From 6c1e484d36ffdcb654b0cf3ef0a74924ad120cb4 Mon Sep 17 00:00:00 2001
From: Michael Stack <stack@apache.org>
Date: Sat, 13 Apr 2013 17:07:14 +0000
Subject: [PATCH] HBASE-8096 [replication] NPE while replicating a log that is
 acquiring a new block from HDFS

git-svn-id: https://svn.apache.org/repos/asf/hbase/trunk@1467662 13f79535-47bb-0310-9956-ffa450edef68
---
 .../regionserver/ReplicationHLogReaderManager.java    |  6 +++++-
 .../replication/regionserver/ReplicationSource.java   | 11 ++++++++---
 .../hbase/replication/TestReplicationSmallTests.java  |  2 +-
 3 files changed, 14 insertions(+), 5 deletions(-)

diff --git a/hbase-server/src/main/java/org/apache/hadoop/hbase/replication/regionserver/ReplicationHLogReaderManager.java b/hbase-server/src/main/java/org/apache/hadoop/hbase/replication/regionserver/ReplicationHLogReaderManager.java
index 82b7023bec2..fe6924c91e0 100644
--- a/hbase-server/src/main/java/org/apache/hadoop/hbase/replication/regionserver/ReplicationHLogReaderManager.java
+++ b/hbase-server/src/main/java/org/apache/hadoop/hbase/replication/regionserver/ReplicationHLogReaderManager.java
@@ -68,7 +68,11 @@ public class ReplicationHLogReaderManager {
       this.reader = HLogFactory.createReader(this.fs, path, this.conf);
       this.lastPath = path;
     } else {
-      this.reader.reset();
+      try {
+        this.reader.reset();
+      } catch (NullPointerException npe) {
+        throw new IOException("NPE resetting reader, likely HDFS-4380", npe);
+      }
     }
     return this.reader;
   }
diff --git a/hbase-server/src/main/java/org/apache/hadoop/hbase/replication/regionserver/ReplicationSource.java b/hbase-server/src/main/java/org/apache/hadoop/hbase/replication/regionserver/ReplicationSource.java
index cac32ae01fe..bde35466ac7 100644
--- a/hbase-server/src/main/java/org/apache/hadoop/hbase/replication/regionserver/ReplicationSource.java
+++ b/hbase-server/src/main/java/org/apache/hadoop/hbase/replication/regionserver/ReplicationSource.java
@@ -622,9 +622,14 @@ public class ReplicationSource extends Thread
     } catch (IOException ioe) {
       LOG.warn(peerClusterZnode + " Got: ", ioe);
       this.reader = null;
-      // TODO Need a better way to determinate if a file is really gone but
-      // TODO without scanning all logs dir
-      if (sleepMultiplier == this.maxRetriesMultiplier) {
+      if (ioe.getCause() instanceof NullPointerException) {
+        // Workaround for race condition in HDFS-4380
+        // which throws a NPE if we open a file before any data node has the most recent block
+        // Just sleep and retry. Will require re-reading compressed HLogs for compressionContext.
+        LOG.warn("Got NPE opening reader, will retry.");
+      } else if (sleepMultiplier == this.maxRetriesMultiplier) {
+        // TODO Need a better way to determine if a file is really gone but
+        // TODO without scanning all logs dir
         LOG.warn("Waited too long for this file, considering dumping");
         return !processEndOfFile();
       }
diff --git a/hbase-server/src/test/java/org/apache/hadoop/hbase/replication/TestReplicationSmallTests.java b/hbase-server/src/test/java/org/apache/hadoop/hbase/replication/TestReplicationSmallTests.java
index 374ea6848de..b9219668db4 100644
--- a/hbase-server/src/test/java/org/apache/hadoop/hbase/replication/TestReplicationSmallTests.java
+++ b/hbase-server/src/test/java/org/apache/hadoop/hbase/replication/TestReplicationSmallTests.java
@@ -428,7 +428,7 @@ public class TestReplicationSmallTests extends TestReplicationBase {
     Result[] res = scanner.next(NB_ROWS_IN_BIG_BATCH);
     scanner.close();
 
-    assertEquals(NB_ROWS_IN_BATCH *10, res.length);
+    assertEquals(NB_ROWS_IN_BIG_BATCH, res.length);
 
     scan = new Scan();