPrint());
|
||||
logger.info("Final cluster state: {}", state);
|
||||
}
|
||||
|
||||
/**
|
||||
|
|
|
@ -149,7 +149,8 @@ public class FullRollingRestartIT extends ESIntegTestCase {
|
|||
ClusterState state = client().admin().cluster().prepareState().get().getState();
|
||||
RecoveryResponse recoveryResponse = client().admin().indices().prepareRecoveries("test").get();
|
||||
for (RecoveryState recoveryState : recoveryResponse.shardRecoveryStates().get("test")) {
|
||||
assertTrue("relocated from: " + recoveryState.getSourceNode() + " to: " + recoveryState.getTargetNode() + "\n" + state.prettyPrint(), recoveryState.getRecoverySource().getType() != RecoverySource.Type.PEER || recoveryState.getPrimary() == false);
|
||||
assertTrue("relocated from: " + recoveryState.getSourceNode() + " to: " + recoveryState.getTargetNode() + "\n" + state,
|
||||
recoveryState.getRecoverySource().getType() != RecoverySource.Type.PEER || recoveryState.getPrimary() == false);
|
||||
}
|
||||
internalCluster().restartRandomDataNode();
|
||||
ensureGreen();
|
||||
|
@ -157,7 +158,8 @@ public class FullRollingRestartIT extends ESIntegTestCase {
|
|||
|
||||
recoveryResponse = client().admin().indices().prepareRecoveries("test").get();
|
||||
for (RecoveryState recoveryState : recoveryResponse.shardRecoveryStates().get("test")) {
|
||||
assertTrue("relocated from: " + recoveryState.getSourceNode() + " to: " + recoveryState.getTargetNode()+ "-- \nbefore: \n" + state.prettyPrint() + "\nafter: \n" + afterState.prettyPrint(), recoveryState.getRecoverySource().getType() != RecoverySource.Type.PEER || recoveryState.getPrimary() == false);
|
||||
assertTrue("relocated from: " + recoveryState.getSourceNode() + " to: " + recoveryState.getTargetNode()+ "-- \nbefore: \n" + state,
|
||||
recoveryState.getRecoverySource().getType() != RecoverySource.Type.PEER || recoveryState.getPrimary() == false);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
|
|
@ -146,7 +146,7 @@ public abstract class ESAllocationTestCase extends ESTestCase {
|
|||
ClusterState lastClusterState;
|
||||
do {
|
||||
lastClusterState = clusterState;
|
||||
logger.debug("ClusterState: {}", clusterState.getRoutingNodes().prettyPrint());
|
||||
logger.debug("ClusterState: {}", clusterState.getRoutingNodes());
|
||||
clusterState = service.applyStartedShards(clusterState, clusterState.getRoutingNodes().shardsWithState(INITIALIZING));
|
||||
} while (lastClusterState.equals(clusterState) == false);
|
||||
return clusterState;
|
||||
|
|
|
@ -164,7 +164,6 @@ import java.util.concurrent.atomic.AtomicInteger;
|
|||
import java.util.concurrent.atomic.AtomicLong;
|
||||
import java.util.function.BooleanSupplier;
|
||||
import java.util.function.Function;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
import static org.elasticsearch.client.Requests.syncedFlushRequest;
|
||||
import static org.elasticsearch.cluster.metadata.IndexMetaData.SETTING_NUMBER_OF_REPLICAS;
|
||||
|
@ -536,9 +535,7 @@ public abstract class ESIntegTestCase extends ESTestCase {
|
|||
for (Discovery discovery : internalCluster().getInstances(Discovery.class)) {
|
||||
if (discovery instanceof ZenDiscovery) {
|
||||
final ZenDiscovery zenDiscovery = (ZenDiscovery) discovery;
|
||||
assertBusy(() -> assertThat(zenDiscovery.localNode().getName() + " still having pending states: \n" +
|
||||
Arrays.stream(zenDiscovery.pendingClusterStates()).map(ClusterState::prettyPrint)
|
||||
.collect(Collectors.joining("\n")),
|
||||
assertBusy(() -> assertThat(zenDiscovery.localNode().getName() + " still having pending states",
|
||||
zenDiscovery.pendingClusterStates(), emptyArray()));
|
||||
}
|
||||
}
|
||||
|
@ -758,17 +755,14 @@ public abstract class ESIntegTestCase extends ESTestCase {
|
|||
*/
|
||||
public void waitNoPendingTasksOnAll() throws Exception {
|
||||
assertNoTimeout(client().admin().cluster().prepareHealth().setWaitForEvents(Priority.LANGUID).get());
|
||||
assertBusy(new Runnable() {
|
||||
@Override
|
||||
public void run() {
|
||||
for (Client client : clients()) {
|
||||
ClusterHealthResponse clusterHealth = client.admin().cluster().prepareHealth().setLocal(true).get();
|
||||
assertThat("client " + client + " still has in flight fetch", clusterHealth.getNumberOfInFlightFetch(), equalTo(0));
|
||||
PendingClusterTasksResponse pendingTasks = client.admin().cluster().preparePendingClusterTasks().setLocal(true).get();
|
||||
assertThat("client " + client + " still has pending tasks " + pendingTasks.prettyPrint(), pendingTasks, Matchers.emptyIterable());
|
||||
clusterHealth = client.admin().cluster().prepareHealth().setLocal(true).get();
|
||||
assertThat("client " + client + " still has in flight fetch", clusterHealth.getNumberOfInFlightFetch(), equalTo(0));
|
||||
}
|
||||
assertBusy(() -> {
|
||||
for (Client client : clients()) {
|
||||
ClusterHealthResponse clusterHealth = client.admin().cluster().prepareHealth().setLocal(true).get();
|
||||
assertThat("client " + client + " still has in flight fetch", clusterHealth.getNumberOfInFlightFetch(), equalTo(0));
|
||||
PendingClusterTasksResponse pendingTasks = client.admin().cluster().preparePendingClusterTasks().setLocal(true).get();
|
||||
assertThat("client " + client + " still has pending tasks " + pendingTasks, pendingTasks, Matchers.emptyIterable());
|
||||
clusterHealth = client.admin().cluster().prepareHealth().setLocal(true).get();
|
||||
assertThat("client " + client + " still has in flight fetch", clusterHealth.getNumberOfInFlightFetch(), equalTo(0));
|
||||
}
|
||||
});
|
||||
assertNoTimeout(client().admin().cluster().prepareHealth().setWaitForEvents(Priority.LANGUID).get());
|
||||
|
@ -872,7 +866,8 @@ public abstract class ESIntegTestCase extends ESTestCase {
|
|||
ClusterHealthResponse actionGet = client().admin().cluster()
|
||||
.health(Requests.clusterHealthRequest(indices).timeout(timeout).waitForGreenStatus().waitForEvents(Priority.LANGUID).waitForNoRelocatingShards(true)).actionGet();
|
||||
if (actionGet.isTimedOut()) {
|
||||
logger.info("ensureGreen timed out, cluster state:\n{}\n{}", client().admin().cluster().prepareState().get().getState().prettyPrint(), client().admin().cluster().preparePendingClusterTasks().get().prettyPrint());
|
||||
logger.info("ensureGreen timed out, cluster state:\n{}\n{}",
|
||||
client().admin().cluster().prepareState().get().getState(), client().admin().cluster().preparePendingClusterTasks().get());
|
||||
fail("timed out waiting for green state");
|
||||
}
|
||||
assertThat(actionGet.getStatus(), equalTo(ClusterHealthStatus.GREEN));
|
||||
|
@ -899,7 +894,8 @@ public abstract class ESIntegTestCase extends ESTestCase {
|
|||
ClusterHealthResponse actionGet = client().admin().cluster()
|
||||
.health(request).actionGet();
|
||||
if (actionGet.isTimedOut()) {
|
||||
logger.info("waitForRelocation timed out (status={}), cluster state:\n{}\n{}", status, client().admin().cluster().prepareState().get().getState().prettyPrint(), client().admin().cluster().preparePendingClusterTasks().get().prettyPrint());
|
||||
logger.info("waitForRelocation timed out (status={}), cluster state:\n{}\n{}", status,
|
||||
client().admin().cluster().prepareState().get().getState(), client().admin().cluster().preparePendingClusterTasks().get());
|
||||
assertThat("timed out waiting for relocation", actionGet.isTimedOut(), equalTo(false));
|
||||
}
|
||||
if (status != null) {
|
||||
|
@ -996,7 +992,8 @@ public abstract class ESIntegTestCase extends ESTestCase {
|
|||
ClusterHealthResponse actionGet = client().admin().cluster()
|
||||
.health(Requests.clusterHealthRequest(indices).waitForNoRelocatingShards(true).waitForYellowStatus().waitForEvents(Priority.LANGUID)).actionGet();
|
||||
if (actionGet.isTimedOut()) {
|
||||
logger.info("ensureYellow timed out, cluster state:\n{}\n{}", client().admin().cluster().prepareState().get().getState().prettyPrint(), client().admin().cluster().preparePendingClusterTasks().get().prettyPrint());
|
||||
logger.info("ensureYellow timed out, cluster state:\n{}\n{}",
|
||||
client().admin().cluster().prepareState().get().getState(), client().admin().cluster().preparePendingClusterTasks().get());
|
||||
assertThat("timed out waiting for yellow", actionGet.isTimedOut(), equalTo(false));
|
||||
}
|
||||
logger.debug("indices {} are yellow", indices.length == 0 ? "[_all]" : indices);
|
||||
|
@ -1007,7 +1004,8 @@ public abstract class ESIntegTestCase extends ESTestCase {
|
|||
* Prints the current cluster state as debug logging.
|
||||
*/
|
||||
public void logClusterState() {
|
||||
logger.debug("cluster state:\n{}\n{}", client().admin().cluster().prepareState().get().getState().prettyPrint(), client().admin().cluster().preparePendingClusterTasks().get().prettyPrint());
|
||||
logger.debug("cluster state:\n{}\n{}",
|
||||
client().admin().cluster().prepareState().get().getState(), client().admin().cluster().preparePendingClusterTasks().get());
|
||||
}
|
||||
|
||||
/**
|
||||
|
@ -1108,7 +1106,7 @@ public abstract class ESIntegTestCase extends ESTestCase {
|
|||
if (clusterHealthResponse.isTimedOut()) {
|
||||
ClusterStateResponse stateResponse = client(viaNode).admin().cluster().prepareState().get();
|
||||
fail("failed to reach a stable cluster of [" + nodeCount + "] nodes. Tried via [" + viaNode + "]. last cluster state:\n"
|
||||
+ stateResponse.getState().prettyPrint());
|
||||
+ stateResponse.getState());
|
||||
}
|
||||
assertThat(clusterHealthResponse.isTimedOut(), is(false));
|
||||
}
|
||||
|
|
|
@ -318,7 +318,8 @@ public abstract class ESSingleNodeTestCase extends ESTestCase {
|
|||
ClusterHealthResponse actionGet = client().admin().cluster()
|
||||
.health(Requests.clusterHealthRequest(indices).timeout(timeout).waitForGreenStatus().waitForEvents(Priority.LANGUID).waitForNoRelocatingShards(true)).actionGet();
|
||||
if (actionGet.isTimedOut()) {
|
||||
logger.info("ensureGreen timed out, cluster state:\n{}\n{}", client().admin().cluster().prepareState().get().getState().prettyPrint(), client().admin().cluster().preparePendingClusterTasks().get().prettyPrint());
|
||||
logger.info("ensureGreen timed out, cluster state:\n{}\n{}", client().admin().cluster().prepareState().get().getState(),
|
||||
client().admin().cluster().preparePendingClusterTasks().get());
|
||||
assertThat("timed out waiting for green state", actionGet.isTimedOut(), equalTo(false));
|
||||
}
|
||||
assertThat(actionGet.getStatus(), equalTo(ClusterHealthStatus.GREEN));
|
||||
|
|
Loading…
Reference in New Issue