From 52047104cc40f0feeab90eb33ffd9fd9755f5efa Mon Sep 17 00:00:00 2001 From: Filippo Berto Date: Wed, 30 Mar 2022 09:35:53 +0200 Subject: [PATCH] Fix spark python + spark history kerberos --- nixos/big_data.nix | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) diff --git a/nixos/big_data.nix b/nixos/big_data.nix index 7566112..eef0133 100644 --- a/nixos/big_data.nix +++ b/nixos/big_data.nix @@ -12,7 +12,7 @@ let extraPythonPackages = pysparkPackageSelector pkgs.python3.pkgs; }; sparkConfDir = pkgs.stdenv.mkDerivation { - name = "spark-config"; + name = "spark-conf"; dontUnpack = true; installPhase = '' # source standard environment @@ -36,8 +36,8 @@ let export JAVA_HOME="${pkgs.jdk8}" export SPARK_HOME="${pkgs.spark}/lib/${pkgs.spark.untarDir}" export SPARK_DIST_CLASSPATH=$(${pkgs.hadoop}/bin/hadoop classpath) - export PYSPARK_PYTHON="${pysparkEnv.outPath}/bin/${pysparkEnv.executable}:" - export PYSPARK_DRIVER_PYTHON="${pysparkEnv.outPath}/bin/${pysparkEnv.executable}:" + export PYSPARK_PYTHON="${pysparkEnv.outPath}/bin/${pysparkEnv.executable}" + export PYSPARK_DRIVER_PYTHON="${pysparkEnv.outPath}/bin/${pysparkEnv.executable}" export PYTHONPATH="\$PYTHONPATH:$PYTHONPATH" export HADOOP_CONF_DIR="${hadoopConfDir}" export SPARKR_R_SHELL="${pkgs.R}/bin/R" @@ -45,11 +45,17 @@ let STOP cat > $out/spark-defaults.conf <<- STOP - spark.eventLog.enabled true - spark.eventLog.dir hdfs://localhost:/logs/spark - spark.history.fs.logDirectory hdfs://localhost:/logs/spark - # spark.yarn.keytab ${keytab_path} - # spark.yarn.principal spark/my.engine@MY.ENGINE + spark.eventLog.enabled true + spark.eventLog.dir hdfs://localhost:/logs/spark + spark.history.fs.logDirectory hdfs://localhost:/logs/spark + # spark.yarn.keytab ${keytab_path} + # spark.yarn.principal spark/my.engine@MY.ENGINE + spark.history.ui.acls.enable true + spark.history.kerberos.enabled true + spark.history.kerberos.keytab ${keytab_path} + spark.history.kerberos.principal spark/my.engine@MY.ENGINE + spark.yarn.appMasterEnv.PYSPARK_PYTHON ${pysparkEnv.outPath}/bin/${pysparkEnv.executable} + spark.yarn.appMasterEnv.PYTHONPATH ${pysparkEnv.outPath}/lib/${pysparkEnv.executable}/site-packages STOP ''; }; @@ -174,7 +180,7 @@ in namenode = { enable = true; formatOnInit = true; restartIfChanged = true; }; datanode = { enable = true; restartIfChanged = true; }; journalnode = { enable = true; restartIfChanged = true; }; - zkfc = { enable = true; restartIfChanged = true; }; + zkfc = { enable = false; restartIfChanged = true; }; # ZOOKEEPER DISABLED, not using High Availability setup httpfs = { enable = true; restartIfChanged = true; }; }; yarn = {