当查询具有 GROUP BY 条款时，Hive 查询抛出“来自 org.apache.hadoop.hive.ql.exec.tez.TezTask 的代码 2”异常

问题描述

我在 HDP 3.1.4 上安装了 Hive + LLAP

Hive 和 Tez 配置是：

yarn.nodemanager.resource.memory-mb = 40960
yarn.scheduler.minimum-allocation-mb = 1024
yarn.scheduler.maximum-allocation-mb = 40960
hive.tez.container.size = 4096
num_llap_nodes=4
hive.llap.daemon.num.executors=8
hive.llap.daemon.yarn.container.mb = 35840
llap_headroom_space=2048
llap_heap_size=32768
hive.llap.io.memory.size=1024
tez.am.resource.memory.mb=4096
hive.tez.java.opts=-server -Djava.net.preferIPv4Stack=true -XX:NewRatio=8 -XX:+UseNUMA -XX:+UseG1GC -XX:+ResizeTLAB -XX:+PrintGCDetails -verbose:gc -XX:+PrintGCTimeStamps **-Xmx3276m**
tez.runtime.io.sort.mb= 1638
tez.runtime.unordered.output.buffer.size-mb=409

以下查询运行正常：

select count(*) from balance;

但是当在以下查询中使用 group by 表达式时：

select count(*),jobdate from balance group by jobdate;

我 我尝试了很多配置，但抛出了这个长异常：

ERROR: Error while processing statement: **FAILED: Execution Error,return code 2 from org.apache.hadoop.hive.ql.exec.tez.TezTask.** 
    Vertex failed,vertexName=Map 1,vertexId=vertex_1617520101397_0014_1_00,diagnostics=[Task 
    failed,taskId=task_1617520101397_0014_1_00_000013,diagnostics=[TaskAttempt 0 failed,**info=[Error: Error while 
    running task ( failure ) : java.lang.NoClassDefFoundError: Could 
    not initialize class 
    org.apache.tez.runtime.library.api.TezRuntimeConfiguration**    at 

    **BLABLA**
        at java.lang.Thread.run(Thread.java:748) ]],Task failed,taskId=task_1617520101397_0014_1_00_000006,info=[Error: Error while 
    running task ( failure ) : java.lang.NoClassDefFoundError: Could 
    not initialize class 
    org.apache.tez.runtime.library.api.TezRuntimeConfiguration  at 
        at java.lang.Thread.run(Thread.java:748) ]],taskId=task_1617520101397_0014_1_00_000005,info=[Error: Error while 
    running task ( failure ) : java.lang.NoClassDefFoundError: Could 
    not initialize class 
    org.apache.tez.runtime.library.api.TezRuntimeConfiguration  at 
    org.apache.tez.runtime.library.output.OrderedPartitionedKVOutput.start(OrderedPartitionedKVOutput.java:111) 
    java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) 
        at java.lang.Thread.run(Thread.java:748) ]],**Vertex did not 
    succeed due to OWN_TASK_FAILURE,failedTasks:9 killedTasks:31761,Vertex vertex_1617520101397_0014_1_00 [Map 1] killed/failed due 
    to:OWN_TASK_FAILURE]Vertex killed,vertexName=Reducer 2,vertexId=vertex_1617520101397_0014_1_01,diagnostics=[Vertex 
    received Kill while in RUNNING state.,Vertex did not succeed due 
    to OTHER_VERTEX_FAILURE,failedTasks:0 killedTasks:18,Vertex 
    vertex_1617520101397_0014_1_01 [Reducer 2] killed/failed due 
    to:OTHER_VERTEX_FAILURE]DAG did not succeed due to 
    VERTEX_FAILURE. failedVertices:1 killedVertices:1 Error Code: 2**

解决方法

Ambari Hive Config 页面中有两个部分用于设置 hive.tez.container.size。其中一个出现在 SETTINGS 标签中，另一个与 LLAP 相关的出现在 ADVANCED 标签中的 Advanced hive-interactive-site 下.我正在尝试使用 hive.tez.container.size 值设置选项卡而不是 Advanced hive-interactive-site 部分。最后，我设置了以下配置并解决了错误：

set hive.tez.container.size=10240;
set hive.tez.java.opts=-Xmx9216m;
set tez.runtime.io.sort.mb=3072;
set tez.runtime.unordered.output.buffer.size-mb=1024;