bin/interpreter.sh

#!/bin/bash
#
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements.  See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You under the Apache License, Version 2.0
# (the "License"); you may not use this file except in compliance with
# the License.  You may obtain a copy of the License at
#
#    http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#


bin="$(dirname "${BASH_SOURCE-$0}")"
bin="$(cd "${bin}">/dev/null; pwd)"

function usage() {
    echo "usage) $0 -p <port> -r <intp_port> -d <interpreter dir to load> -l <local interpreter repo dir to load> -g <interpreter group name>"
}

function downloadInterpreterLibraries() {
    mkdir -p ${LOCAL_INTERPRETER_REPO}
    IFS=' ' read -r -a JAVA_INTP_OPTS_ARRAY <<< "${JAVA_INTP_OPTS}"
    ZEPPELIN_DOWNLOADER="org.apache.zeppelin.interpreter.remote.RemoteInterpreterDownloader"
    INTERPRETER_DOWNLOAD_COMMAND+=("${ZEPPELIN_RUNNER}" "${JAVA_INTP_OPTS_ARRAY[@]}" "-cp" "${ZEPPELIN_INTP_CLASSPATH_OVERRIDES}:${ZEPPELIN_INTP_CLASSPATH}" "${ZEPPELIN_DOWNLOADER}" "${CALLBACK_HOST}" "${PORT}" "${INTERPRETER_SETTING_NAME}" "${LOCAL_INTERPRETER_REPO}")
    echo "Interpreter download command: ${INTERPRETER_DOWNLOAD_COMMAND[@]}"
    eval "${INTERPRETER_DOWNLOAD_COMMAND[@]}"
}

# pre-requisites for checking that we're running in container
if [ -f /proc/self/cgroup ] && [ -n "$(command -v getent)" ]; then
    # checks if we're running in container...
    if awk -F: '/cpu/ && $3 ~ /^\/$/{ c=1 } END { exit c }' /proc/self/cgroup; then
        # Check whether there is a passwd entry for the container UID
        myuid="$(id -u)"
        mygid="$(id -g)"
        # turn off -e for getent because it will return error code in anonymous uid case
        set +e
        uidentry="$(getent passwd "$myuid")"
        set -e

        # If there is no passwd entry for the container UID, attempt to create one
        if [ -z "$uidentry" ] ; then
            if [ -w /etc/passwd ] ; then
                echo "zeppelin:x:$myuid:$mygid:anonymous uid:$ZEPPELIN_HOME:/bin/false" >> /etc/passwd
            else
                echo "Container ENTRYPOINT failed to add passwd entry for anonymous UID"
            fi
        fi
    fi
fi

while getopts "hc:p:r:i:d:l:v:u:g:" o; do
    case ${o} in
        h)
            usage
            exit 0
            ;;
        d)
            INTERPRETER_DIR=${OPTARG}
            ;;
        c)
            CALLBACK_HOST=${OPTARG} # This will be used callback host
            ;;
        p)
            PORT=${OPTARG} # This will be used for callback port
            ;;
        r)
            INTP_PORT=${OPTARG} # This will be used for interpreter process port
            ;;
        i)
            INTP_GROUP_ID=${OPTARG} # This will be used for interpreter group id
            ;;
        l)
            LOCAL_INTERPRETER_REPO=${OPTARG}
            ;;
        v)
            . "${bin}/common.sh"
            getZeppelinVersion
            ;;
        u)
            ZEPPELIN_IMPERSONATE_USER="${OPTARG}"
            ;;
        g)
            INTERPRETER_SETTING_NAME=${OPTARG}
            ;;
        esac
done


if [ -z "${PORT}" ] || [ -z "${INTERPRETER_DIR}" ]; then
    usage
    exit 1
fi

. "${bin}/common.sh"

check_java_version

ZEPPELIN_INTERPRETER_API_JAR=$(find "${ZEPPELIN_HOME}/interpreter" -name 'zeppelin-interpreter-shaded-*.jar')
ZEPPELIN_INTP_CLASSPATH+=":${CLASSPATH}:${ZEPPELIN_INTERPRETER_API_JAR}"

# construct classpath
if [[ -d "${ZEPPELIN_HOME}/zeppelin-interpreter/target/classes" ]]; then
  ZEPPELIN_INTP_CLASSPATH+=":${ZEPPELIN_HOME}/zeppelin-interpreter/target/classes"
fi

# add test classes for unittest
if [[ -d "${ZEPPELIN_HOME}/zeppelin-zengine/target/test-classes" ]]; then
  ZEPPELIN_INTP_CLASSPATH+=":${ZEPPELIN_HOME}/zeppelin-zengine/target/test-classes"
  addJarInDirForIntp "${ZEPPELIN_HOME}/zeppelin-zengine/target/test-classes"
fi

addJarInDirForIntp "${ZEPPELIN_HOME}/zeppelin-interpreter-shaded/target"

HOSTNAME=$(hostname)
ZEPPELIN_SERVER=org.apache.zeppelin.interpreter.remote.RemoteInterpreterServer

INTERPRETER_ID=$(basename "${INTERPRETER_DIR}")
if [[ "${INTERPRETER_ID}" != "flink" ]]; then
  # don't add interpreter jar for flink, FlinkInterpreterLauncher will choose the right interpreter jar based
  # on scala version of current FLINK_HOME.
  addJarInDirForIntp "${INTERPRETER_DIR}"
fi

ZEPPELIN_PID="${ZEPPELIN_PID_DIR}/zeppelin-interpreter-${INTP_GROUP_ID}-${ZEPPELIN_IDENT_STRING}-${HOSTNAME}-${PORT}.pid"

if [[ "${ZEPPELIN_INTERPRETER_LAUNCHER}" == "yarn" ]]; then
    # {LOG_DIRS} is env name in yarn container which point to the log dirs of container
    # split the log dirs to array and use the first one
    IFS=','
    read -ra LOG_DIRS_ARRAY <<< "${LOG_DIRS}"
    ZEPPELIN_LOG_DIR=${LOG_DIRS_ARRAY[0]}
fi

ZEPPELIN_LOGFILE="${ZEPPELIN_LOG_DIR}/zeppelin-interpreter-${INTERPRETER_GROUP_ID}-"

if [[ -z "$ZEPPELIN_IMPERSONATE_CMD" ]]; then
    if [[ "${INTERPRETER_ID}" != "spark" || "$ZEPPELIN_IMPERSONATE_SPARK_PROXY_USER" == "false" ]]; then
        ZEPPELIN_IMPERSONATE_RUN_CMD=("ssh" "${ZEPPELIN_IMPERSONATE_USER}@localhost")
    fi
else
    ZEPPELIN_IMPERSONATE_RUN_CMD=$(eval "echo ${ZEPPELIN_IMPERSONATE_CMD} ")
fi


if [[ -n "$ZEPPELIN_IMPERSONATE_USER" ]]; then
    ZEPPELIN_LOGFILE+="${ZEPPELIN_IMPERSONATE_USER}-"
fi
ZEPPELIN_LOGFILE+="${ZEPPELIN_IDENT_STRING}-${HOSTNAME}.log"
JAVA_INTP_OPTS+=" -Dzeppelin.log.file=${ZEPPELIN_LOGFILE}"

if [[ ! -d "${ZEPPELIN_LOG_DIR}" ]]; then
  echo "Log dir doesn't exist, create ${ZEPPELIN_LOG_DIR}"
  mkdir -p "${ZEPPELIN_LOG_DIR}"
fi

# set spark related env variables
if [[ "${INTERPRETER_ID}" == "spark" ]]; then

  # run kinit
  if [[ -n "${ZEPPELIN_SERVER_KERBEROS_KEYTAB}" ]] && [[ -n "${ZEPPELIN_SERVER_KERBEROS_PRINCIPAL}" ]]; then
    kinit -kt "${ZEPPELIN_SERVER_KERBEROS_KEYTAB}" "${ZEPPELIN_SERVER_KERBEROS_PRINCIPAL}"
  fi
  if [[ -n "${SPARK_HOME}" ]]; then
    export SPARK_SUBMIT="${SPARK_HOME}/bin/spark-submit"
    SPARK_APP_JAR="$(ls "${ZEPPELIN_HOME}"/interpreter/spark/spark-interpreter*.jar)"
    # This will evantually passes SPARK_APP_JAR to classpath of SparkIMain
    ZEPPELIN_INTP_CLASSPATH+=":${SPARK_APP_JAR}"

    py4j=("${SPARK_HOME}"/python/lib/py4j-*-src.zip)
    # pick the first match py4j zip - there should only be one
    export PYTHONPATH="$SPARK_HOME/python/:$PYTHONPATH"
    export PYTHONPATH="${py4j[0]}:$PYTHONPATH"
  else
    echo "No SPARK_HOME is specified"
    exit -1
  fi

  if [[ -n "${HADOOP_CONF_DIR}" ]] && [[ -d "${HADOOP_CONF_DIR}" ]]; then
    ZEPPELIN_INTP_CLASSPATH+=":${HADOOP_CONF_DIR}"
    export HADOOP_CONF_DIR=${HADOOP_CONF_DIR}
  else
    # autodetect HADOOP_CONF_HOME by heuristic
    if [[ -n "${HADOOP_HOME}" ]] && [[ -z "${HADOOP_CONF_DIR}" ]]; then
      if [[ -d "${HADOOP_HOME}/etc/hadoop" ]]; then
        export HADOOP_CONF_DIR="${HADOOP_HOME}/etc/hadoop"
      elif [[ -d "/etc/hadoop/conf" ]]; then
        export HADOOP_CONF_DIR="/etc/hadoop/conf"
      fi
    fi
  fi

elif [[ "${INTERPRETER_ID}" == "hbase" ]]; then
  if [[ -n "${HBASE_CONF_DIR}" ]]; then
    ZEPPELIN_INTP_CLASSPATH+=":${HBASE_CONF_DIR}"
  elif [[ -n "${HBASE_HOME}" ]]; then
    ZEPPELIN_INTP_CLASSPATH+=":${HBASE_HOME}/conf"
  else
    echo "HBASE_HOME and HBASE_CONF_DIR are not set, configuration might not be loaded"
  fi

elif [[ "${INTERPRETER_ID}" == "flink" ]]; then
  addEachJarInDirRecursiveForIntp "${FLINK_HOME}/lib"

  FLINK_PYTHON_JAR=$(find "${FLINK_HOME}/opt" -name 'flink-python_*.jar')
  ZEPPELIN_INTP_CLASSPATH+=":${FLINK_PYTHON_JAR}:${FLINK_APP_JAR}"

  if [[ -n "${HADOOP_CONF_DIR}" ]] && [[ -d "${HADOOP_CONF_DIR}" ]]; then
    ZEPPELIN_INTP_CLASSPATH+=":${HADOOP_CONF_DIR}"
    # Don't use `hadoop classpath` if flink-hadoop-shaded in in lib folder
    flink_hadoop_shaded_jar=$(find "${FLINK_HOME}/lib" -name 'flink-shaded-hadoop-*.jar')
    if [[ -n "$flink_hadoop_shaded_jar" ]]; then
      echo ""
    else
      if [[ ! ( -x "$(command -v hadoop)" ) && ( "${ZEPPELIN_INTERPRETER_LAUNCHER}" != "yarn" ) ]]; then
        echo 'Error: hadoop is not in PATH when HADOOP_CONF_DIR is specified and no flink-shaded-hadoop jar '
        exit 1
      fi
      ZEPPELIN_INTP_CLASSPATH+=":$(hadoop classpath)"
    fi
    export HADOOP_CONF_DIR=${HADOOP_CONF_DIR}
  else
    # autodetect HADOOP_CONF_HOME by heuristic
    if [[ -n "${HADOOP_HOME}" ]] && [[ -z "${HADOOP_CONF_DIR}" ]]; then
      if [[ -d "${HADOOP_HOME}/etc/hadoop" ]]; then
        export HADOOP_CONF_DIR="${HADOOP_HOME}/etc/hadoop"
      elif [[ -d "/etc/hadoop/conf" ]]; then
        export HADOOP_CONF_DIR="/etc/hadoop/conf"
      fi
    fi
  fi

fi

downloadInterpreterLibraries
addJarInDirForIntp "${LOCAL_INTERPRETER_REPO}"

if [[ -n "$ZEPPELIN_IMPERSONATE_USER" ]]; then
  if [[ "${INTERPRETER_ID}" != "spark" || "$ZEPPELIN_IMPERSONATE_SPARK_PROXY_USER" == "false" ]]; then
    suid="$(id -u "${ZEPPELIN_IMPERSONATE_USER}")"
    if [[ -n  "${suid}" || -z "${SPARK_SUBMIT}" ]]; then
       INTERPRETER_RUN_COMMAND+=("${ZEPPELIN_IMPERSONATE_RUN_CMD[@]}")
       if [[ -f "${ZEPPELIN_CONF_DIR}/zeppelin-env.sh" ]]; then
           INTERPRETER_RUN_COMMAND+=("source" "${ZEPPELIN_CONF_DIR}/zeppelin-env.sh;")
       fi
    fi
  fi
fi

if [[ -n "${SPARK_SUBMIT}" ]]; then
  IFS=' ' read -r -a SPARK_SUBMIT_OPTIONS_ARRAY <<< "${SPARK_SUBMIT_OPTIONS}"
  IFS='|' read -r -a ZEPPELIN_SPARK_CONF_ARRAY <<< "${ZEPPELIN_SPARK_CONF}"
  if [[ "${ZEPPELIN_SPARK_YARN_CLUSTER}" == "true"  ]]; then
    INTERPRETER_RUN_COMMAND+=("${SPARK_SUBMIT}" "--class" "${ZEPPELIN_SERVER}" "--driver-java-options" "${SPARK_DRIVER_EXTRAJAVAOPTIONS_CONF} ${JAVA_INTP_OPTS}" "${SPARK_SUBMIT_OPTIONS_ARRAY[@]}" "${ZEPPELIN_SPARK_CONF_ARRAY[@]}" "${SPARK_APP_JAR}" "${CALLBACK_HOST}" "${PORT}" "${INTP_GROUP_ID}" "${INTP_PORT}")
  else
    INTERPRETER_RUN_COMMAND+=("${SPARK_SUBMIT}" "--class" "${ZEPPELIN_SERVER}" "--driver-class-path" "${ZEPPELIN_INTP_CLASSPATH_OVERRIDES}:${ZEPPELIN_INTP_CLASSPATH}" "--driver-java-options" "${SPARK_DRIVER_EXTRAJAVAOPTIONS_CONF} ${JAVA_INTP_OPTS}" "${SPARK_SUBMIT_OPTIONS_ARRAY[@]}" "${ZEPPELIN_SPARK_CONF_ARRAY[@]}" "${SPARK_APP_JAR}" "${CALLBACK_HOST}" "${PORT}" "${INTP_GROUP_ID}" "${INTP_PORT}")
  fi
elif [[ -n "${ZEPPELIN_FLINK_APPLICATION_MODE}" ]]; then
  IFS='|' read -r -a ZEPPELIN_FLINK_APPLICATION_MODE_CONF_ARRAY <<< "${ZEPPELIN_FLINK_APPLICATION_MODE_CONF}"
  INTERPRETER_RUN_COMMAND+=("${FLINK_HOME}/bin/flink" "run-application" "-c" "${ZEPPELIN_SERVER}" "-t" "${ZEPPELIN_FLINK_APPLICATION_MODE}" "${ZEPPELIN_FLINK_APPLICATION_MODE_CONF_ARRAY[@]}" "${FLINK_APP_JAR}" "${CALLBACK_HOST}" "${PORT}" "${INTP_GROUP_ID}" "${INTP_PORT}")
else
  IFS=' ' read -r -a JAVA_INTP_OPTS_ARRAY <<< "${JAVA_INTP_OPTS}"
  IFS=' ' read -r -a ZEPPELIN_INTP_MEM_ARRAY <<< "${ZEPPELIN_INTP_MEM}"
  INTERPRETER_RUN_COMMAND+=("${ZEPPELIN_RUNNER}" "${JAVA_INTP_OPTS_ARRAY[@]}" "${ZEPPELIN_INTP_MEM_ARRAY[@]}" "-cp" "${ZEPPELIN_INTP_CLASSPATH_OVERRIDES}:${ZEPPELIN_INTP_CLASSPATH}" "${ZEPPELIN_SERVER}" "${CALLBACK_HOST}" "${PORT}" "${INTP_GROUP_ID}" "${INTP_PORT}")
fi

# Don't remove this echo, it is for diagnose, this line of output will be redirected to java log4j output.
# Output that starts with `[INFO]` will be redirected to log4j INFO output. Other outputs from interpreter.sh
# will be redirected to log4j DEBUG output.
echo "[INFO] Interpreter launch command: ${INTERPRETER_RUN_COMMAND[@]}"
exec "${INTERPRETER_RUN_COMMAND[@]}"