[ https://issues.apache.org/jira/browse/HIVE-27087?focusedWorklogId=851210&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-851210 ]
ASF GitHub Bot logged work on HIVE-27087: ----------------------------------------- Author: ASF GitHub Bot Created on: 15/Mar/23 20:26 Start Date: 15/Mar/23 20:26 Worklog Time Spent: 10m Work Description: vihangk1 commented on PR #4067: URL: https://github.com/apache/hive/pull/4067#issuecomment-1470792448 > @vihangk1 Did you debug the creation of the file locally to see which codepath sets the replication factor to 1? Maybe there is some logic to adapt the replication factor based on other things leading to the different behavior when running in CI. @zabetak I tried to look into this but there was nothing obvious that came out. I am not convinced that this is a bug yet. The replication factor is decided by the HDFS configuration `dfs.replication.factor` and I don't see any code in HoS or Hive which would affect that. Can we consider masking the replication factor for these tests for spark tests? If this was a real bug, someone would have reported it by now since it also happens on 3.1.3 so there is a high likelihood that this is related to our test environment. Issue Time Tracking ------------------- Worklog Id: (was: 851210) Time Spent: 3h 20m (was: 3h 10m) > Fix TestMiniSparkOnYarnCliDriver test failures on branch-3 > ---------------------------------------------------------- > > Key: HIVE-27087 > URL: https://issues.apache.org/jira/browse/HIVE-27087 > Project: Hive > Issue Type: Sub-task > Reporter: Vihang Karajgaonkar > Assignee: Vihang Karajgaonkar > Priority: Major > Labels: pull-request-available > Time Spent: 3h 20m > Remaining Estimate: 0h > > TestMiniSparkOnYarnCliDriver are failing with the error below > [ERROR] 2023-02-16 14:13:08.991 [Driver] SparkContext - Error initializing > SparkContext. > java.lang.RuntimeException: java.lang.NoSuchFieldException: > DEFAULT_TINY_CACHE_SIZE > at > org.apache.spark.network.util.NettyUtils.getPrivateStaticField(NettyUtils.java:131) > ~[spark-network-common_2.11-2.3.0.jar:2.3.0] > at > org.apache.spark.network.util.NettyUtils.createPooledByteBufAllocator(NettyUtils.java:118) > ~[spark-network-common_2.11-2.3.0.jar:2.3.0] > at > org.apache.spark.network.server.TransportServer.init(TransportServer.java:94) > ~[spark-network-common_2.11-2.3.0.jar:2.3.0] > at > org.apache.spark.network.server.TransportServer.<init>(TransportServer.java:73) > ~[spark-network-common_2.11-2.3.0.jar:2.3.0] > at > org.apache.spark.network.TransportContext.createServer(TransportContext.java:114) > ~[spark-network-common_2.11-2.3.0.jar:2.3.0] > at org.apache.spark.rpc.netty.NettyRpcEnv.startServer(NettyRpcEnv.scala:119) > ~[spark-core_2.11-2.3.0.jar:2.3.0] > at > org.apache.spark.rpc.netty.NettyRpcEnvFactory$$anonfun$4.apply(NettyRpcEnv.scala:465) > ~[spark-core_2.11-2.3.0.jar:2.3.0] > at > org.apache.spark.rpc.netty.NettyRpcEnvFactory$$anonfun$4.apply(NettyRpcEnv.scala:464) > ~[spark-core_2.11-2.3.0.jar:2.3.0] > at > org.apache.spark.util.Utils$$anonfun$startServiceOnPort$1.apply$mcVI$sp(Utils.scala:2271) > ~[spark-core_2.11-2.3.0.jar:2.3.0] > at scala.collection.immutable.Range.foreach$mVc$sp(Range.scala:160) > ~[scala-library-2.11.8.jar:?] > at org.apache.spark.util.Utils$.startServiceOnPort(Utils.scala:2263) > ~[spark-core_2.11-2.3.0.jar:2.3.0] > at > org.apache.spark.rpc.netty.NettyRpcEnvFactory.create(NettyRpcEnv.scala:469) > ~[spark-core_2.11-2.3.0.jar:2.3.0] > at org.apache.spark.rpc.RpcEnv$.create(RpcEnv.scala:57) > ~[spark-core_2.11-2.3.0.jar:2.3.0] > at org.apache.spark.SparkEnv$.create(SparkEnv.scala:249) > ~[spark-core_2.11-2.3.0.jar:2.3.0] > at org.apache.spark.SparkEnv$.createDriverEnv(SparkEnv.scala:175) > ~[spark-core_2.11-2.3.0.jar:2.3.0] > at org.apache.spark.SparkContext.createSparkEnv(SparkContext.scala:256) > [spark-core_2.11-2.3.0.jar:2.3.0] > at org.apache.spark.SparkContext.<init>(SparkContext.scala:423) > [spark-core_2.11-2.3.0.jar:2.3.0] > at > org.apache.spark.api.java.JavaSparkContext.<init>(JavaSparkContext.scala:58) > [spark-core_2.11-2.3.0.jar:2.3.0] > at org.apache.hive.spark.client.RemoteDriver.<init>(RemoteDriver.java:161) > [hive-exec-3.2.0-SNAPSHOT.jar:3.2.0-SNAPSHOT] > at org.apache.hive.spark.client.RemoteDriver.main(RemoteDriver.java:536) > [hive-exec-3.2.0-SNAPSHOT.jar:3.2.0-SNAPSHOT] > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) ~[?:1.8.0_322] > at > sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) > ~[?:1.8.0_322] > The root cause of the problem is that we upgrade the netty library from > 4.1.17.Final to 4.1.69.Final. The upgraded library does not have > `DEFAULT_TINY_CACHE_SIZE` field > [here|https://github.com/netty/netty/blob/netty-4.1.51.Final/buffer/src/main/java/io/netty/buffer/PooledByteBufAllocator.java#L46] > which was removed in 4.1.52.Final -- This message was sent by Atlassian Jira (v8.20.10#820010)