HDDS-11696. Limit max number of entries in list keys/status response (a…

…pache#7431)

hadoop-hdds/common/src/main/resources/ozone-default.xml

@@ -4556,4 +4556,12 @@
     </description>
   </property>
 
+  <property>
+    <name>ozone.om.server.list.max.size</name>
+    <value>1000</value>
+    <tag>OZONE, OM</tag>
+    <description>
+      Configuration property to configure the max server side response size for list calls on om.
+    </description>
+  </property>
 </configuration>

hadoop-ozone/client/src/main/java/org/apache/hadoop/ozone/client/OzoneBucket.java

@@ -1808,7 +1808,6 @@ private boolean getChildrenKeys(String keyPrefix, String startKey,
       // 1. Get immediate children of keyPrefix, starting with startKey
       List<OzoneFileStatusLight> statuses = proxy.listStatusLight(volumeName,
           name, keyPrefix, false, startKey, listCacheSize, true);
-      boolean reachedLimitCacheSize = statuses.size() == listCacheSize;
 
       // 2. Special case: ListKey expects keyPrefix element should present in
       // the resultList, only if startKey is blank. If startKey is not blank
@@ -1840,7 +1839,7 @@ private boolean getChildrenKeys(String keyPrefix, String startKey,
           // Return it so that the next iteration will be
           // started using the stacked items.
           return true;
-        } else if (reachedLimitCacheSize && indx == statuses.size() - 1) {
+        } else if (indx == statuses.size() - 1) {
           // The last element is a FILE and reaches the listCacheSize.
           // Now, sets next seek key to this element
           stack.push(new ImmutablePair<>(keyPrefix, keyInfo.getKeyName()));

hadoop-ozone/common/src/main/java/org/apache/hadoop/ozone/OFSPath.java

@@ -301,6 +301,19 @@ public boolean isSnapshotPath() {
     return false;
   }
 
+  /**
+   * If the path is a snapshot path get the snapshot name from the key name.
+   */
+  public String getSnapshotName() {
+    if (keyName.startsWith(OM_SNAPSHOT_INDICATOR)) {
+      if (!bucketName.isEmpty() && !volumeName.isEmpty()) {
+        String[] keyNames = keyName.split(OZONE_URI_DELIMITER);
+        return keyNames.length > 1 ? keyNames[1] : null;
+      }
+    }
+    return null;
+  }
+
   /**
    * If key name is not empty, the given path is a key.
    * e.g. /volume1/bucket2/key3 is a key.

hadoop-ozone/common/src/main/java/org/apache/hadoop/ozone/om/OMConfigKeys.java

@@ -625,4 +625,9 @@ private OMConfigKeys() {
   public static final String OZONE_OM_MAX_BUCKET =
       "ozone.om.max.buckets";
   public static final int OZONE_OM_MAX_BUCKET_DEFAULT = 100000;
+  /**
+   * Configuration property to configure the max server side response size for list calls.
+   */
+  public static final String OZONE_OM_SERVER_LIST_MAX_SIZE = "ozone.om.server.list.max.size";
+  public static final int OZONE_OM_SERVER_LIST_MAX_SIZE_DEFAULT = 1000;
 }

hadoop-ozone/integration-test/src/test/java/org/apache/hadoop/fs/ozone/AbstractOzoneFileSystemTest.java

@@ -117,6 +117,7 @@
 import static org.apache.hadoop.ozone.OzoneConfigKeys.OZONE_FS_ITERATE_BATCH_SIZE;
 import static org.apache.hadoop.ozone.OzoneConsts.OM_KEY_PREFIX;
 import static org.apache.hadoop.ozone.OzoneConsts.OZONE_URI_DELIMITER;
+import static org.apache.hadoop.ozone.om.OMConfigKeys.OZONE_OM_SERVER_LIST_MAX_SIZE;
 import static org.apache.hadoop.ozone.om.exceptions.OMException.ResultCodes.KEY_NOT_FOUND;
 import static org.apache.hadoop.ozone.om.helpers.BucketLayout.FILE_SYSTEM_OPTIMIZED;
 import static org.assertj.core.api.Assertions.assertThat;
@@ -184,7 +185,7 @@ void init() throws Exception {
     conf.setFloat(OMConfigKeys.OZONE_FS_TRASH_INTERVAL_KEY, TRASH_INTERVAL);
     conf.setFloat(FS_TRASH_INTERVAL_KEY, TRASH_INTERVAL);
     conf.setFloat(FS_TRASH_CHECKPOINT_INTERVAL_KEY, TRASH_INTERVAL / 2);
-
+    conf.setInt(OZONE_OM_SERVER_LIST_MAX_SIZE, 2);
     conf.setBoolean(OMConfigKeys.OZONE_OM_RATIS_ENABLE_KEY, omRatisEnabled);
     conf.setBoolean(OZONE_ACL_ENABLED, true);
     conf.setBoolean(OzoneConfigKeys.OZONE_HBASE_ENHANCEMENTS_ALLOWED, true);
@@ -2093,8 +2094,8 @@ void testListStatus2() throws IOException {
     final long initialListStatusCount = omMetrics.getNumListStatus();
     FileStatus[] statusList = fs.listStatus(createPath("/"));
     assertEquals(1, statusList.length);
-    assertChange(initialStats, statistics, Statistic.OBJECTS_LIST.getSymbol(), 1);
-    assertEquals(initialListStatusCount + 1, omMetrics.getNumListStatus());
+    assertChange(initialStats, statistics, Statistic.OBJECTS_LIST.getSymbol(), 2);
+    assertEquals(initialListStatusCount + 2, omMetrics.getNumListStatus());
     assertEquals(fs.getFileStatus(path), statusList[0]);
 
     dirPath = RandomStringUtils.randomAlphanumeric(5);
@@ -2105,8 +2106,8 @@ void testListStatus2() throws IOException {
 
     statusList = fs.listStatus(createPath("/"));
     assertEquals(2, statusList.length);
-    assertChange(initialStats, statistics, Statistic.OBJECTS_LIST.getSymbol(), 2);
-    assertEquals(initialListStatusCount + 2, omMetrics.getNumListStatus());
+    assertChange(initialStats, statistics, Statistic.OBJECTS_LIST.getSymbol(), 4);
+    assertEquals(initialListStatusCount + 4, omMetrics.getNumListStatus());
     for (Path p : paths) {
       assertThat(Arrays.asList(statusList)).contains(fs.getFileStatus(p));
     }

hadoop-ozone/integration-test/src/test/java/org/apache/hadoop/fs/ozone/TestOzoneFsSnapshot.java

@@ -24,6 +24,8 @@
 import java.nio.file.Files;
 import java.nio.file.Path;
 import java.nio.file.Paths;
+import java.util.ArrayList;
+import java.util.List;
 import java.util.concurrent.TimeUnit;
 import java.util.stream.Stream;
 import java.util.concurrent.atomic.AtomicInteger;
@@ -49,10 +51,13 @@
 
 import static java.nio.charset.StandardCharsets.UTF_8;
 import static org.apache.hadoop.fs.FileSystem.FS_DEFAULT_NAME_KEY;
+import static org.apache.hadoop.ozone.OzoneConfigKeys.OZONE_FS_LISTING_PAGE_SIZE;
+import static org.apache.hadoop.ozone.OzoneConfigKeys.OZONE_FS_LISTING_PAGE_SIZE_DEFAULT;
 import static org.apache.hadoop.ozone.OzoneConfigKeys.OZONE_SNAPSHOT_DELETING_SERVICE_INTERVAL;
 import static org.apache.hadoop.ozone.OzoneConsts.OM_KEY_PREFIX;
 import static org.apache.hadoop.ozone.OzoneConsts.OZONE_OFS_URI_SCHEME;
 import static org.apache.hadoop.ozone.OzoneConsts.OM_SNAPSHOT_INDICATOR;
+import static org.apache.hadoop.ozone.om.OMConfigKeys.OZONE_OM_SERVER_LIST_MAX_SIZE;
 import static org.apache.hadoop.ozone.om.OMConfigKeys.OZONE_SNAPSHOT_SST_FILTERING_SERVICE_INTERVAL;
 import static org.apache.hadoop.ozone.om.OmSnapshotManager.getSnapshotPath;
 import static org.assertj.core.api.Assertions.assertThat;
@@ -91,6 +96,8 @@ static void initClass() throws Exception {
     conf.setBoolean(OMConfigKeys.OZONE_FILESYSTEM_SNAPSHOT_ENABLED_KEY, true);
     conf.setTimeDuration(OZONE_SNAPSHOT_DELETING_SERVICE_INTERVAL, 1, TimeUnit.SECONDS);
     conf.setInt(OZONE_SNAPSHOT_SST_FILTERING_SERVICE_INTERVAL, KeyManagerImpl.DISABLE_VALUE);
+    conf.setInt(OZONE_OM_SERVER_LIST_MAX_SIZE, 20);
+    conf.setInt(OZONE_FS_LISTING_PAGE_SIZE, 30);
 
     // Start the cluster
     cluster = MiniOzoneCluster.newHABuilder(conf)
@@ -289,6 +296,13 @@ void testFsLsSnapshot(@TempDir Path tempDir) throws Exception {
       String snapshotPath2 = BUCKET_WITH_SNAPSHOT_INDICATOR_PATH +
           OM_KEY_PREFIX + snapshotName2;
       String snapshotKeyPath2 = snapshotPath2 + OM_KEY_PREFIX + key2;
+      List<String> snapshotNames = new ArrayList<>();
+      for (int i = 0; i < cluster.getConf().getInt(OZONE_FS_LISTING_PAGE_SIZE,
+          OZONE_FS_LISTING_PAGE_SIZE_DEFAULT) * 2; i++) {
+        snapshotNames.add(createSnapshot());
+      }
+      String snapshotName3 = createSnapshot();
+
 
       int res = ToolRunner.run(shell,
           new String[]{"-deleteSnapshot", BUCKET_PATH, snapshotName1});
@@ -313,6 +327,10 @@ void testFsLsSnapshot(@TempDir Path tempDir) throws Exception {
 
       assertThat(listSnapOut).doesNotContain(snapshotName1);
       assertThat(listSnapOut).contains(snapshotName2);
+      assertThat(listSnapOut).contains(snapshotName3);
+      for (String snapshotName : snapshotNames) {
+        assertThat(listSnapOut).contains(snapshotName);
+      }
 
       // Check for snapshot keys with "ozone fs -ls"
       String listSnapKeyOut = execShellCommandAndGetOutput(1,

hadoop-ozone/integration-test/src/test/java/org/apache/hadoop/ozone/freon/TestHadoopDirTreeGenerator.java

@@ -46,6 +46,9 @@
 import java.io.IOException;
 import java.net.URI;
 import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
 
 import static org.assertj.core.api.Assertions.assertThat;
 import static org.junit.jupiter.api.Assertions.assertEquals;
@@ -150,7 +153,7 @@ private void verifyDirTree(String volumeName, String bucketName, int depth,
     FileStatus[] fileStatuses = fileSystem.listStatus(rootDir);
     // verify the num of peer directories, expected span count is 1
     // as it has only one dir at root.
-    verifyActualSpan(1, fileStatuses);
+    verifyActualSpan(1, Arrays.asList(fileStatuses));
     for (FileStatus fileStatus : fileStatuses) {
       int actualDepth =
           traverseToLeaf(fileSystem, fileStatus.getPath(), 1, depth, span,
@@ -164,14 +167,16 @@ private int traverseToLeaf(FileSystem fs, Path dirPath, int depth,
                              int expectedFileCnt, StorageSize perFileSize)
           throws IOException {
     FileStatus[] fileStatuses = fs.listStatus(dirPath);
+    List<FileStatus> fileStatusList = new ArrayList<>();
+    Collections.addAll(fileStatusList, fileStatuses);
     // check the num of peer directories except root and leaf as both
     // has less dirs.
     if (depth < expectedDepth - 1) {
-      verifyActualSpan(expectedSpanCnt, fileStatuses);
+      verifyActualSpan(expectedSpanCnt, fileStatusList);
     }
     int actualNumFiles = 0;
     ArrayList <String> files = new ArrayList<>();
-    for (FileStatus fileStatus : fileStatuses) {
+    for (FileStatus fileStatus : fileStatusList) {
       if (fileStatus.isDirectory()) {
         ++depth;
         return traverseToLeaf(fs, fileStatus.getPath(), depth, expectedDepth,
@@ -192,7 +197,7 @@ private int traverseToLeaf(FileSystem fs, Path dirPath, int depth,
   }
 
   private int verifyActualSpan(int expectedSpanCnt,
-                               FileStatus[] fileStatuses) {
+                               List<FileStatus> fileStatuses) {
     int actualSpan = 0;
     for (FileStatus fileStatus : fileStatuses) {
       if (fileStatus.isDirectory()) {

hadoop-ozone/integration-test/src/test/java/org/apache/hadoop/ozone/om/TestListKeys.java

@@ -48,6 +48,7 @@
 import static com.google.common.collect.Lists.newLinkedList;
 import static org.apache.hadoop.ozone.OzoneConfigKeys.OZONE_CLIENT_LIST_CACHE_SIZE;
 import static org.apache.hadoop.ozone.OzoneConfigKeys.OZONE_FS_ITERATE_BATCH_SIZE;
+import static org.apache.hadoop.ozone.om.OMConfigKeys.OZONE_OM_SERVER_LIST_MAX_SIZE;
 import static org.junit.jupiter.params.provider.Arguments.of;
 import static org.junit.jupiter.api.Assertions.assertEquals;
 
@@ -80,6 +81,7 @@ public static void init() throws Exception {
     // Set the number of keys to be processed during batch operate.
     conf.setInt(OZONE_FS_ITERATE_BATCH_SIZE, 3);
     conf.setInt(OZONE_CLIENT_LIST_CACHE_SIZE, 3);
+    conf.setInt(OZONE_OM_SERVER_LIST_MAX_SIZE, 2);
     cluster = MiniOzoneCluster.newBuilder(conf).build();
     cluster.waitForClusterToBeReady();
     client = cluster.newClient();

hadoop-ozone/integration-test/src/test/java/org/apache/hadoop/ozone/om/TestListKeysWithFSO.java

@@ -47,6 +47,7 @@
 
 import static org.apache.hadoop.ozone.OzoneConfigKeys.OZONE_CLIENT_LIST_CACHE_SIZE;
 import static org.apache.hadoop.ozone.OzoneConfigKeys.OZONE_FS_ITERATE_BATCH_SIZE;
+import static org.apache.hadoop.ozone.om.OMConfigKeys.OZONE_OM_SERVER_LIST_MAX_SIZE;
 import static org.junit.jupiter.api.Assertions.assertEquals;
 
 /**
@@ -81,6 +82,7 @@ public static void init() throws Exception {
     // Set the number of keys to be processed during batch operate.
     conf.setInt(OZONE_FS_ITERATE_BATCH_SIZE, 3);
     conf.setInt(OZONE_CLIENT_LIST_CACHE_SIZE, 3);
+    conf.setInt(OZONE_OM_SERVER_LIST_MAX_SIZE, 2);
     cluster = MiniOzoneCluster.newBuilder(conf).build();
     cluster.waitForClusterToBeReady();
     client = cluster.newClient();

hadoop-ozone/ozone-manager/src/main/java/org/apache/hadoop/ozone/protocolPB/OzoneManagerRequestHandler.java

@@ -146,6 +146,8 @@
 
 import com.google.common.collect.Lists;
 
+import static org.apache.hadoop.ozone.om.OMConfigKeys.OZONE_OM_SERVER_LIST_MAX_SIZE;
+import static org.apache.hadoop.ozone.om.OMConfigKeys.OZONE_OM_SERVER_LIST_MAX_SIZE_DEFAULT;
 import static org.apache.hadoop.ozone.om.upgrade.OMLayoutFeature.HBASE_SUPPORT;
 import static org.apache.hadoop.ozone.om.upgrade.OMLayoutFeature.MULTITENANCY_SCHEMA;
 import static org.apache.hadoop.ozone.om.upgrade.OMLayoutFeature.FILESYSTEM_SNAPSHOT;
@@ -181,9 +183,16 @@ public class OzoneManagerRequestHandler implements RequestHandler {
       LoggerFactory.getLogger(OzoneManagerRequestHandler.class);
   private final OzoneManager impl;
   private FaultInjector injector;
+  private long maxKeyListSize;
+
 
   public OzoneManagerRequestHandler(OzoneManager om) {
     this.impl = om;
+    this.maxKeyListSize = om.getConfiguration().getLong(OZONE_OM_SERVER_LIST_MAX_SIZE,
+      OZONE_OM_SERVER_LIST_MAX_SIZE_DEFAULT);
+    if (this.maxKeyListSize <= 0) {
+      this.maxKeyListSize = OZONE_OM_SERVER_LIST_MAX_SIZE_DEFAULT;
+    }
   }
 
   //TODO simplify it to make it shorter
@@ -745,7 +754,7 @@ private ListKeysResponse listKeys(ListKeysRequest request, int clientVersion)
         request.getBucketName(),
         request.getStartKey(),
         request.getPrefix(),
-        request.getCount());
+        (int)Math.min(this.maxKeyListSize, request.getCount()));
     for (OmKeyInfo key : listKeysResult.getKeys()) {
       resp.addKeyInfo(key.getProtobuf(true, clientVersion));
     }
@@ -763,7 +772,7 @@ private ListKeysLightResponse listKeysLight(ListKeysRequest request)
         request.getBucketName(),
         request.getStartKey(),
         request.getPrefix(),
-        request.getCount());
+        (int)Math.min(this.maxKeyListSize, request.getCount()));
     for (BasicOmKeyInfo key : listKeysLightResult.getKeys()) {
       resp.addBasicKeyInfo(key.getProtobuf());
     }
@@ -1234,7 +1243,7 @@ private ListStatusResponse listStatus(
         request.hasAllowPartialPrefix() && request.getAllowPartialPrefix();
     List<OzoneFileStatus> statuses =
         impl.listStatus(omKeyArgs, request.getRecursive(),
-            request.getStartKey(), request.getNumEntries(),
+            request.getStartKey(), Math.min(this.maxKeyListSize, request.getNumEntries()),
             allowPartialPrefixes);
     ListStatusResponse.Builder
         listStatusResponseBuilder =
@@ -1260,7 +1269,7 @@ private ListStatusLightResponse listStatusLight(
         request.hasAllowPartialPrefix() && request.getAllowPartialPrefix();
     List<OzoneFileStatusLight> statuses =
         impl.listStatusLight(omKeyArgs, request.getRecursive(),
-            request.getStartKey(), request.getNumEntries(),
+            request.getStartKey(), Math.min(this.maxKeyListSize, request.getNumEntries()),
             allowPartialPrefixes);
     ListStatusLightResponse.Builder
         listStatusLightResponseBuilder =
@@ -1488,7 +1497,7 @@ private OzoneManagerProtocolProtos.ListSnapshotResponse getSnapshots(
       throws IOException {
     ListSnapshotResponse implResponse = impl.listSnapshot(
         request.getVolumeName(), request.getBucketName(), request.getPrefix(),
-        request.getPrevSnapshot(), request.getMaxListResult());
+        request.getPrevSnapshot(), (int)Math.min(request.getMaxListResult(), maxKeyListSize));
 
     List<OzoneManagerProtocolProtos.SnapshotInfo> snapshotInfoList = implResponse.getSnapshotInfos()
         .stream().map(SnapshotInfo::getProtobuf).collect(Collectors.toList());

hadoop-ozone/ozone-manager/src/test/java/org/apache/hadoop/ozone/om/TestOMMultiTenantManager.java

@@ -131,7 +131,8 @@ public void testMultiTenancyRequestsWhenDisabled() throws IOException {
 
     final OzoneManager ozoneManager = mock(OzoneManager.class);
     doCallRealMethod().when(ozoneManager).checkS3MultiTenancyEnabled();
-
+    final OzoneConfiguration conf = new OzoneConfiguration();
+    when(ozoneManager.getConfiguration()).thenReturn(conf);
     when(ozoneManager.isS3MultiTenancyEnabled()).thenReturn(false);
 
     final String tenantId = "test-tenant";