You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-commits@hadoop.apache.org by sj...@apache.org on 2015/09/26 18:05:56 UTC

[49/50] [abbrv] hadoop git commit: YARN-2019. Retrospect on decision of making RM crashed if any exception throw in ZKRMStateStore. Contributed by Jian He.

YARN-2019. Retrospect on decision of making RM crashed if any exception throw in ZKRMStateStore. Contributed by Jian He.

(cherry picked from commit db57d91ac91e895bcb9a23fa50af0b2fbcb1db5a)


Project: http://git-wip-us.apache.org/repos/asf/hadoop/repo
Commit: http://git-wip-us.apache.org/repos/asf/hadoop/commit/d27f09c9
Tree: http://git-wip-us.apache.org/repos/asf/hadoop/tree/d27f09c9
Diff: http://git-wip-us.apache.org/repos/asf/hadoop/diff/d27f09c9

Branch: refs/heads/branch-2.6
Commit: d27f09c9369d9f2604eb31deaff50f4bd5aa98ac
Parents: c09bb46
Author: Xuan <xg...@apache.org>
Authored: Mon Sep 7 17:34:33 2015 -0700
Committer: Sangjin Lee <sj...@apache.org>
Committed: Fri Sep 25 16:30:49 2015 -0700

----------------------------------------------------------------------
 hadoop-yarn-project/CHANGES.txt                     |  3 +++
 .../apache/hadoop/yarn/conf/YarnConfiguration.java  | 11 +++++++++++
 .../src/main/resources/yarn-default.xml             | 16 ++++++++++++++++
 .../resourcemanager/recovery/RMStateStore.java      |  9 +++++++--
 4 files changed, 37 insertions(+), 2 deletions(-)
----------------------------------------------------------------------


http://git-wip-us.apache.org/repos/asf/hadoop/blob/d27f09c9/hadoop-yarn-project/CHANGES.txt
----------------------------------------------------------------------
diff --git a/hadoop-yarn-project/CHANGES.txt b/hadoop-yarn-project/CHANGES.txt
index 6d016b7..15d082b 100644
--- a/hadoop-yarn-project/CHANGES.txt
+++ b/hadoop-yarn-project/CHANGES.txt
@@ -6,6 +6,9 @@ Release 2.6.2 - UNRELEASED
 
   NEW FEATURES
 
+    YARN-2019. Retrospect on decision of making RM crashed if any exception throw 
+    in ZKRMStateStore. (Jian He via junping_du)
+
   IMPROVEMENTS
 
     YARN-4092. Fixed UI redirection to print useful messages when both RMs are

http://git-wip-us.apache.org/repos/asf/hadoop/blob/d27f09c9/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-api/src/main/java/org/apache/hadoop/yarn/conf/YarnConfiguration.java
----------------------------------------------------------------------
diff --git a/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-api/src/main/java/org/apache/hadoop/yarn/conf/YarnConfiguration.java b/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-api/src/main/java/org/apache/hadoop/yarn/conf/YarnConfiguration.java
index 881ecb5..471297c 100644
--- a/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-api/src/main/java/org/apache/hadoop/yarn/conf/YarnConfiguration.java
+++ b/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-api/src/main/java/org/apache/hadoop/yarn/conf/YarnConfiguration.java
@@ -380,6 +380,11 @@ public class YarnConfiguration extends Configuration {
   public static final String RECOVERY_ENABLED = RM_PREFIX + "recovery.enabled";
   public static final boolean DEFAULT_RM_RECOVERY_ENABLED = false;
 
+  public static final String YARN_FAIL_FAST = YARN_PREFIX + "fail-fast";
+  public static final boolean DEFAULT_YARN_FAIL_FAST = true;
+
+  public static final String RM_FAIL_FAST = RM_PREFIX + "fail-fast";
+
   @Private
   public static final String RM_WORK_PRESERVING_RECOVERY_ENABLED = RM_PREFIX
       + "work-preserving-recovery.enabled";
@@ -1568,6 +1573,12 @@ public class YarnConfiguration extends Configuration {
             YARN_HTTP_POLICY_DEFAULT));
   }
 
+  public static boolean shouldRMFailFast(Configuration conf) {
+    return conf.getBoolean(YarnConfiguration.RM_FAIL_FAST,
+        conf.getBoolean(YarnConfiguration.YARN_FAIL_FAST,
+            YarnConfiguration.DEFAULT_YARN_FAIL_FAST));
+  }
+
   @Private
   public static String getClusterId(Configuration conf) {
     String clusterId = conf.get(YarnConfiguration.RM_CLUSTER_ID);

http://git-wip-us.apache.org/repos/asf/hadoop/blob/d27f09c9/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/resources/yarn-default.xml
----------------------------------------------------------------------
diff --git a/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/resources/yarn-default.xml b/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/resources/yarn-default.xml
index 2ffa8b8..fa2e5cc 100644
--- a/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/resources/yarn-default.xml
+++ b/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-common/src/main/resources/yarn-default.xml
@@ -276,6 +276,22 @@
   </property>
 
   <property>
+    <description>Should RM fail fast if it encounters any errors. By defalt, it
+      points to ${yarn.fail-fast}. Errors include:
+      1) exceptions when state-store write/read operations fails.
+    </description>
+    <name>yarn.resourcemanager.fail-fast</name>
+    <value>${yarn.fail-fast}</value>
+  </property>
+
+  <property>
+    <description>Should YARN fail fast if it encounters any errors.
+    </description>
+    <name>yarn.fail-fast</name>
+    <value>true</value>
+  </property>
+
+  <property>
     <description>Enable RM work preserving recovery. This configuration is private
     to YARN for experimenting the feature.
     </description>

http://git-wip-us.apache.org/repos/asf/hadoop/blob/d27f09c9/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/recovery/RMStateStore.java
----------------------------------------------------------------------
diff --git a/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/recovery/RMStateStore.java b/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/recovery/RMStateStore.java
index 983cc81..37855f7 100644
--- a/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/recovery/RMStateStore.java
+++ b/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/recovery/RMStateStore.java
@@ -42,6 +42,7 @@ import org.apache.hadoop.yarn.api.records.Container;
 import org.apache.hadoop.yarn.api.records.ContainerExitStatus;
 import org.apache.hadoop.yarn.api.records.FinalApplicationStatus;
 import org.apache.hadoop.yarn.api.records.impl.pb.ApplicationSubmissionContextPBImpl;
+import org.apache.hadoop.yarn.conf.YarnConfiguration;
 import org.apache.hadoop.yarn.event.AsyncDispatcher;
 import org.apache.hadoop.yarn.event.Dispatcher;
 import org.apache.hadoop.yarn.event.EventHandler;
@@ -820,14 +821,18 @@ public abstract class RMStateStore extends AbstractService {
    * @param failureCause the exception due to which the operation failed
    */
   protected void notifyStoreOperationFailed(Exception failureCause) {
+    LOG.error("State store operation failed ", failureCause);
     if (failureCause instanceof StoreFencedException) {
       Thread standByTransitionThread =
           new Thread(new StandByTransitionThread());
       standByTransitionThread.setName("StandByTransitionThread Handler");
       standByTransitionThread.start();
     } else {
-      rmDispatcher.getEventHandler().handle(
-        new RMFatalEvent(RMFatalEventType.STATE_STORE_OP_FAILED, failureCause));
+      if (YarnConfiguration.shouldRMFailFast(getConfig())) {
+        rmDispatcher.getEventHandler().handle(
+            new RMFatalEvent(RMFatalEventType.STATE_STORE_OP_FAILED,
+                failureCause));
+      }
     }
   }