[GitHub] [hbase] wchevreuil commented on a change in pull request #798: HBASE-23257: Track clusterID in stand by masters

[GitBox <gi...@apache.org>]

wchevreuil commented on a change in pull request #798: HBASE-23257: Track clusterID in stand by masters
URL: https://github.com/apache/hbase/pull/798#discussion_r343279534
 
 

 ##########
 File path: hbase-server/src/main/java/org/apache/hadoop/hbase/master/CachedClusterId.java
 ##########
 @@ -0,0 +1,146 @@
+/*
+ * 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.hbase.master;
+
+import java.io.IOException;
+import java.util.concurrent.atomic.AtomicBoolean;
+import java.util.concurrent.atomic.AtomicInteger;
+import java.util.concurrent.locks.ReadWriteLock;
+import java.util.concurrent.locks.ReentrantReadWriteLock;
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.FileSystem;
+import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.hbase.ClusterId;
+import org.apache.hadoop.hbase.util.FSUtils;
+import org.apache.yetus.audience.InterfaceAudience;
+import org.apache.hbase.thirdparty.com.google.common.annotations.VisibleForTesting;
+import org.apache.hbase.thirdparty.com.google.common.base.Preconditions;
+
+/**
+ * Caches the cluster ID of the cluster. For standby masters, this is used to serve the client
+ * RPCs that fetch the cluster ID. ClusterID is only created by an active master if one does not
+ * already exist. Standby masters just read the information from the file system. This class is
+ * thread-safe.
+ *
+ * TODO: Make it a singleton without affecting concurrent junit tests.
+ */
+@InterfaceAudience.Private
+public class CachedClusterId {
+
+  private static int MAX_FETCH_TIMEOUT_MS = 10000;
+
+  private Path rootDir;
+  private FileSystem fs;
+
+  // When true, indicates that a FileSystem fetch of ClusterID is in progress. This is used to
+  // avoid multiple fetches from FS and let only one thread fetch the information.
+  AtomicBoolean fetchInProgress = new AtomicBoolean(false);
+
+  // Protects accesses to clusterID.
+  private ReadWriteLock clusterIdLock = new ReentrantReadWriteLock();
+  private ClusterId clusterId;
+
+  // cache stats for testing.
+  private AtomicInteger cacheMisses = new AtomicInteger(0);
+
+  public CachedClusterId(Configuration conf) throws IOException {
+    rootDir = FSUtils.getRootDir(conf);
+    fs = rootDir.getFileSystem(conf);
+  }
+
+  /**
+   * Succeeds only once, when the cached id = null. Overwrites are not allowed and throws a RTE.
+   */
+  private void setClusterId(ClusterId id) {
+    clusterIdLock.writeLock().lock();
+    try {
+      // Make sure we write the cluster ID only once.
+      Preconditions.checkState(clusterId == null, "Error setting clusterID since it is non-null");
+      clusterId = id;
+    } finally {
+      clusterIdLock.writeLock().unlock();
+    }
+  }
+
+  /**
+   * Returns a cached copy of the cluster ID. null if the cache is not populated.
+   */
+  private String getClusterId() {
+    clusterIdLock.readLock().lock();
+    try {
+      if (clusterId == null) {
+        return null;
+      }
+      return clusterId.toString();
+    } finally {
+      clusterIdLock.readLock().unlock();
+    }
+  }
+
+  /**
+   * Fetches the ClusterId from FS if it is not cached locally. Atomically updates the cached
+   * copy and is thread-safe. Optimized to do a single fetch when there are multiple threads are
+   * trying get from a clean cache.
+   *
+   * @return ClusterId by reading from FileSystem or null in any error case or cluster ID does
+   *     not exist on the file system.
+   * @throws IOException when there is an FS issue fetching the cluster ID.
+   */
+  public String getFromCacheOrFetch() {
+    String id = getClusterId();
+    if (id != null) {
+      return id;
+    }
+    // It is not cached here, attempt a fetch if no fetch is in progress.
+    if (fetchInProgress.compareAndSet(false, true)) {
+      // A fetch is not in progress, so try fetching the cluster ID synchronously and then notify
+      // the waiting threads.
+      try {
+        cacheMisses.incrementAndGet();
+        setClusterId(FSUtils.getClusterId(fs, rootDir));
+      } catch (IOException e) {
+        return null;
 
 Review comment:
   nit: worth a debug or trace log?

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services