使用linux脚本shell检查大数据各节点服务是否运行正常

Posted 2020-12-11 cool小伙

脚本通过ssh检查大数据环境中各节点的相关实例是否运行正常，包括：zk、hadoop-ha、hdfs、yarn、hbase、hive-metastore、hiveserver2、kafka。

注意修改成自己的集群节点名称及kafka版本

#!/bin/bash
# 定义机器名前缀
HN_PREFIX="hadoop.test"
KFK_NAME="kafka_2.12-1.1.0"
 
# 各服务对应节点
# zk 节点
ZK_NODES="$HN_PREFIX.n1 $HN_PREFIX.n2 $HN_PREFIX.n3" 
# hadoop HA DFSZKFailoverController 节点
DFC_NODES="$HN_PREFIX.m1 $HN_PREFIX.m2" 
# hadoop HA journalnode 节点
JOURNAL_NODES="$HN_PREFIX.n1 $HN_PREFIX.n2 $HN_PREFIX.n3 $HN_PREFIX.n4 $HN_PREFIX.n5" 
# DFS namenode 节点
NAME_NODES="$HN_PREFIX.m1 $HN_PREFIX.m2" 
# DFS datanode 节点
DATA_NODES="$HN_PREFIX.n1 $HN_PREFIX.n2 $HN_PREFIX.n3 $HN_PREFIX.n4 $HN_PREFIX.n5" 
# YARN resourcemanager 节点
RM_NODES="$HN_PREFIX.m1 $HN_PREFIX.m2" 
# YARN nodemanager 节点
NM_NODES="$HN_PREFIX.n1 $HN_PREFIX.n2 $HN_PREFIX.n3 $HN_PREFIX.n4 $HN_PREFIX.n5" 
# hadoop jobhistoryserver 节点
JOBHIS_NODES="$HN_PREFIX.m1" 
# hbase hmaster 节点
HMASTER_NODES="$HN_PREFIX.m1 $HN_PREFIX.m2" 
# hbase hregionserver 节点
HREGION_NODES="$HN_PREFIX.n1 $HN_PREFIX.n2 $HN_PREFIX.n3 $HN_PREFIX.n4 $HN_PREFIX.n5" 
# hive metastore 节点
META_NODES="$HN_PREFIX.n1" 
# hive hiveserver2 节点
HIVE2_NODES="$HN_PREFIX.n1" 
# kafka 节点
KAFKA_NODES="$HN_PREFIX.n1 $HN_PREFIX.n2 $HN_PREFIX.n3" 
 
# 检测 zookeeper
echo "-------------------------------------------------------"
echo "zk - QuorumPeerMain ："
for ip in $ZK_NODES
do
    ssh hadoop@$ip "ps -fe|grep QuorumPeerMain |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hadoop HA DFSZKFailoverController
echo "-------------------------------------------------------"
echo "hadoop HA - DFSZKFailoverController ："
for ip in $DFC_NODES
do
    ssh hadoop@$ip "ps -fe|grep DFSZKFailoverController |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hadoop HA JournalNode
echo "-------------------------------------------------------"
echo "hadoop HA - JournalNode ："
for ip in $JOURNAL_NODES
do
    ssh hadoop@$ip "ps -fe|grep JournalNode |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hadoop HDFS NameNode
echo "-------------------------------------------------------"
echo "hadoop HDFS - NameNode ："
for ip in $NAME_NODES
do
    ssh hadoop@$ip "ps -fe|grep NameNode |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hadoop HDFS DataNode
echo "-------------------------------------------------------"
echo "hadoop HDFS - DataNode ："
for ip in $DATA_NODES
do
    ssh hadoop@$ip "ps -fe|grep DataNode |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hadoop YARN ResourceManager
echo "-------------------------------------------------------"
echo "hadoop YARN - ResourceManager ："
for ip in $RM_NODES
do
    ssh hadoop@$ip "ps -fe|grep ResourceManager |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hadoop YARN NodeManager
echo "-------------------------------------------------------"
echo "hadoop YARN - NodeManager ："
for ip in $NM_NODES
do
    ssh hadoop@$ip "ps -fe|grep NodeManager |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hadoop JobHistoryServer
echo "-------------------------------------------------------"
echo "hadoop - JobHistoryServer ："
for ip in $JOBHIS_NODES
do
    ssh hadoop@$ip "ps -fe|grep JobHistoryServer |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hbase HMaster
echo "-------------------------------------------------------"
echo "hbase - HMaster ："
for ip in $HMASTER_NODES
do
    ssh hadoop@$ip "ps -fe|grep HMaster |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hbase HRegionServer
echo "-------------------------------------------------------"
echo "hbase - HRegionServer ："
for ip in $HREGION_NODES
do
    ssh hadoop@$ip "ps -fe|grep HRegionServer |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hive metastore
echo "-------------------------------------------------------"
echo "hive - HiveMetaStore ："
for ip in $META_NODES
do
    ssh hadoop@$ip "ps -fe|grep HiveMetaStore |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 hive hiveserver2
echo "-------------------------------------------------------"
echo "hive - HiveServer2 ："
for ip in $HIVE2_NODES
do
    ssh hadoop@$ip "ps -fe|grep HiveServer2 |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done
 
# 检测 Kafka
echo "-------------------------------------------------------"
echo "Kafka - $KFK_NAME ："
for ip in $KAFKA_NODES
do
    ssh hadoop@$ip "ps -fe|grep $KFK_NAME |grep -v grep >/dev/null 2>&1"
    if [ $? -ne 0 ]
    then
        echo -e "$ip： [\\033[31m未启动\\033[0m]"
    else
        echo -e "$ip： [\\033[32m正常\\033[0m]"
    fi
done

运行效果：