From 7d6b133af3f7ee4b986bdd2b266b83693614b6ab Mon Sep 17 00:00:00 2001
From: Stephen O'Donnell <stephen.odonnell@gmail.com>
Date: Wed, 1 Jun 2022 20:36:01 +0100
Subject: [PATCH] HDFS-16610. Make fsck read timeout configurable (#4384)

(cherry picked from commit 34a973a90ef89b633c9b5c13a79aa1ac11c92eb5)

 Conflicts:
	hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml
---
 .../hdfs/client/HdfsClientConfigKeys.java     |  8 ++++++++
 .../org/apache/hadoop/hdfs/tools/DFSck.java   | 13 ++++++++++++-
 .../src/main/resources/hdfs-default.xml       | 19 +++++++++++++++++++
 3 files changed, 39 insertions(+), 1 deletion(-)

diff --git a/hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/client/HdfsClientConfigKeys.java b/hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/client/HdfsClientConfigKeys.java
index cf12793c868..e6af2dbecc5 100755
--- a/hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/client/HdfsClientConfigKeys.java
+++ b/hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/client/HdfsClientConfigKeys.java
@@ -261,6 +261,14 @@ public interface HdfsClientConfigKeys {
   String DFS_LEASE_HARDLIMIT_KEY = "dfs.namenode.lease-hard-limit-sec";
   long DFS_LEASE_HARDLIMIT_DEFAULT = 20 * 60;
 
+  String DFS_CLIENT_FSCK_CONNECT_TIMEOUT =
+      "dfs.client.fsck.connect.timeout";
+  int DFS_CLIENT_FSCK_CONNECT_TIMEOUT_DEFAULT = 60 * 1000;
+
+  String DFS_CLIENT_FSCK_READ_TIMEOUT =
+      "dfs.client.fsck.read.timeout";
+  int DFS_CLIENT_FSCK_READ_TIMEOUT_DEFAULT = 60 * 1000;
+
   /**
    * These are deprecated config keys to client code.
    */
diff --git a/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/tools/DFSck.java b/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/tools/DFSck.java
index 8a2ef8b5920..33a117f314e 100644
--- a/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/tools/DFSck.java
+++ b/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/tools/DFSck.java
@@ -27,6 +27,7 @@
 import java.net.URLConnection;
 import java.net.URLEncoder;
 import java.security.PrivilegedExceptionAction;
+import java.util.concurrent.TimeUnit;
 
 import org.apache.hadoop.classification.InterfaceAudience;
 import org.apache.hadoop.conf.Configuration;
@@ -37,6 +38,7 @@
 import org.apache.hadoop.hdfs.DistributedFileSystem;
 import org.apache.hadoop.hdfs.HAUtil;
 import org.apache.hadoop.hdfs.HdfsConfiguration;
+import org.apache.hadoop.hdfs.client.HdfsClientConfigKeys;
 import org.apache.hadoop.hdfs.server.namenode.NamenodeFsck;
 import org.apache.hadoop.hdfs.web.URLConnectionFactory;
 import org.apache.hadoop.security.UserGroupInformation;
@@ -137,8 +139,17 @@ public DFSck(Configuration conf, PrintStream out) throws IOException {
     super(conf);
     this.ugi = UserGroupInformation.getCurrentUser();
     this.out = out;
+    int connectTimeout = (int) conf.getTimeDuration(
+        HdfsClientConfigKeys.DFS_CLIENT_FSCK_CONNECT_TIMEOUT,
+        HdfsClientConfigKeys.DFS_CLIENT_FSCK_CONNECT_TIMEOUT_DEFAULT,
+        TimeUnit.MILLISECONDS);
+    int readTimeout = (int) conf.getTimeDuration(
+        HdfsClientConfigKeys.DFS_CLIENT_FSCK_READ_TIMEOUT,
+        HdfsClientConfigKeys.DFS_CLIENT_FSCK_READ_TIMEOUT_DEFAULT,
+        TimeUnit.MILLISECONDS);
+
     this.connectionFactory = URLConnectionFactory
-        .newDefaultURLConnectionFactory(conf);
+        .newDefaultURLConnectionFactory(connectTimeout, readTimeout, conf);
     this.isSpnegoEnabled = UserGroupInformation.isSecurityEnabled();
   }
 
diff --git a/hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml b/hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml
index 0b573ddaf64..f8abb59daa0 100755
--- a/hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml
+++ b/hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml
@@ -6174,4 +6174,23 @@
       set on fs.protected.directories.
     </description>
   </property>
+
+  <property>
+    <name>dfs.client.fsck.connect.timeout</name>
+    <value>60000ms</value>
+    <description>
+      The amount of time the fsck client will wait to connect to the namenode
+      before timing out.
+    </description>
+  </property>
+
+  <property>
+    <name>dfs.client.fsck.read.timeout</name>
+    <value>60000ms</value>
+    <description>
+      The amount of time the fsck client will wait to read from the namenode
+      before timing out. If the namenode does not report progress more
+      frequently than this time, the client will give up waiting.
+    </description>
+  </property>
 </configuration>