EdColeman commented on a change in pull request #2569:
URL: https://github.com/apache/accumulo/pull/2569#discussion_r829579676



##########
File path: 
server/base/src/main/java/org/apache/accumulo/server/conf/store/impl/PropStoreWatcher.java
##########
@@ -0,0 +1,253 @@
+/*
+ * 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.accumulo.server.conf.store.impl;
+
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.Map;
+import java.util.Set;
+import java.util.concurrent.ExecutorService;
+import java.util.concurrent.locks.ReentrantReadWriteLock;
+import java.util.stream.Collectors;
+
+import org.apache.accumulo.core.util.threads.ThreadPools;
+import org.apache.accumulo.server.conf.store.PropCacheId;
+import org.apache.accumulo.server.conf.store.PropChangeListener;
+import org.apache.zookeeper.WatchedEvent;
+import org.apache.zookeeper.Watcher;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+/**
+ * This class serves as a translator between ZooKeeper events and converts 
them to PropStore events.
+ * Using this as an intermediary, the external listeners do not need to set / 
manage external
+ * ZooKeeper watchers, they can register for PropStore events if they need to 
take active action on
+ * change detection.
+ * <p>
+ * Users of the PropStore.get() will get properties that match what is stored 
in ZooKeeper for each
+ * call and do not need to manage any caching. However, the ability to receive 
active notification
+ * without needed to register / manage ZooKeeper watchers external to the 
PropStore is provided in
+ * case other code is relying on active notifications.
+ * <p>
+ * The notification occurs on a separate thread from the ZooKeeper 
notification handling, but
+ * listeners should not perform lengthy operations on the notification thread 
so that other listener
+ * notifications are not delayed.
+ */
+public class PropStoreWatcher implements Watcher {
+
+  private static final Logger log = 
LoggerFactory.getLogger(PropStoreWatcher.class);
+
+  private final ExecutorService executorService =
+      ThreadPools.getServerThreadPools().createFixedThreadPool(1, 
"zoo_change_update", false);
+
+  private final ReentrantReadWriteLock listenerLock = new 
ReentrantReadWriteLock();
+  private final ReentrantReadWriteLock.ReadLock listenerReadLock = 
listenerLock.readLock();
+  private final ReentrantReadWriteLock.WriteLock listenerWriteLock = 
listenerLock.writeLock();
+
+  // access should be guarded by acquiring the listener read or write lock
+  private final Map<PropCacheId,Set<PropChangeListener>> listeners = new 
HashMap<>();
+
+  private final ReadyMonitor zkReadyMonitor;
+
+  public PropStoreWatcher(final ReadyMonitor zkReadyMonitor) {
+    this.zkReadyMonitor = zkReadyMonitor;
+  }
+
+  public void registerListener(final PropCacheId propCacheId, final 
PropChangeListener listener) {
+    listenerWriteLock.lock();
+    try {
+      Set<PropChangeListener> set = listeners.computeIfAbsent(propCacheId, s 
-> new HashSet<>());
+      set.add(listener);
+    } finally {
+      listenerWriteLock.unlock();
+    }
+  }
+
+  /**
+   * Process a ZooKeeper event. This method does not reset the watcher. 
Subscribers are notified of
+   * the change - if they call get to update and respond to the change the 
watcher will be (re)set
+   * then. This helps clean up watchers by not automatically re-adding the 
watcher on the event but
+   * only if being used.
+   *
+   * @param event
+   *          ZooKeeper event.
+   */
+  @SuppressWarnings("FutureReturnValueIgnored") // currently, tasks are fire 
and forget
+  @Override
+  public void process(final WatchedEvent event) {
+
+    String path;
+    switch (event.getType()) {
+      case NodeDataChanged:
+        path = event.getPath();
+        log.trace("handle change event for path: {}", path);
+        PropCacheId.fromPath(path).ifPresent(this::signalZkChangeEvent);
+        break;
+      case NodeDeleted:
+        path = event.getPath();
+        log.trace("handle delete event for path: {}", path);
+        PropCacheId.fromPath(path).ifPresent(cacheId -> {
+          // notify listeners
+          Set<PropChangeListener> snapshot = getListenerSnapshot(cacheId);
+          if (snapshot != null) {
+            executorService
+                .submit(new 
PropStoreEventTask.PropStoreDeleteEventTask(cacheId, snapshot));
+          }
+
+          listenerCleanup(cacheId);
+
+        });
+
+        break;
+      case None:
+        Event.KeeperState state = event.getState();
+        switch (state) {
+          // pause - could reconnect
+          case ConnectedReadOnly:
+          case Disconnected:
+            log.debug("ZooKeeper disconnected event received");
+            zkReadyMonitor.clearReady();
+            executorService.submit(new 
PropStoreEventTask.PropStoreConnectionEventTask(null,
+                getAllListenersSnapshot()));
+            break;
+
+          // okay
+          case SyncConnected:
+            log.debug("ZooKeeper connected event received");
+            zkReadyMonitor.setReady();
+            break;
+
+          // terminal - never coming back.
+          case Expired:
+          case Closed:
+            log.info("ZooKeeper connection closed event received");
+            zkReadyMonitor.clearReady();
+            zkReadyMonitor.setClosed(); // terminal condition
+            executorService.submit(new 
PropStoreEventTask.PropStoreConnectionEventTask(null,
+                getAllListenersSnapshot()));
+            break;
+
+          default:
+            log.trace("ignoring zooKeeper state: {}", state);
+        }
+        break;
+      default:
+        break;
+    }
+
+  }
+
+  /**
+   * Submit task to notify registered listeners that the propCacheId node 
received an event
+   * notification from ZooKeeper and should be updated. The process can be 
initiated either by a
+   * ZooKeeper notification or a change detected in the cache based on a 
ZooKeeper event.
+   *
+   * @param propCacheId
+   *          the cache id
+   */
+  @SuppressWarnings("FutureReturnValueIgnored") // currently, tasks are fire 
and forget
+  public void signalZkChangeEvent(final PropCacheId propCacheId) {
+    log.trace("signal ZooKeeper change event: {}", propCacheId);
+    Set<PropChangeListener> snapshot = getListenerSnapshot(propCacheId);
+    log.trace("Sending change event to: {}", snapshot);
+    if (snapshot != null) {
+      executorService
+          .submit(new 
PropStoreEventTask.PropStoreZkChangeEventTask(propCacheId, snapshot));
+    }
+  }
+
+  /**
+   * Submit task to notify registered listeners that the propCacheId node 
change was detected should
+   * be updated.
+   *
+   * @param propCacheId
+   *          the cache id
+   */
+  @SuppressWarnings("FutureReturnValueIgnored") // currently, tasks are fire 
and forget
+  public void signalCacheChangeEvent(final PropCacheId propCacheId) {
+    log.trace("cache change event: {}", propCacheId);
+    Set<PropChangeListener> snapshot = getListenerSnapshot(propCacheId);
+    if (snapshot != null) {
+      executorService
+          .submit(new 
PropStoreEventTask.PropStoreCacheChangeEventTask(propCacheId, snapshot));
+    }
+  }
+
+  /**
+   * Clean-up the active listeners set when an entry is removed from the 
cache, remove it from the
+   * active listeners.
+   *
+   * @param propCacheId
+   *          the cache id
+   */
+  public void listenerCleanup(final PropCacheId propCacheId) {
+    listenerWriteLock.lock();
+    try {
+      listeners.remove(propCacheId);
+    } finally {
+      listenerWriteLock.unlock();
+    }
+  }
+
+  /**
+   * Get an immutable snapshot of the listeners for a prop cache id. The set 
is intended for
+   * notification of changes for a specific prop cache id.
+   *
+   * @param PropCacheId
+   *          the prop cache id
+   * @return an immutable copy of listeners.
+   */
+  private Set<PropChangeListener> getListenerSnapshot(final PropCacheId 
PropCacheId) {
+
+    Set<PropChangeListener> snapshot = null;
+    listenerReadLock.lock();
+    try {
+      Set<PropChangeListener> set = listeners.get(PropCacheId);
+      if (set != null) {
+        snapshot = Set.copyOf(set);
+      }
+
+    } finally {
+      listenerReadLock.unlock();
+    }
+    return snapshot;
+  }
+
+  /**
+   * Get an immutable snapshot of the all listeners registered for event. The 
set is intended for
+   * connection event notifications that are not specific to an individual 
prop cache id.
+   *
+   * @return an immutable copy of all registered listeners.
+   */
+  private Set<PropChangeListener> getAllListenersSnapshot() {
+
+    Set<PropChangeListener> snapshot;
+    listenerReadLock.lock();
+    try {
+
+      snapshot = listeners.keySet().stream().flatMap(key -> 
listeners.get(key).stream())
+          .collect(Collectors.toSet());
+
+    } finally {
+      listenerReadLock.unlock();
+    }
+    return Collections.unmodifiableSet(snapshot);

Review comment:
       Fixed in eafaa7da6e - also addressed a checkstyle issue on connection 
events that did not use prop cache key by adding second constructor.




-- 
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]


Reply via email to