smengcl commented on a change in pull request #2389:
URL: https://github.com/apache/ozone/pull/2389#discussion_r665334841
##########
File path:
hadoop-ozone/recon/src/main/java/org/apache/hadoop/ozone/recon/ReconUtils.java
##########
@@ -299,4 +299,28 @@ public static void upsertGlobalStatsTable(Configuration
sqlConfiguration,
globalStatsDao.update(newRecord);
}
}
+
+ public static long getFileSizeUpperBound(long fileSize) {
+ if (fileSize >= ReconConstants.MAX_FILE_SIZE_UPPER_BOUND) {
+ return Long.MAX_VALUE;
+ }
+ // The smallest file size being tracked for count
+ // is 1 KB i.e. 1024 = 2 ^ 10.
+ int binIndex = getBinIndex(fileSize);
+ return (long) Math.pow(2, (10 + binIndex));
+ }
+
+ public static int getBinIndex(long fileSize) {
+ int index = nextClosestPowerIndexOfTwo(fileSize);
+ return index < 10 ? 0 : index - 10;
+ }
+
+ private static int nextClosestPowerIndexOfTwo(long dataSize) {
+ int index = 0;
+ while(dataSize != 0) {
Review comment:
nit
```suggestion
while (dataSize != 0) {
```
##########
File path:
hadoop-ozone/recon/src/main/java/org/apache/hadoop/ozone/recon/tasks/NSSummaryTask.java
##########
@@ -0,0 +1,205 @@
+/*
+ * 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
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * 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.recon.tasks;
+import org.apache.commons.lang3.tuple.ImmutablePair;
+import org.apache.commons.lang3.tuple.Pair;
+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.OmKeyInfo;
+import org.apache.hadoop.ozone.recon.ReconConstants;
+import org.apache.hadoop.ozone.recon.ReconUtils;
+import org.apache.hadoop.ozone.recon.api.types.NSSummary;
+import org.apache.hadoop.ozone.recon.spi.ReconNamespaceSummaryManager;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import javax.inject.Inject;
+import java.io.IOException;
+import java.util.Collection;
+import java.util.Collections;
+import java.util.Iterator;
+
+import static org.apache.hadoop.ozone.om.OmMetadataManagerImpl.KEY_TABLE;
+
+/**
+ * Task to query data from OMDB and write into Recon RocksDB.
+ */
+public class NSSummaryTask implements ReconOmTask {
+ private static final Logger LOG =
+ LoggerFactory.getLogger(NSSummaryTask.class);
+ private ReconNamespaceSummaryManager reconNamespaceSummaryManager;
+
+ @Inject
+ public NSSummaryTask(ReconNamespaceSummaryManager
+ reconNamespaceSummaryManager) {
+ this.reconNamespaceSummaryManager = reconNamespaceSummaryManager;
+ }
+
+ @Override
+ public String getTaskName() {
+ return "NSSummaryTask";
+ }
+
+ public Collection<String> getTaskTables() {
+ return Collections.singletonList(KEY_TABLE);
+ }
+
+ @Override
+ public Pair<String, Boolean> process(OMUpdateEventBatch events) {
+ Iterator<OMDBUpdateEvent> eventIterator = events.getIterator();
+ final Collection<String> taskTables = getTaskTables();
+
+ while (eventIterator.hasNext()) {
+ OMDBUpdateEvent<String, OmKeyInfo> omdbUpdateEvent =
eventIterator.next();
+ // we only process updates on OM's KeyTable.
+ if (!taskTables.contains(omdbUpdateEvent.getTable())) {
+ continue;
+ }
+ String updatedKey = omdbUpdateEvent.getKey();
+ OmKeyInfo updatedKeyValue = omdbUpdateEvent.getValue();
+
+ try {
+ switch (omdbUpdateEvent.getAction()) {
+ case PUT:
+ writeOmKeyInfoOnNamespaceDB(updatedKeyValue);
+ break;
+
+ case DELETE:
+ deleteOmKeyInfoOnNamespaceDB(updatedKeyValue);
+ break;
+
+ case UPDATE:
+ if (omdbUpdateEvent.getOldValue() != null) {
+ // delete first, then put
+ deleteOmKeyInfoOnNamespaceDB(omdbUpdateEvent.getOldValue());
+ } else {
+ LOG.warn("Update event does not have the old Key Info for {}.",
+ updatedKey);
+ }
+ writeOmKeyInfoOnNamespaceDB(updatedKeyValue);
+ break;
+
+ default:
+ LOG.debug("Skipping DB update event : {}",
+ omdbUpdateEvent.getAction());
+ }
+
+ } catch (IOException ioEx) {
+ LOG.error("Unable to process Namespace Summary data in Recon DB. ",
+ ioEx);
+ return new ImmutablePair<>(getTaskName(), false);
+ }
+ }
+ LOG.info("Completed a process run of NSSummaryTask");
+ return new ImmutablePair<>(getTaskName(), true);
+ }
+
+ @Override
+ public Pair<String, Boolean> reprocess(OMMetadataManager omMetadataManager) {
+ Table keyTable = omMetadataManager.getKeyTable();
+ TableIterator<String, ? extends
+ Table.KeyValue<String, OmKeyInfo>> tableIter = keyTable.iterator();
+
+ try {
+ // reinit Recon RocksDB's namespace CF.
+ reconNamespaceSummaryManager.initNSSummaryTable();
+
+ while (tableIter.hasNext()) {
+ Table.KeyValue<String, OmKeyInfo> kv = tableIter.next();
+ OmKeyInfo keyInfo = kv.getValue();
+ writeOmKeyInfoOnNamespaceDB(keyInfo);
+ }
+ } catch (IOException ioEx) {
+ LOG.error("Unable to reprocess Namespace Summary data in Recon DB. ",
+ ioEx);
+ return new ImmutablePair<>(getTaskName(), false);
+ }
+
+ LOG.info("Completed a reprocess run of NSSummaryTask");
+ return new ImmutablePair<>(getTaskName(), true);
+ }
+
+ private void writeOmKeyInfoOnNamespaceDB(OmKeyInfo keyInfo)
+ throws IOException {
+ long parentObjectId = keyInfo.getParentObjectID();
+ NSSummary nsSummary = reconNamespaceSummaryManager
+ .getNSSummary(parentObjectId);
+ if (nsSummary == null) {
+ nsSummary = getEmptyNSSummary();
+ }
+ int numOfFile = nsSummary.getNumOfFiles();
+ long sizeOfFile = nsSummary.getSizeOfFiles();
+ int[] fileBucket = nsSummary.getFileSizeBucket();
+ nsSummary.setNumOfFiles(numOfFile + 1);
+ long dataSize = keyInfo.getDataSize();
+ nsSummary.setSizeOfFiles(sizeOfFile + dataSize);
+ int binIndex = ReconUtils.getBinIndex(dataSize);
+
+ // make sure the file is within our scope of tracking.
+ if (binIndex >= 0 && binIndex < ReconConstants.NUM_OF_BINS) {
+ ++fileBucket[binIndex];
+ nsSummary.setFileSizeBucket(fileBucket);
+ } else {
+ LOG.warn("File size beyond our tracking scope.");
Review comment:
Better throw here? Warnings are often overlooked.
##########
File path:
hadoop-ozone/recon/src/main/java/org/apache/hadoop/ozone/recon/tasks/NSSummaryTask.java
##########
@@ -0,0 +1,205 @@
+/*
+ * 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
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * 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.recon.tasks;
+import org.apache.commons.lang3.tuple.ImmutablePair;
+import org.apache.commons.lang3.tuple.Pair;
+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.OmKeyInfo;
+import org.apache.hadoop.ozone.recon.ReconConstants;
+import org.apache.hadoop.ozone.recon.ReconUtils;
+import org.apache.hadoop.ozone.recon.api.types.NSSummary;
+import org.apache.hadoop.ozone.recon.spi.ReconNamespaceSummaryManager;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import javax.inject.Inject;
+import java.io.IOException;
+import java.util.Collection;
+import java.util.Collections;
+import java.util.Iterator;
+
+import static org.apache.hadoop.ozone.om.OmMetadataManagerImpl.KEY_TABLE;
+
+/**
+ * Task to query data from OMDB and write into Recon RocksDB.
+ */
+public class NSSummaryTask implements ReconOmTask {
+ private static final Logger LOG =
+ LoggerFactory.getLogger(NSSummaryTask.class);
+ private ReconNamespaceSummaryManager reconNamespaceSummaryManager;
+
+ @Inject
+ public NSSummaryTask(ReconNamespaceSummaryManager
+ reconNamespaceSummaryManager) {
+ this.reconNamespaceSummaryManager = reconNamespaceSummaryManager;
+ }
+
+ @Override
+ public String getTaskName() {
+ return "NSSummaryTask";
+ }
+
+ public Collection<String> getTaskTables() {
+ return Collections.singletonList(KEY_TABLE);
+ }
+
+ @Override
+ public Pair<String, Boolean> process(OMUpdateEventBatch events) {
+ Iterator<OMDBUpdateEvent> eventIterator = events.getIterator();
+ final Collection<String> taskTables = getTaskTables();
+
+ while (eventIterator.hasNext()) {
+ OMDBUpdateEvent<String, OmKeyInfo> omdbUpdateEvent =
eventIterator.next();
+ // we only process updates on OM's KeyTable.
+ if (!taskTables.contains(omdbUpdateEvent.getTable())) {
Review comment:
Have we made sure that we are only tracking FSO-only buckets here?
##########
File path:
hadoop-ozone/recon/src/main/java/org/apache/hadoop/ozone/recon/codec/NSSummaryCodec.java
##########
@@ -36,15 +37,18 @@
private final Codec<Integer> integerCodec = new IntegerCodec();
private final Codec<Short> shortCodec = new ShortCodec();
- // 2 int fields + 41-length int array
- private static final int NUM_OF_INTS = 2 + ReconConstants.NUM_OF_BINS;
+ private final Codec<Long> longCodec = new LongCodec();
+ // 1 int fields + 41-length int array
+ private static final int NUM_OF_INTS = 1 + ReconConstants.NUM_OF_BINS;
Review comment:
One field changed from int to long hence -1.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
