(hive) branch master updated: HIVE-28569: ReExecuteLostAMQueryPlugin doesn't kick in if the message contains line break and diagnostics (#5499) (Laszlo Bodor reviewed by Denys Kuzmenko)

[abstractdog]

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

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


The following commit(s) were added to refs/heads/master by this push:
     new 4d964df5918 HIVE-28569: ReExecuteLostAMQueryPlugin doesn't kick in if 
the message contains line break and diagnostics (#5499) (Laszlo Bodor reviewed 
by Denys Kuzmenko)
4d964df5918 is described below

commit 4d964df5918e36c7cf51d923db895fcce0258ca8
Author: Bodor Laszlo <bodorlaszlo0...@gmail.com>
AuthorDate: Thu Oct 10 13:19:17 2024 +0200

    HIVE-28569: ReExecuteLostAMQueryPlugin doesn't kick in if the message 
contains line break and diagnostics (#5499) (Laszlo Bodor reviewed by Denys 
Kuzmenko)
---
 .../hadoop/hive/ql/reexec/ReExecuteLostAMQueryPlugin.java      |  2 +-
 .../hadoop/hive/ql/reexec/TestReExecuteLostAMQueryPlugin.java  | 10 ++++++++++
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git 
a/ql/src/java/org/apache/hadoop/hive/ql/reexec/ReExecuteLostAMQueryPlugin.java 
b/ql/src/java/org/apache/hadoop/hive/ql/reexec/ReExecuteLostAMQueryPlugin.java
index c3a6c8a363b..8e2b6adbbc9 100644
--- 
a/ql/src/java/org/apache/hadoop/hive/ql/reexec/ReExecuteLostAMQueryPlugin.java
+++ 
b/ql/src/java/org/apache/hadoop/hive/ql/reexec/ReExecuteLostAMQueryPlugin.java
@@ -39,7 +39,7 @@ public class ReExecuteLostAMQueryPlugin implements 
IReExecutionPlugin {
   // Lost am container have exit code -100, due to node failures. This pattern 
of exception is thrown when
   // AM is managed by HS2.
   private static final Pattern LOST_AM_CONTAINER_ERROR_PATTERN =
-      Pattern.compile(".*AM Container for .* exited .* exitCode: -100.*");
+      Pattern.compile(".*AM Container for .* exited .* exitCode: -100.*", 
Pattern.DOTALL);
   // When HS2 does not manage the AMs, tez AMs are registered with zookeeper 
and HS2 discovers it,
   // failure of unmanaged AMs will throw AM record not being found in 
zookeeper.
   private static final String UNMANAGED_AM_FAILURE = "AM record not found 
(likely died)";
diff --git 
a/ql/src/test/org/apache/hadoop/hive/ql/reexec/TestReExecuteLostAMQueryPlugin.java
 
b/ql/src/test/org/apache/hadoop/hive/ql/reexec/TestReExecuteLostAMQueryPlugin.java
index 1d29f324945..da926fe4484 100644
--- 
a/ql/src/test/org/apache/hadoop/hive/ql/reexec/TestReExecuteLostAMQueryPlugin.java
+++ 
b/ql/src/test/org/apache/hadoop/hive/ql/reexec/TestReExecuteLostAMQueryPlugin.java
@@ -31,6 +31,16 @@ public class TestReExecuteLostAMQueryPlugin {
     testReExecuteWithExceptionMessage("AM record not found (likely died)");
   }
 
+  @Test
+  public void testRetryOnLostAmContainerMessageWithLineBreak() throws 
Exception {
+    testReExecuteWithExceptionMessage("Application 
application_1728328561547_0042 failed 1 times (global limit =5; " +
+        "local limit is =1) due to AM Container for 
appattempt_1728328561547_0042_000001 exited with  exitCode: " +
+        "-100\nFailing this attempt.Diagnostics: Container released on a 
*lost* nodeFor more detailed output, check " +
+        "the application tracking page: 
https://v2h0231.sjc.cloudera.com:8090/cluster/app/application_1728328561547_0042";
 +
+        " Then click on links to logs of each attempt.\n" +
+        ". Failing the application.");
+  }
+
   @Test
   public void testRetryOnNoCurrentDAGException() throws Exception {
     testReExecuteWithExceptionMessage("No running DAG at present");