You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ozone.apache.org by "szetszwo (via GitHub)" <gi...@apache.org> on 2025/06/12 20:14:09 UTC
Re: [PR] HDDS-8742. Quota repair counts of namespace and space usages [ozone]
szetszwo commented on code in PR #4817:
URL: https://github.com/apache/ozone/pull/4817#discussion_r2143581488
##########
hadoop-ozone/ozone-manager/src/main/java/org/apache/hadoop/ozone/om/service/QuotaRepairTask.java:
##########
@@ -0,0 +1,326 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+
+package org.apache.hadoop.ozone.om.service;
+
+import com.google.common.util.concurrent.UncheckedExecutionException;
+import java.io.IOException;
+import java.io.UncheckedIOException;
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.concurrent.ArrayBlockingQueue;
+import java.util.concurrent.BlockingQueue;
+import java.util.concurrent.ConcurrentHashMap;
+import java.util.concurrent.ExecutionException;
+import java.util.concurrent.ExecutorService;
+import java.util.concurrent.Executors;
+import java.util.concurrent.Future;
+import java.util.concurrent.TimeUnit;
+import java.util.concurrent.atomic.AtomicBoolean;
+import java.util.concurrent.atomic.AtomicLong;
+import org.apache.hadoop.hdds.utils.db.BatchOperation;
+import org.apache.hadoop.hdds.utils.db.Table;
+import org.apache.hadoop.hdds.utils.db.TableIterator;
+import org.apache.hadoop.ozone.om.OMMetadataManager;
+import org.apache.hadoop.ozone.om.helpers.BucketLayout;
+import org.apache.hadoop.ozone.om.helpers.OmBucketInfo;
+import org.apache.hadoop.ozone.om.helpers.OmKeyInfo;
+import org.apache.hadoop.ozone.om.helpers.OmVolumeArgs;
+import org.apache.hadoop.ozone.om.lock.IOzoneManagerLock;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import static org.apache.hadoop.ozone.OzoneConsts.OM_KEY_PREFIX;
+import static org.apache.hadoop.ozone.om.lock.OzoneManagerLock.Resource.BUCKET_LOCK;
+
+/**
+ * Quota repair task.
+ */
+public class QuotaRepairTask {
+ private static final Logger LOG = LoggerFactory.getLogger(
+ QuotaRepairTask.class);
+ private static final int BATCH_SIZE = 5000;
+ private static final int TASK_THREAD_CNT = 3;
+ private final OMMetadataManager metadataManager;
+ private final Map<String, OmBucketInfo> nameBucketInfoMap = new HashMap<>();
+ private final Map<String, OmBucketInfo> idBucketInfoMap = new HashMap<>();
+ private ExecutorService executor;
+ private final Map<String, CountPair> keyCountMap = new ConcurrentHashMap<>();
+ private final Map<String, CountPair> fileCountMap
+ = new ConcurrentHashMap<>();
+ private final Map<String, CountPair> directoryCountMap
+ = new ConcurrentHashMap<>();
+
+ public QuotaRepairTask(OMMetadataManager metadataManager) {
+ this.metadataManager = metadataManager;
+ }
+
+ public void repair() throws Exception {
+ LOG.info("Starting quota repair task");
+ prepareAllVolumeBucketInfo();
+
+ IOzoneManagerLock lock = metadataManager.getLock();
+ // thread pool with 3 Table type * (1 task each + 3 thread each)
+ executor = Executors.newFixedThreadPool(12);
+ try {
+ nameBucketInfoMap.values().stream().forEach(e -> lock.acquireReadLock(
+ BUCKET_LOCK, e.getVolumeName(), e.getBucketName()));
+ repairCount();
+ } finally {
+ nameBucketInfoMap.values().stream().forEach(e -> lock.releaseReadLock(
+ BUCKET_LOCK, e.getVolumeName(), e.getBucketName()));
+ executor.shutdown();
+ LOG.info("Completed quota repair task");
+ }
+ }
+
+ private void prepareAllVolumeBucketInfo() throws IOException {
+ try (TableIterator<String, ? extends Table.KeyValue<String, OmVolumeArgs>>
+ iterator = metadataManager.getVolumeTable().iterator()) {
+
+ OmVolumeArgs omVolumeArgs;
+ while (iterator.hasNext()) {
+ Table.KeyValue<String, OmVolumeArgs> entry =
+ iterator.next();
+ omVolumeArgs = entry.getValue();
+ getAllBuckets(omVolumeArgs.getVolume(), omVolumeArgs.getObjectID());
+ }
+ }
+ }
+
+ private void getAllBuckets(String volumeName, long volumeId)
+ throws IOException {
+ List<OmBucketInfo> bucketList = metadataManager.listBuckets(
+ volumeName, null, null, Integer.MAX_VALUE, false);
+ for (OmBucketInfo bucketInfo : bucketList) {
+ bucketInfo.incrUsedNamespace(-bucketInfo.getUsedNamespace());
+ bucketInfo.incrUsedBytes(-bucketInfo.getUsedBytes());
+ nameBucketInfoMap.put(buildNamePath(volumeName,
+ bucketInfo.getBucketName()), bucketInfo);
+ idBucketInfoMap.put(buildIdPath(volumeId, bucketInfo.getObjectID()),
+ bucketInfo);
+ }
+ }
+
+ private String buildNamePath(String volumeName, String bucketName) {
+ final StringBuilder builder = new StringBuilder();
+ builder.append(OM_KEY_PREFIX)
+ .append(volumeName)
+ .append(OM_KEY_PREFIX)
+ .append(bucketName)
+ .append(OM_KEY_PREFIX);
+ return builder.toString();
+ }
+
+ private String buildIdPath(long volumeId, long bucketId) {
+ final StringBuilder builder = new StringBuilder();
+ builder.append(OM_KEY_PREFIX)
+ .append(volumeId)
+ .append(OM_KEY_PREFIX)
+ .append(bucketId)
+ .append(OM_KEY_PREFIX);
+ return builder.toString();
+ }
+
+ private void repairCount() throws Exception {
+ LOG.info("Starting quota repair for all keys, files and directories");
+ try {
+ nameBucketInfoMap.keySet().stream().forEach(e -> keyCountMap.put(e,
+ new CountPair()));
+ idBucketInfoMap.keySet().stream().forEach(e -> fileCountMap.put(e,
+ new CountPair()));
+ idBucketInfoMap.keySet().stream().forEach(e -> directoryCountMap.put(e,
+ new CountPair()));
+
+ List<Future<?>> tasks = new ArrayList<>();
+ tasks.add(executor.submit(() -> recalculateUsages(
+ metadataManager.getKeyTable(BucketLayout.OBJECT_STORE),
+ keyCountMap, "Key usages", true)));
+ tasks.add(executor.submit(() -> recalculateUsages(
+ metadataManager.getKeyTable(BucketLayout.FILE_SYSTEM_OPTIMIZED),
+ fileCountMap, "File usages", true)));
+ tasks.add(executor.submit(() -> recalculateUsages(
+ metadataManager.getDirectoryTable(),
+ directoryCountMap, "Directory usages", false)));
+ for (Future<?> f : tasks) {
+ f.get();
+ }
+ } catch (UncheckedIOException ex) {
+ LOG.error("quota repair failure", ex.getCause());
+ throw ex.getCause();
+ } catch (UncheckedExecutionException ex) {
+ LOG.error("quota repair failure", ex.getCause());
+ throw new Exception(ex.getCause());
+ }
+
+ // persist bucket info
+ updateCountToBucketInfo(nameBucketInfoMap, keyCountMap);
+ updateCountToBucketInfo(idBucketInfoMap, fileCountMap);
+ updateCountToBucketInfo(idBucketInfoMap, directoryCountMap);
+
+ try (BatchOperation batchOperation = metadataManager.getStore()
+ .initBatchOperation()) {
+ for (Map.Entry<String, OmBucketInfo> entry
+ : nameBucketInfoMap.entrySet()) {
+ String bucketKey = metadataManager.getBucketKey(
+ entry.getValue().getVolumeName(),
+ entry.getValue().getBucketName());
+ metadataManager.getBucketTable().putWithBatch(batchOperation,
+ bucketKey, entry.getValue());
+ }
+ metadataManager.getStore().commitBatchOperation(batchOperation);
+ }
+ LOG.info("Completed quota repair for all keys, files and directories");
+ }
+
+ private <VALUE> void recalculateUsages(
+ Table<String, VALUE> table, Map<String, CountPair> prefixUsageMap,
+ String strType, boolean haveValue) throws UncheckedIOException,
+ UncheckedExecutionException {
+ LOG.info("Starting recalculate {}", strType);
+
+ List<Table.KeyValue<String, VALUE>> kvList = new ArrayList<>(BATCH_SIZE);
+ BlockingQueue<List<Table.KeyValue<String, VALUE>>> q
+ = new ArrayBlockingQueue<>(TASK_THREAD_CNT);
+ List<Future<?>> tasks = new ArrayList<>();
+ AtomicBoolean isRunning = new AtomicBoolean(true);
+ for (int i = 0; i < TASK_THREAD_CNT; ++i) {
+ tasks.add(executor.submit(() -> captureCount(
+ prefixUsageMap, q, isRunning, haveValue)));
+ }
+ int count = 0;
+ long startTime = System.currentTimeMillis();
+ try (TableIterator<String, ? extends Table.KeyValue<String, VALUE>>
+ keyIter = table.iterator()) {
+ while (keyIter.hasNext()) {
+ count++;
+ kvList.add(keyIter.next());
+ if (kvList.size() == BATCH_SIZE) {
+ q.put(kvList);
+ kvList = new ArrayList<>(BATCH_SIZE);
+ }
+ }
+ q.put(kvList);
+ isRunning.set(false);
+ for (Future<?> f : tasks) {
+ f.get();
+ }
+ LOG.info("Recalculate {} completed, count {} time {}ms", strType,
+ count, (System.currentTimeMillis() - startTime));
+ } catch (IOException ex) {
+ throw new UncheckedIOException(ex);
+ } catch (InterruptedException ex) {
+ Thread.currentThread().interrupt();
+ } catch (ExecutionException ex) {
+ throw new UncheckedExecutionException(ex);
+ }
+ }
+
+ private <VALUE> void captureCount(
+ Map<String, CountPair> prefixUsageMap,
+ BlockingQueue<List<Table.KeyValue<String, VALUE>>> q,
+ AtomicBoolean isRunning, boolean haveValue) throws UncheckedIOException {
+ try {
+ while (isRunning.get() || !q.isEmpty()) {
+ List<Table.KeyValue<String, VALUE>> kvList
+ = q.poll(100, TimeUnit.MILLISECONDS);
+ if (null != kvList) {
+ for (Table.KeyValue<String, VALUE> kv : kvList) {
+ extractCount(kv, prefixUsageMap, haveValue);
+ }
+ }
+ }
+ } catch (InterruptedException ex) {
+ Thread.currentThread().interrupt();
+ }
+ }
+
+ private <VALUE> void extractCount(
+ Table.KeyValue<String, VALUE> kv,
+ Map<String, CountPair> prefixUsageMap,
+ boolean haveValue) {
+ try {
+ String prefix = getVolumeBucketPrefix(kv.getKey());
+ CountPair usage = prefixUsageMap.get(prefix);
+ if (null == usage) {
+ return;
+ }
+ usage.incrNamespace(1L);
+ // avoid decode of value
+ if (haveValue) {
Review Comment:
@sumitagrawl, This may not work since `OmKeyInfo`/`OmDirectoryInfo` supported CodecBuffer -- they were always decoded. The correct way is to support key only iterator. Cc @swamirishi , @adoroszlai
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org