001/*
002 * Licensed to the Apache Software Foundation (ASF) under one
003 * or more contributor license agreements.  See the NOTICE file
004 * distributed with this work for additional information
005 * regarding copyright ownership.  The ASF licenses this file
006 * to you under the Apache License, Version 2.0 (the
007 * "License"); you may not use this file except in compliance
008 * with the License.  You may obtain a copy of the License at
009 *
010 *     http://www.apache.org/licenses/LICENSE-2.0
011 *
012 * Unless required by applicable law or agreed to in writing, software
013 * distributed under the License is distributed on an "AS IS" BASIS,
014 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
015 * See the License for the specific language governing permissions and
016 * limitations under the License.
017 */
018package org.apache.hadoop.hbase.master;
019
020import static org.junit.jupiter.api.Assertions.assertEquals;
021import static org.junit.jupiter.api.Assertions.fail;
022
023import java.io.IOException;
024import java.util.AbstractMap.SimpleImmutableEntry;
025import java.util.List;
026import org.apache.hadoop.hbase.HBaseTestingUtil;
027import org.apache.hadoop.hbase.TableName;
028import org.apache.hadoop.hbase.client.ColumnFamilyDescriptorBuilder;
029import org.apache.hadoop.hbase.client.RegionInfo;
030import org.apache.hadoop.hbase.client.TableDescriptor;
031import org.apache.hadoop.hbase.client.TableDescriptorBuilder;
032import org.apache.hadoop.hbase.master.assignment.RegionStates;
033import org.apache.hadoop.hbase.quotas.SpaceQuotaSnapshot;
034import org.apache.hadoop.hbase.quotas.SpaceQuotaSnapshot.SpaceQuotaStatus;
035import org.apache.hadoop.hbase.quotas.SpaceViolationPolicy;
036import org.apache.hadoop.hbase.testclassification.MasterTests;
037import org.apache.hadoop.hbase.testclassification.MediumTests;
038import org.apache.hadoop.hbase.util.Bytes;
039import org.apache.hadoop.hbase.util.EnvironmentEdgeManager;
040import org.apache.hadoop.hbase.util.PairOfSameType;
041import org.apache.hadoop.hbase.util.Threads;
042import org.junit.jupiter.api.AfterAll;
043import org.junit.jupiter.api.BeforeAll;
044import org.junit.jupiter.api.Tag;
045import org.junit.jupiter.api.Test;
046import org.slf4j.Logger;
047import org.slf4j.LoggerFactory;
048
049@Tag(MasterTests.TAG)
050@Tag(MediumTests.TAG)
051public class TestMasterMetricsWrapper {
052
053  private static final Logger LOG = LoggerFactory.getLogger(TestMasterMetricsWrapper.class);
054
055  private static final HBaseTestingUtil TEST_UTIL = new HBaseTestingUtil();
056  private static final int NUM_RS = 4;
057
058  @BeforeAll
059  public static void setup() throws Exception {
060    TEST_UTIL.startMiniCluster(NUM_RS);
061  }
062
063  @AfterAll
064  public static void teardown() throws Exception {
065    TEST_UTIL.shutdownMiniCluster();
066  }
067
068  @Test
069  public void testInfo() throws IOException {
070    HMaster master = TEST_UTIL.getHBaseCluster().getMaster();
071    MetricsMasterWrapperImpl info = new MetricsMasterWrapperImpl(master);
072    assertEquals(master.getRegionNormalizerManager().getSplitPlanCount(), info.getSplitPlanCount(),
073      0);
074    assertEquals(master.getRegionNormalizerManager().getMergePlanCount(), info.getMergePlanCount(),
075      0);
076    assertEquals(master.getAverageLoad(), info.getAverageLoad(), 0);
077    assertEquals(master.getClusterId(), info.getClusterId());
078    assertEquals(master.getMasterActiveTime(), info.getActiveTime());
079    assertEquals(master.getMasterStartTime(), info.getStartTime());
080    assertEquals(master.getMasterCoprocessors().length, info.getCoprocessors().length);
081    assertEquals(master.getServerManager().getOnlineServersList().size(),
082      info.getNumRegionServers());
083    assertEquals(master.getMasterWalManager().getOldWALsDirSize(), info.getOldWALsDirSize());
084    int regionServerCount = NUM_RS;
085    assertEquals(regionServerCount, info.getNumRegionServers());
086
087    String zkServers = info.getZookeeperQuorum();
088    assertEquals(zkServers.split(",").length, TEST_UTIL.getZkCluster().getZooKeeperServerNum());
089
090    final int index = 3;
091    LOG.info("Stopping " + TEST_UTIL.getMiniHBaseCluster().getRegionServer(index));
092    TEST_UTIL.getMiniHBaseCluster().stopRegionServer(index, false);
093    TEST_UTIL.getMiniHBaseCluster().waitOnRegionServer(index);
094    // We stopped the regionserver but could take a while for the master to notice it so hang here
095    // until it does... then move forward to see if metrics wrapper notices.
096    while (
097      TEST_UTIL.getHBaseCluster().getMaster().getServerManager().getOnlineServers().size()
098          == regionServerCount
099    ) {
100      Threads.sleep(10);
101    }
102    assertEquals(regionServerCount - 1, info.getNumRegionServers());
103    assertEquals(1, info.getNumDeadRegionServers());
104    // now we do not expose this information as WALProcedureStore is not the only ProcedureStore
105    // implementation any more.
106    assertEquals(0, info.getNumWALFiles());
107    // We decommission the first online region server and verify the metrics.
108    TEST_UTIL.getMiniHBaseCluster().getMaster().decommissionRegionServers(
109      master.getServerManager().getOnlineServersList().subList(0, 1), false);
110    assertEquals(1, info.getNumDrainingRegionServers());
111    assertEquals(master.getServerManager().getOnlineServersList().get(0).toString(),
112      info.getDrainingRegionServers());
113  }
114
115  @Test
116  public void testQuotaSnapshotConversion() {
117    MetricsMasterWrapperImpl info =
118      new MetricsMasterWrapperImpl(TEST_UTIL.getHBaseCluster().getMaster());
119    assertEquals(new SimpleImmutableEntry<Long, Long>(1024L, 2048L), info
120      .convertSnapshot(new SpaceQuotaSnapshot(SpaceQuotaStatus.notInViolation(), 1024L, 2048L)));
121    assertEquals(new SimpleImmutableEntry<Long, Long>(4096L, 2048L), info.convertSnapshot(
122      new SpaceQuotaSnapshot(new SpaceQuotaStatus(SpaceViolationPolicy.NO_INSERTS), 4096L, 2048L)));
123  }
124
125  /**
126   * tests online and offline region number
127   */
128  @Test
129  public void testOfflineRegion() throws Exception {
130    HMaster master = TEST_UTIL.getHBaseCluster().getMaster();
131    MetricsMasterWrapperImpl info = new MetricsMasterWrapperImpl(master);
132    TableName table = TableName.valueOf("testRegionNumber");
133    try {
134      RegionInfo hri;
135      byte[] FAMILY = Bytes.toBytes("FAMILY");
136      TableDescriptor tableDescriptor = TableDescriptorBuilder.newBuilder(table)
137        .setColumnFamily(ColumnFamilyDescriptorBuilder.of(FAMILY)).build();
138      TEST_UTIL.getAdmin().createTable(tableDescriptor, Bytes.toBytes("A"), Bytes.toBytes("Z"), 5);
139
140      // wait till the table is assigned
141      long timeoutTime = EnvironmentEdgeManager.currentTime() + 1000;
142      while (true) {
143        List<RegionInfo> regions =
144          master.getAssignmentManager().getRegionStates().getRegionsOfTable(table);
145        if (regions.size() > 3) {
146          hri = regions.get(2);
147          break;
148        }
149        long now = EnvironmentEdgeManager.currentTime();
150        if (now > timeoutTime) {
151          fail("Could not find an online region");
152        }
153        Thread.sleep(10);
154      }
155
156      PairOfSameType<Integer> regionNumberPair = info.getRegionCounts();
157      assertEquals(5, regionNumberPair.getFirst().intValue());
158      assertEquals(0, regionNumberPair.getSecond().intValue());
159
160      TEST_UTIL.getAdmin().offline(hri.getRegionName());
161
162      timeoutTime = EnvironmentEdgeManager.currentTime() + 800;
163      RegionStates regionStates = master.getAssignmentManager().getRegionStates();
164      while (true) {
165        if (
166          regionStates.getRegionByStateOfTable(table).get(RegionState.State.OFFLINE).contains(hri)
167        ) {
168          break;
169        }
170        long now = EnvironmentEdgeManager.currentTime();
171        if (now > timeoutTime) {
172          fail("Failed to offline the region in time");
173          break;
174        }
175        Thread.sleep(10);
176      }
177      regionNumberPair = info.getRegionCounts();
178      assertEquals(4, regionNumberPair.getFirst().intValue());
179      assertEquals(1, regionNumberPair.getSecond().intValue());
180    } finally {
181      TEST_UTIL.deleteTable(table);
182    }
183  }
184}