leventov commented on a change in pull request #9172: Fix huge number of 
watches in zk
URL: https://github.com/apache/druid/pull/9172#discussion_r365811794
 
 

 ##########
 File path: 
server/src/main/java/org/apache/druid/curator/announcement/NodeAnnouncer.java
 ##########
 @@ -0,0 +1,350 @@
+/*
+ * 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.druid.curator.announcement;
+
+import com.google.common.annotations.VisibleForTesting;
+import org.apache.curator.framework.CuratorFramework;
+import org.apache.curator.framework.api.transaction.CuratorOp;
+import org.apache.curator.framework.recipes.cache.ChildData;
+import org.apache.curator.framework.recipes.cache.NodeCache;
+import org.apache.druid.java.util.common.IAE;
+import org.apache.druid.java.util.common.ISE;
+import org.apache.druid.java.util.common.guava.CloseQuietly;
+import org.apache.druid.java.util.common.io.Closer;
+import org.apache.druid.java.util.common.lifecycle.LifecycleStart;
+import org.apache.druid.java.util.common.lifecycle.LifecycleStop;
+import org.apache.druid.java.util.common.logger.Logger;
+import org.apache.druid.server.ZKPathsUtils;
+import org.apache.zookeeper.CreateMode;
+import org.apache.zookeeper.KeeperException;
+
+import javax.annotation.concurrent.GuardedBy;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.List;
+import java.util.Set;
+import java.util.concurrent.ConcurrentHashMap;
+import java.util.concurrent.ConcurrentMap;
+import java.util.concurrent.CopyOnWriteArrayList;
+
+/**
+ * NodeAnnouncer announces single node on Zookeeper and only watches this node,
+ * while {@link Announcer} watches all child paths, not only this node
+ */
+public class NodeAnnouncer
+{
+  private static final Logger log = new Logger(NodeAnnouncer.class);
+
+  private final CuratorFramework curator;
+
+  /**
+   * In case a path is added to this collection in {@link #announce} before zk 
is connected,
+   * should remember the path and do announce in {@link #start} later.
+   */
+  private final List<Announceable> toAnnounce = new ArrayList<>();
+  /**
+   * In case a path is added to this collection in {@link #update} before zk 
is connected,
+   * should remember the path and do update in {@link #start} later.
+   */
+  @GuardedBy("toAnnounce")
+  private final List<Announceable> toUpdate = new ArrayList<>();
+  private final ConcurrentMap<String, NodeCache> listeners = new 
ConcurrentHashMap<>();
+  private final ConcurrentMap<String, byte[]> announcedPaths = new 
ConcurrentHashMap<>();
+  /**
+   * Only the one created the parent path can drop the parent path, so should 
remember these created parents.
+   */
+  private final List<String> pathsCreatedInThisAnnouncer = new 
CopyOnWriteArrayList<>();
+
+  private boolean started = false;
+
+  public NodeAnnouncer(CuratorFramework curator)
+  {
+    this.curator = curator;
+  }
+
+  @VisibleForTesting
+  Set<String> getAddedPaths()
+  {
+    return announcedPaths.keySet();
+  }
+
+  @LifecycleStart
+  @SuppressWarnings("DuplicatedCode")
+  public void start()
+  {
+    log.info("Starting announcer");
+    synchronized (toAnnounce) {
+      if (started) {
+        return;
+      }
+
+      started = true;
+
+      for (Announceable announceable : toAnnounce) {
+        announce(announceable.path, announceable.bytes, 
announceable.removeParentsIfCreated);
+      }
+      toAnnounce.clear();
+
+      for (Announceable announceable : toUpdate) {
+        update(announceable.path, announceable.bytes);
+      }
+      toUpdate.clear();
+    }
+  }
+
+  @LifecycleStop
+  public void stop()
+  {
+    log.info("Stopping announcer");
+    synchronized (toAnnounce) {
+      if (!started) {
+        return;
+      }
+
+      started = false;
+
+      Closer closer = Closer.create();
+      for (NodeCache cache : listeners.values()) {
+        closer.register(cache);
+      }
+      CloseQuietly.close(closer);
+
+      for (String announcementPath : announcedPaths.keySet()) {
+        unannounce(announcementPath);
+      }
+
+      if (!pathsCreatedInThisAnnouncer.isEmpty()) {
+        final List<CuratorOp> deleteOps = new 
ArrayList<>(pathsCreatedInThisAnnouncer.size());
+        for (String parent : pathsCreatedInThisAnnouncer) {
+          try {
+            deleteOps.add(curator.transactionOp().delete().forPath(parent));
+          }
+          catch (Exception e) {
+            log.error(e, "Unable to delete parent[%s].", parent);
+          }
+        }
+        try {
+          curator.transaction().forOperations(deleteOps);
+        }
+        catch (Exception e) {
+          log.error(e, "Unable to commit transaction.");
+        }
+      }
+    }
+  }
+
+  /**
+   * Like announce(path, bytes, true).
+   */
+  public void announce(String path, byte[] bytes)
+  {
+    announce(path, bytes, true);
+  }
+
+  /**
+   * Announces the provided bytes at the given path.  Announcement means that 
it will create an ephemeral node
+   * and monitor it to make sure that it always exists until it is unannounced 
or this object is closed.
+   *
+   * @param path                  The path to announce at
+   * @param bytes                 The payload to announce
+   * @param removeParentIfCreated remove parent of "path" if we had created 
that parent
+   */
+  public void announce(String path, byte[] bytes, boolean 
removeParentIfCreated)
+  {
+    synchronized (toAnnounce) {
+      if (!started) {
+        toAnnounce.add(new Announceable(path, bytes, removeParentIfCreated));
+        return;
+      }
+    }
+
+    final String parentPath = ZKPathsUtils.getParentPath(path);
+    boolean buildParentPath = false;
+
+    byte[] value = announcedPaths.get(path);
+
+    if (value == null) {
+      try {
+        if (curator.checkExists().forPath(parentPath) == null) {
+          buildParentPath = true;
+        }
+      }
+      catch (Exception e) {
+        log.debug(e, "Problem checking if the parent existed, ignoring.");
+      }
+
+      // Synchronize to make sure that I only create a listener once.
+      synchronized (toAnnounce) {
+        if (!listeners.containsKey(path)) {
+          final NodeCache cache = new NodeCache(curator, path, true);
+          cache.getListenable().addListener(
+              () -> {
+                ChildData currentData = cache.getCurrentData();
+                if (currentData == null) {
+                  final byte[] value1 = announcedPaths.get(path);
+                  if (value1 != null) {
+                    log.info("Node[%s] dropped, reinstating.", path);
+                    createAnnouncement(path, value1);
+                  }
+                }
+              }
+          );
+
+          if (started) {
+            if (buildParentPath) {
+              createPath(parentPath, removeParentIfCreated);
+            }
+            startCache(cache);
+            listeners.put(path, cache);
+          }
+        }
+      }
+    }
+
+    boolean created = false;
+    synchronized (toAnnounce) {
+      if (started) {
+        byte[] oldBytes = announcedPaths.putIfAbsent(path, bytes);
 
 Review comment:
   The above suggestion is bad if we assume that `announce()` could be called 
concurrently with stop(). If this is the case, this should be explained in 
concurrent control flow documentation for this class and/or methods.

----------------------------------------------------------------
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:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to