HADOOP-16792: Make S3 client request timeout configurable.

Contributed by Mustafa Iman.

This adds a new configuration option fs.s3a.connection.request.timeout
to declare the time out on HTTP requests to the AWS service;
0 means no timeout.
Measured in seconds; the usual time suffixes are all supported

Important: this is the maximum duration of any AWS service call,
including upload and copy operations. If non-zero, it must be larger
than the time to upload multi-megabyte blocks to S3 from the client,
and to rename many-GB files. Use with care.

Change-Id: I407745341068b702bf8f401fb96450a9f987c51c

hadoop-common-project/hadoop-common/src/main/resources/core-default.xml

@@ -1940,6 +1940,23 @@
   </description>
 </property>
 
+<property>
+  <name>fs.s3a.connection.request.timeout</name>
+  <value>0</value>
+  <description>
+    Time out on HTTP requests to the AWS service; 0 means no timeout.
+    Measured in seconds; the usual time suffixes are all supported
+
+    Important: this is the maximum duration of any AWS service call,
+    including upload and copy operations. If non-zero, it must be larger
+    than the time to upload multi-megabyte blocks to S3 from the client,
+    and to rename many-GB files. Use with care.
+
+    Values that are larger than Integer.MAX_VALUE milliseconds are
+    converged to Integer.MAX_VALUE milliseconds
+  </description>
+</property>
+
 <property>
   <name>fs.s3a.etag.checksum.enabled</name>
   <value>false</value>

hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/Constants.java

@@ -187,6 +187,11 @@ private Constants() {
   public static final String SOCKET_TIMEOUT = "fs.s3a.connection.timeout";
   public static final int DEFAULT_SOCKET_TIMEOUT = 200000;
 
+  // milliseconds until a request is timed-out
+  public static final String REQUEST_TIMEOUT =
+      "fs.s3a.connection.request.timeout";
+  public static final int DEFAULT_REQUEST_TIMEOUT = 0;
+
   // socket send buffer to be used in Amazon client
   public static final String SOCKET_SEND_BUFFER = "fs.s3a.socket.send.buffer";
   public static final int DEFAULT_SOCKET_SEND_BUFFER = 8 * 1024;

hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/S3AUtils.java

@@ -82,6 +82,7 @@
 import java.util.Optional;
 import java.util.Set;
 import java.util.concurrent.ExecutionException;
+import java.util.concurrent.TimeUnit;
 
 import static org.apache.commons.lang3.StringUtils.isEmpty;
 import static org.apache.hadoop.fs.s3a.Constants.*;
@@ -1284,6 +1285,15 @@ public static void initConnectionSettings(Configuration conf,
         DEFAULT_SOCKET_SEND_BUFFER, 2048);
     int sockRecvBuffer = intOption(conf, SOCKET_RECV_BUFFER,
         DEFAULT_SOCKET_RECV_BUFFER, 2048);
+    long requestTimeoutMillis = conf.getTimeDuration(REQUEST_TIMEOUT,
+        DEFAULT_REQUEST_TIMEOUT, TimeUnit.SECONDS, TimeUnit.MILLISECONDS);
+
+    if (requestTimeoutMillis > Integer.MAX_VALUE) {
+      LOG.debug("Request timeout is too high({} ms). Setting to {} ms instead",
+          requestTimeoutMillis, Integer.MAX_VALUE);
+      requestTimeoutMillis = Integer.MAX_VALUE;
+    }
+    awsConf.setRequestTimeout((int) requestTimeoutMillis);
     awsConf.setSocketBufferSizeHints(sockSendBuffer, sockRecvBuffer);
     String signerOverride = conf.getTrimmed(SIGNING_ALGORITHM, "");
     if (!signerOverride.isEmpty()) {

hadoop-tools/hadoop-aws/src/site/markdown/tools/hadoop-aws/index.md

@@ -983,6 +983,23 @@ options are covered in [Testing](./testing.md).
   <description>Select which version of the S3 SDK's List Objects API to use.
   Currently support 2 (default) and 1 (older API).</description>
 </property>
+
+<property>
+  <name>fs.s3a.connection.request.timeout</name>
+  <value>0</value>
+  <description>
+  Time out on HTTP requests to the AWS service; 0 means no timeout.
+  Measured in seconds; the usual time suffixes are all supported
+
+  Important: this is the maximum duration of any AWS service call,
+  including upload and copy operations. If non-zero, it must be larger
+  than the time to upload multi-megabyte blocks to S3 from the client,
+  and to rename many-GB files. Use with care.
+
+  Values that are larger than Integer.MAX_VALUE milliseconds are
+  converged to Integer.MAX_VALUE milliseconds
+  </description>
+</property>
 ```
 
 ## <a name="retry_and_recovery"></a>Retry and Recovery

hadoop-tools/hadoop-aws/src/site/markdown/tools/hadoop-aws/troubleshooting_s3a.md

@@ -1384,3 +1384,43 @@ For this reason, the number of retry events are limited.
   </description>
 </property>
 ```
+
+### <a name="aws-timeouts"></a> Tuning AWS request timeouts
+
+It is possible to configure a global timeout for AWS service calls using following property:
+
+```xml
+<property>
+  <name>fs.s3a.connection.request.timeout</name>
+  <value>0</value>
+  <description>
+    Time out on HTTP requests to the AWS service; 0 means no timeout.
+    Measured in seconds; the usual time suffixes are all supported
+
+    Important: this is the maximum duration of any AWS service call,
+    including upload and copy operations. If non-zero, it must be larger
+    than the time to upload multi-megabyte blocks to S3 from the client,
+    and to rename many-GB files. Use with care.
+
+    Values that are larger than Integer.MAX_VALUE milliseconds are
+    converged to Integer.MAX_VALUE milliseconds
+  </description>
+</property>
+```
+
+If this value is configured too low, user may encounter `SdkClientException`s due to many requests
+timing-out.
+
+```
+com.amazonaws.SdkClientException: Unable to execute HTTP request:
+  Request did not complete before the request timeout configuration.:
+  Unable to execute HTTP request: Request did not complete before the request timeout configuration.
+  at org.apache.hadoop.fs.s3a.S3AUtils.translateException(S3AUtils.java:205)
+  at org.apache.hadoop.fs.s3a.Invoker.once(Invoker.java:112)
+  at org.apache.hadoop.fs.s3a.Invoker.lambda$retry$4(Invoker.java:315)
+  at org.apache.hadoop.fs.s3a.Invoker.retryUntranslated(Invoker.java:407)
+  at org.apache.hadoop.fs.s3a.Invoker.retry(Invoker.java:311)
+```
+
+When this happens, try to set `fs.s3a.connection.request.timeout` to a larger value or disable it
+completely by setting it to `0`.

hadoop-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/s3a/ITestS3AConfiguration.java

@@ -390,6 +390,19 @@ public void testCustomUserAgent() throws Exception {
         awsConf.getUserAgentPrefix());
   }
 
+  @Test
+  public void testRequestTimeout() throws Exception {
+    conf = new Configuration();
+    conf.set(REQUEST_TIMEOUT, "120");
+    fs = S3ATestUtils.createTestFileSystem(conf);
+    AmazonS3 s3 = fs.getAmazonS3ClientForTesting("Request timeout (ms)");
+    ClientConfiguration awsConf = getField(s3, ClientConfiguration.class,
+        "clientConfiguration");
+    assertEquals("Configured " + REQUEST_TIMEOUT +
+        " is different than what AWS sdk configuration uses internally",
+        120000, awsConf.getRequestTimeout());
+  }
+
   @Test
   public void testCloseIdempotent() throws Throwable {
     conf = new Configuration();