[zeppelin] branch master updated: [ZEPPELIN-5017]. Add retry in PooledRemoteClient

[zj...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

zjffdu pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/zeppelin.git


The following commit(s) were added to refs/heads/master by this push:
     new 621e257  [ZEPPELIN-5017]. Add retry in PooledRemoteClient
621e257 is described below

commit 621e2579d4a69b6837619ae3a4a130eb314c6fcd
Author: Jeff Zhang <zj...@apache.org>
AuthorDate: Wed Aug 26 14:34:13 2020 +0800

    [ZEPPELIN-5017]. Add retry in PooledRemoteClient
    
    ### What is this PR for?
    
    Trivial PR to add retry in PooledRemoteClient in case there's some network issue.
    
    ### What type of PR is it?
    [Improvement ]
    
    ### Todos
    * [ ] - Task
    
    ### What is the Jira issue?
    * https://issues.apache.org/jira/browse/ZEPPELIN-5017
    
    ### How should this be tested?
    * CI pass
    
    ### Screenshots (if appropriate)
    
    ### Questions:
    * Does the licenses files need update? No
    * Is there breaking changes for older versions? No
    * Does this needs documentation? No
    
    Author: Jeff Zhang <zj...@apache.org>
    
    Closes #3890 from zjffdu/ZEPPELIN-5017 and squashes the following commits:
    
    b964b884a [Jeff Zhang] add retry
    13cbde46b [Jeff Zhang] [ZEPPELIN-5017]. Add retry in PooledRemoteClient
---
 .../interpreter/remote/PooledRemoteClient.java     | 35 +++++++++++++---------
 1 file changed, 21 insertions(+), 14 deletions(-)

diff --git a/zeppelin-interpreter/src/main/java/org/apache/zeppelin/interpreter/remote/PooledRemoteClient.java b/zeppelin-interpreter/src/main/java/org/apache/zeppelin/interpreter/remote/PooledRemoteClient.java
index a219036..7d5a764 100644
--- a/zeppelin-interpreter/src/main/java/org/apache/zeppelin/interpreter/remote/PooledRemoteClient.java
+++ b/zeppelin-interpreter/src/main/java/org/apache/zeppelin/interpreter/remote/PooledRemoteClient.java
@@ -33,6 +33,7 @@ import org.slf4j.LoggerFactory;
 public class PooledRemoteClient<T extends TServiceClient> {
 
   private static final Logger LOGGER = LoggerFactory.getLogger(PooledRemoteClient.class);
+  private static final int RETRY_COUNT = 3;
 
   private GenericObjectPool<T> clientPool;
   private RemoteClientFactory<T> remoteClientFactory;
@@ -83,23 +84,29 @@ public class PooledRemoteClient<T extends TServiceClient> {
   }
 
   public <R> R callRemoteFunction(RemoteFunction<R, T> func) {
-    T client = null;
     boolean broken = false;
-    try {
-      client = getClient();
-      if (client != null) {
-        return func.call(client);
-      }
-    } catch (TException e) {
-      broken = true;
-      throw new RuntimeException(e);
-    } catch (Exception e1) {
-      throw new RuntimeException(e1);
-    } finally {
-      if (client != null) {
-        releaseClient(client, broken);
+    for (int i = 0;i < RETRY_COUNT; ++ i) {
+      T client = null;
+      broken = false;
+      try {
+        client = getClient();
+        if (client != null) {
+          return func.call(client);
+        }
+      } catch (TException e) {
+        broken = true;
+        continue;
+      } catch (Exception e1) {
+        throw new RuntimeException(e1);
+      } finally {
+        if (client != null) {
+          releaseClient(client, broken);
+        }
       }
     }
+    if (broken) {
+      throw new RuntimeException("Fail to callRemoteFunction, because connection is broken");
+    }
     return null;
   }