From 5e83a21cb66c78e89ac5af9a130ab0aee596a9f4 Mon Sep 17 00:00:00 2001
From: Wei-Chiu Chuang <weichiu@apache.org>
Date: Thu, 20 Oct 2016 13:02:16 -0700
Subject: [PATCH] HDFS-10976. Report erasure coding policy of EC files in Fsck.
 Contributed by Wei-Chiu Chuang.

---
 .../hadoop/hdfs/server/namenode/NamenodeFsck.java  | 14 ++++++++++++--
 .../hadoop/hdfs/server/namenode/TestFsck.java      | 14 +++++++++++++-
 2 files changed, 25 insertions(+), 3 deletions(-)

diff --git a/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/NamenodeFsck.java b/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/NamenodeFsck.java
index 8302035348..a2e249d49d 100644
--- a/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/NamenodeFsck.java
+++ b/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/NamenodeFsck.java
@@ -55,6 +55,7 @@
 import org.apache.hadoop.hdfs.protocol.DatanodeInfo;
 import org.apache.hadoop.hdfs.protocol.DatanodeInfoWithStorage;
 import org.apache.hadoop.hdfs.protocol.DirectoryListing;
+import org.apache.hadoop.hdfs.protocol.ErasureCodingPolicy;
 import org.apache.hadoop.hdfs.protocol.ExtendedBlock;
 import org.apache.hadoop.hdfs.protocol.HdfsConstants;
 import org.apache.hadoop.hdfs.protocol.HdfsFileStatus;
@@ -540,11 +541,20 @@ private void collectFileSummary(String path, HdfsFileStatus file, Result res,
     res.totalFiles++;
     res.totalSize += fileLen;
     res.totalBlocks += blocks.locatedBlockCount();
+    String redundancyPolicy;
+    ErasureCodingPolicy ecPolicy = file.getErasureCodingPolicy();
+    if (ecPolicy == null) { // a replicated file
+      redundancyPolicy = "replicated: replication=" +
+          file.getReplication() + ",";
+    } else {
+      redundancyPolicy = "erasure-coded: policy=" + ecPolicy.getName() + ",";
+    }
+
     if (showOpenFiles && isOpen) {
-      out.print(path + " " + fileLen + " bytes, " +
+      out.print(path + " " + fileLen + " bytes, " + redundancyPolicy + " " +
         blocks.locatedBlockCount() + " block(s), OPENFORWRITE: ");
     } else if (showFiles) {
-      out.print(path + " " + fileLen + " bytes, " +
+      out.print(path + " " + fileLen + " bytes, " + redundancyPolicy + " " +
         blocks.locatedBlockCount() + " block(s): ");
     } else if (showprogress) {
       out.print('.');
diff --git a/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestFsck.java b/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestFsck.java
index aa41e9b092..254a86ce39 100644
--- a/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestFsck.java
+++ b/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestFsck.java
@@ -1700,9 +1700,21 @@ public void testECFsck() throws Exception {
     // restart the cluster; bring up namenode but not the data nodes
     cluster = new MiniDFSCluster.Builder(conf)
         .numDataNodes(0).format(false).build();
-    outStr = runFsck(conf, 1, true, "/");
+    outStr = runFsck(conf, 1, true, "/", "-files", "-blocks");
     // expect the result is corrupt
     assertTrue(outStr.contains(NamenodeFsck.CORRUPT_STATUS));
+    String[] outLines = outStr.split("\\r?\\n");
+    for (String line: outLines) {
+      if (line.contains(largeFilePath.toString())) {
+        final HdfsFileStatus file = cluster.getNameNode().getRpcServer().
+            getFileInfo(largeFilePath.toString());
+        assertTrue(line.contains("policy=" +
+            file.getErasureCodingPolicy().getName()));
+      } else if (line.contains(replFilePath.toString())) {
+        assertTrue(line.contains("replication=" + cluster.getFileSystem().
+            getFileStatus(replFilePath).getReplication()));
+      }
+    }
     System.out.println(outStr);
   }