Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Remove deprecated settings to defer cluster recovery #17357

Merged
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions CHANGELOG-3.0.md
Original file line number Diff line number Diff line change
Expand Up @@ -58,6 +58,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
- Remove `index.store.hybrid.mmap.extensions` setting in favor of `index.store.hybrid.nio.extensions` setting ([#9392](https://github.com/opensearch-project/OpenSearch/pull/9392))
- Remove package org.opensearch.action.support.master ([#4856](https://github.com/opensearch-project/OpenSearch/issues/4856))
- Remove transport-nio plugin ([#16887](https://github.com/opensearch-project/OpenSearch/issues/16887))
- Remove deprecated 'gateway' settings used to defer cluster recovery ([#3117](https://github.com/opensearch-project/OpenSearch/issues/3117))

### Fixed
- Fix 'org.apache.hc.core5.http.ParseException: Invalid protocol version' under JDK 16+ ([#4827](https://github.com/opensearch-project/OpenSearch/pull/4827))
Expand Down
2 changes: 1 addition & 1 deletion distribution/src/config/opensearch.yml
Original file line number Diff line number Diff line change
Expand Up @@ -77,7 +77,7 @@ ${path.logs}
#
# Block initial recovery after a full cluster restart until N nodes are started:
#
#gateway.recover_after_nodes: 3
#gateway.recover_after_data_nodes: 3
#
# For more information, consult the gateway module documentation.
#
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -46,6 +46,7 @@
import org.opensearch.common.settings.Settings;
import org.opensearch.common.util.concurrent.OpenSearchExecutors;
import org.opensearch.core.common.unit.ByteSizeValue;
import org.opensearch.gateway.GatewayService;
import org.opensearch.monitor.os.OsStats;
import org.opensearch.node.NodeRoleSettings;
import org.opensearch.test.OpenSearchIntegTestCase;
Expand Down Expand Up @@ -383,19 +384,18 @@ public void testAllocatedProcessors() throws Exception {
}

public void testClusterStatusWhenStateNotRecovered() throws Exception {
internalCluster().startClusterManagerOnlyNode(Settings.builder().put("gateway.recover_after_nodes", 2).build());
internalCluster().startClusterManagerOnlyNode(
Settings.builder().put(GatewayService.RECOVER_AFTER_DATA_NODES_SETTING.getKey(), 2).build()
);
ClusterStatsResponse response = client().admin()
.cluster()
.prepareClusterStats()
.useAggregatedNodeLevelResponses(randomBoolean())
.get();
assertThat(response.getStatus(), equalTo(ClusterHealthStatus.RED));

if (randomBoolean()) {
internalCluster().startClusterManagerOnlyNode();
} else {
internalCluster().startDataOnlyNode();
}
internalCluster().startDataOnlyNodes(2);

// wait for the cluster status to settle
ensureGreen();
response = client().admin().cluster().prepareClusterStats().useAggregatedNodeLevelResponses(randomBoolean()).get();
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -95,7 +95,7 @@ public void testPendingTasksWithClusterNotRecoveredBlock() throws Exception {
internalCluster().fullRestart(new InternalTestCluster.RestartCallback() {
@Override
public Settings onNodeStopped(String nodeName) {
return Settings.builder().put(GatewayService.RECOVER_AFTER_NODES_SETTING.getKey(), nodeCount + 1).build();
return Settings.builder().put(GatewayService.RECOVER_AFTER_DATA_NODES_SETTING.getKey(), nodeCount + 1).build();
}

@Override
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -49,7 +49,7 @@ public class IndicesExistsIT extends OpenSearchIntegTestCase {

public void testIndexExistsWithBlocksInPlace() throws IOException {
internalCluster().setBootstrapClusterManagerNodeIndex(0);
Settings settings = Settings.builder().put(GatewayService.RECOVER_AFTER_NODES_SETTING.getKey(), 99).build();
Settings settings = Settings.builder().put(GatewayService.RECOVER_AFTER_DATA_NODES_SETTING.getKey(), 99).build();
String node = internalCluster().startNode(settings);

assertRequestBuilderThrows(
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -77,7 +77,7 @@ public Client startNode(Settings.Builder settings) {
public void testRecoverAfterNodes() throws Exception {
internalCluster().setBootstrapClusterManagerNodeIndex(0);
logger.info("--> start node (1)");
Client clientNode1 = startNode(Settings.builder().put("gateway.recover_after_nodes", 3));
Client clientNode1 = startNode(Settings.builder().put(GatewayService.RECOVER_AFTER_DATA_NODES_SETTING.getKey(), 3));
assertThat(
clientNode1.admin()
.cluster()
Expand All @@ -92,7 +92,7 @@ public void testRecoverAfterNodes() throws Exception {
);

logger.info("--> start node (2)");
Client clientNode2 = startNode(Settings.builder().put("gateway.recover_after_nodes", 3));
Client clientNode2 = startNode(Settings.builder().put(GatewayService.RECOVER_AFTER_DATA_NODES_SETTING.getKey(), 3));
Thread.sleep(BLOCK_WAIT_TIMEOUT.millis());
assertThat(
clientNode1.admin()
Expand Down Expand Up @@ -120,104 +120,13 @@ public void testRecoverAfterNodes() throws Exception {
);

logger.info("--> start node (3)");
Client clientNode3 = startNode(Settings.builder().put("gateway.recover_after_nodes", 3));
Client clientNode3 = startNode(Settings.builder().put(GatewayService.RECOVER_AFTER_DATA_NODES_SETTING.getKey(), 3));

assertThat(waitForNoBlocksOnNode(BLOCK_WAIT_TIMEOUT, clientNode1).isEmpty(), equalTo(true));
assertThat(waitForNoBlocksOnNode(BLOCK_WAIT_TIMEOUT, clientNode2).isEmpty(), equalTo(true));
assertThat(waitForNoBlocksOnNode(BLOCK_WAIT_TIMEOUT, clientNode3).isEmpty(), equalTo(true));
}

public void testRecoverAfterClusterManagerNodes() throws Exception {
internalCluster().setBootstrapClusterManagerNodeIndex(0);
logger.info("--> start cluster_manager_node (1)");
Client clusterManager1 = startNode(Settings.builder().put("gateway.recover_after_master_nodes", 2).put(clusterManagerOnlyNode()));
assertThat(
clusterManager1.admin()
.cluster()
.prepareState()
.setLocal(true)
.execute()
.actionGet()
.getState()
.blocks()
.global(ClusterBlockLevel.METADATA_WRITE),
hasItem(GatewayService.STATE_NOT_RECOVERED_BLOCK)
);

logger.info("--> start data_node (1)");
Client data1 = startNode(Settings.builder().put("gateway.recover_after_master_nodes", 2).put(dataOnlyNode()));
assertThat(
clusterManager1.admin()
.cluster()
.prepareState()
.setLocal(true)
.execute()
.actionGet()
.getState()
.blocks()
.global(ClusterBlockLevel.METADATA_WRITE),
hasItem(GatewayService.STATE_NOT_RECOVERED_BLOCK)
);
assertThat(
data1.admin()
.cluster()
.prepareState()
.setLocal(true)
.execute()
.actionGet()
.getState()
.blocks()
.global(ClusterBlockLevel.METADATA_WRITE),
hasItem(GatewayService.STATE_NOT_RECOVERED_BLOCK)
);

logger.info("--> start data_node (2)");
Client data2 = startNode(Settings.builder().put("gateway.recover_after_master_nodes", 2).put(dataOnlyNode()));
assertThat(
clusterManager1.admin()
.cluster()
.prepareState()
.setLocal(true)
.execute()
.actionGet()
.getState()
.blocks()
.global(ClusterBlockLevel.METADATA_WRITE),
hasItem(GatewayService.STATE_NOT_RECOVERED_BLOCK)
);
assertThat(
data1.admin()
.cluster()
.prepareState()
.setLocal(true)
.execute()
.actionGet()
.getState()
.blocks()
.global(ClusterBlockLevel.METADATA_WRITE),
hasItem(GatewayService.STATE_NOT_RECOVERED_BLOCK)
);
assertThat(
data2.admin()
.cluster()
.prepareState()
.setLocal(true)
.execute()
.actionGet()
.getState()
.blocks()
.global(ClusterBlockLevel.METADATA_WRITE),
hasItem(GatewayService.STATE_NOT_RECOVERED_BLOCK)
);

logger.info("--> start cluster_manager_node (2)");
Client clusterManager2 = startNode(Settings.builder().put("gateway.recover_after_master_nodes", 2).put(clusterManagerOnlyNode()));
assertThat(waitForNoBlocksOnNode(BLOCK_WAIT_TIMEOUT, clusterManager1).isEmpty(), equalTo(true));
assertThat(waitForNoBlocksOnNode(BLOCK_WAIT_TIMEOUT, clusterManager2).isEmpty(), equalTo(true));
assertThat(waitForNoBlocksOnNode(BLOCK_WAIT_TIMEOUT, data1).isEmpty(), equalTo(true));
assertThat(waitForNoBlocksOnNode(BLOCK_WAIT_TIMEOUT, data2).isEmpty(), equalTo(true));
}

public void testRecoverAfterDataNodes() throws Exception {
internalCluster().setBootstrapClusterManagerNodeIndex(0);
logger.info("--> start cluster_manager_node (1)");
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -119,7 +119,7 @@
import static org.opensearch.gateway.GatewayRecoveryTestUtils.corruptShard;
import static org.opensearch.gateway.GatewayRecoveryTestUtils.getDiscoveryNodes;
import static org.opensearch.gateway.GatewayRecoveryTestUtils.prepareRequestMap;
import static org.opensearch.gateway.GatewayService.RECOVER_AFTER_NODES_SETTING;
import static org.opensearch.gateway.GatewayService.RECOVER_AFTER_DATA_NODES_SETTING;
import static org.opensearch.index.query.QueryBuilders.matchAllQuery;
import static org.opensearch.index.query.QueryBuilders.termQuery;
import static org.opensearch.test.hamcrest.OpenSearchAssertions.assertAcked;
Expand Down Expand Up @@ -411,7 +411,7 @@ public void testTwoNodeFirstNodeCleared() throws Exception {
@Override
public Settings onNodeStopped(String nodeName) {
return Settings.builder()
.put(RECOVER_AFTER_NODES_SETTING.getKey(), 2)
.put(RECOVER_AFTER_DATA_NODES_SETTING.getKey(), 2)
.putList(INITIAL_CLUSTER_MANAGER_NODES_SETTING.getKey()) // disable bootstrapping
.build();
}
Expand All @@ -436,7 +436,7 @@ public boolean clearData(String nodeName) {

public void testLatestVersionLoaded() throws Exception {
// clean two nodes
List<String> nodes = internalCluster().startNodes(2, Settings.builder().put("gateway.recover_after_nodes", 2).build());
List<String> nodes = internalCluster().startNodes(2, Settings.builder().put(RECOVER_AFTER_DATA_NODES_SETTING.getKey(), 2).build());
Settings node1DataPathSettings = internalCluster().dataPathSettings(nodes.get(0));
Settings node2DataPathSettings = internalCluster().dataPathSettings(nodes.get(1));

Expand Down Expand Up @@ -520,8 +520,8 @@ public void testLatestVersionLoaded() throws Exception {
logger.info("--> starting the two nodes back");

internalCluster().startNodes(
Settings.builder().put(node1DataPathSettings).put("gateway.recover_after_nodes", 2).build(),
Settings.builder().put(node2DataPathSettings).put("gateway.recover_after_nodes", 2).build()
Settings.builder().put(node1DataPathSettings).put(RECOVER_AFTER_DATA_NODES_SETTING.getKey(), 2).build(),
Settings.builder().put(node2DataPathSettings).put(RECOVER_AFTER_DATA_NODES_SETTING.getKey(), 2).build()
);

logger.info("--> running cluster_health (wait for the shards to startup)");
Expand Down Expand Up @@ -710,7 +710,7 @@ public void testStartedShardFoundIfStateNotYetProcessed() throws Exception {
@Override
public Settings onNodeStopped(String nodeName) throws Exception {
// make sure state is not recovered
return Settings.builder().put(RECOVER_AFTER_NODES_SETTING.getKey(), 2).build();
return Settings.builder().put(RECOVER_AFTER_DATA_NODES_SETTING.getKey(), 2).build();
}
});

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -346,11 +346,7 @@ public void apply(Settings value, Settings current, Settings previous) {
NoClusterManagerBlockService.NO_MASTER_BLOCK_SETTING, // deprecated
NoClusterManagerBlockService.NO_CLUSTER_MANAGER_BLOCK_SETTING,
GatewayService.EXPECTED_DATA_NODES_SETTING,
GatewayService.EXPECTED_MASTER_NODES_SETTING,
GatewayService.EXPECTED_NODES_SETTING,
GatewayService.RECOVER_AFTER_DATA_NODES_SETTING,
GatewayService.RECOVER_AFTER_MASTER_NODES_SETTING,
GatewayService.RECOVER_AFTER_NODES_SETTING,
GatewayService.RECOVER_AFTER_TIME_SETTING,
ShardsBatchGatewayAllocator.GATEWAY_ALLOCATOR_BATCH_SIZE,
ShardsBatchGatewayAllocator.PRIMARY_BATCH_ALLOCATOR_TIMEOUT_SETTING,
Expand Down
Loading
Loading