在这里，以两台es集群为例。
es集群健康状况有三种状态，这里我们搭建的es集群，只要两台不同时挂掉，数据不会丢失。
| green | 所有主要分片和复制分片都可用 |
| yellow | 所有主要分片可用，但不是所有复制分片都可用 |
| red | 不是所有的主要分片都可用 |
举个例子：
- 比如说现在集群节点es1位主节点，es2位复制分片节点，默认情况下，两台es都接收logstash传过来的日志，是负载均衡的。
- 如果es1宕掉，es2会被提升为主节点，只有es2接收logstash传来得日志数据，同时整个集群状态由green转为red。
- 修复es1，es1会作为分片节点加入集群中，es1加入到集群中后，会进行修复，es2中新增的数据同步到es1分片节点，同步过程整个集群健康状态转为yellow。
- 当同步完成，集群状态转为green。
安装elasticsearch
官网：https://www.elastic.co/products/elasticsearch
# tar xf elasticsearch-2.3.5.tar.gz -C /usr/local/app/ # ln -sv /usr/local/app/elasticsearch-2.3.5 /usr/local/elasticsearch # chown –R nobody.nobody /usr/local/app/elasticsearch-2.3.5/ # cd /usr/local/elasticsearch
编辑配置文件
es1
cluster.name: myES
node.name: es1
node.master: true
path.data: /data/elasticsearch/data
path.logs: /data/elasticsearch/log
bootstrap.mlockall: True
network.host: 10.80.2.83
http.port: 9200
http.enabled: true
transport.tcp.port: 9300
transport.tcp.compress: true
discovery.zen.ping.unicast.hosts: ["10.80.2.83:9300","10.80.2.84:9300"]
discovery.zen.minimum_master_nodes: 1
discovery.zen.ping.timeout: 3s
discovery.zen.ping.multicast.enabled: false
gateway.recover_after_nodes: 1
gateway.recover_after_time: 5m
gateway.expected_nodes: 2
cluster.routing.allocation.node_initial_primaries_recoveries: 4
indices.recovery.max_size_per_sec: 50mb
index :
analysis :
analyzer :
default :
tokenizer : keyword
config/elasticsearch.yml
#!/bin/sh
# check in case a user was using this mechanism
if [ "x$ES_CLASSPATH" != "x" ]; then
cat >&2 << EOF
Error: Don‘t modify the classpath with ES_CLASSPATH. Best is to add
additional elements via the plugin mechanism, or if code must really be
added to the main classpath, add jars to lib/ (unsupported).
EOF
exit 1
fi
ES_CLASSPATH="$ES_HOME/lib/elasticsearch-2.3.4.jar:$ES_HOME/lib/*"
if [ "x$ES_MIN_MEM" = "x" ]; then
ES_MIN_MEM=8g
fi
if [ "x$ES_MAX_MEM" = "x" ]; then
ES_MAX_MEM=8g
fi
if [ "x$ES_HEAP_SIZE" != "x" ]; then
ES_MIN_MEM=$ES_HEAP_SIZE
ES_MAX_MEM=$ES_HEAP_SIZE
fi
# min and max heap sizes should be set to the same value to avoid
# stop-the-world GC pauses during resize, and so that we can lock the
# heap in memory on startup to prevent any of it from being swapped
# out.
JAVA_OPTS="$JAVA_OPTS -Xms${ES_MIN_MEM}"
JAVA_OPTS="$JAVA_OPTS -Xmx${ES_MAX_MEM}"
# new generation
if [ "x$ES_HEAP_NEWSIZE" != "x" ]; then
JAVA_OPTS="$JAVA_OPTS -Xmn${ES_HEAP_NEWSIZE}"
fi
# max direct memory
if [ "x$ES_DIRECT_SIZE" != "x" ]; then
JAVA_OPTS="$JAVA_OPTS -XX:MaxDirectMemorySize=${ES_DIRECT_SIZE}"
fi
# set to headless, just in case
JAVA_OPTS="$JAVA_OPTS -Djava.awt.headless=true"
# Force the JVM to use IPv4 stack
if [ "x$ES_USE_IPV4" != "x" ]; then
JAVA_OPTS="$JAVA_OPTS -Djava.net.preferIPv4Stack=true"
fi
# Add gc options. ES_GC_OPTS is unsupported, for internal testing
if [ "x$ES_GC_OPTS" = "x" ]; then
ES_GC_OPTS="$ES_GC_OPTS -XX:+UseParNewGC"
ES_GC_OPTS="$ES_GC_OPTS -XX:+UseConcMarkSweepGC"
ES_GC_OPTS="$ES_GC_OPTS -XX:CMSInitiatingOccupancyFraction=75"
ES_GC_OPTS="$ES_GC_OPTS -XX:+UseCMSInitiatingOccupancyOnly"
fi
JAVA_OPTS="$JAVA_OPTS $ES_GC_OPTS"
# GC logging options
if [ -n "$ES_GC_LOG_FILE" ]; then
JAVA_OPTS="$JAVA_OPTS -XX:+PrintGCDetails"
JAVA_OPTS="$JAVA_OPTS -XX:+PrintGCTimeStamps"
JAVA_OPTS="$JAVA_OPTS -XX:+PrintGCDateStamps"
JAVA_OPTS="$JAVA_OPTS -XX:+PrintClassHistogram"
JAVA_OPTS="$JAVA_OPTS -XX:+PrintTenuringDistribution"
JAVA_OPTS="$JAVA_OPTS -XX:+PrintGCApplicationStoppedTime"
JAVA_OPTS="$JAVA_OPTS -Xloggc:$ES_GC_LOG_FILE"
# Ensure that the directory for the log file exists: the JVM will not create it.
mkdir -p "`dirname \"$ES_GC_LOG_FILE\"`"
fi
# Causes the JVM to dump its heap on OutOfMemory.
JAVA_OPTS="$JAVA_OPTS -XX:+HeapDumpOnOutOfMemoryError"
# The path to the heap dump location, note directory must exists and have enough
# space for a full heap dump.
#JAVA_OPTS="$JAVA_OPTS -XX:HeapDumpPath=$ES_HOME/logs/heapdump.hprof"
# Disables explicit GC
JAVA_OPTS="$JAVA_OPTS -XX:+DisableExplicitGC"
# Ensure UTF-8 encoding by default (e.g. filenames)
JAVA_OPTS="$JAVA_OPTS -Dfile.encoding=UTF-8"
# Use our provided JNA always versus the system one
JAVA_OPTS="$JAVA_OPTS -Djna.nosys=true"
bin/elasticsearch.in.sh
es2
#集群名字
cluster.name: myES
#节点名字
node.name: es2
#是否有资格成为主节点
node.master: true
#数据目录
path.data: /data/elasticsearch/data
#日志目录
path.logs: /data/elasticsearch/log
#锁住内存
bootstrap.mlockall: True
#绑定的IP地址，可为0.0.0.0
#network.bind_host: 10.80.2.84
#该节点与其他节点交互的ip地址，
#network.publish_host: 10.80.2.84
#上面两个参数集合
network.host: 10.80.2.84
#对外服务端口
http.port: 9200
#是否使用http协议对外提供服务
http.enabled: true
#节点间交互tcp端口
transport.tcp.port: 9300
#传输数据时压缩
transport.tcp.compress: true
#集群中主节点初始化列表，通过这些节点自动发现加入其他节点
discovery.zen.ping.unicast.hosts: ["10.80.2.83:9300","10.80.2.84:9300"]
#保证集群中节点可以知道其他N各有主节点资格，默认1，es数超过2个，可设置大一些
discovery.zen.minimum_master_nodes: 1
#自动发现其他节点时ping超时时间，网络环境差，提高此参数
discovery.zen.ping.timeout: 3s
#是否打开多播发现节点，默认true
discovery.zen.ping.multicast.enabled: false
#集群中几个节点启动时进行数据恢复
gateway.recover_after_nodes: 1
#初始化数据恢复进程超时时间
gateway.recover_after_time: 5m
#这个es集群中节点数量，一旦这些数目节点启动，进行数据恢复
gateway.expected_nodes: 2
#初始化数据恢复时，并发恢复的线程个数
cluster.routing.allocation.node_initial_primaries_recoveries: 4
#数据恢复时最大带宽
indices.recovery.max_size_per_sec: 50mb
# 禁止分词，
index :
analysis :
analyzer :
default :
tokenizer : keyword
config/elasticsearch.yml
#!/bin/sh
# check in case a user was using this mechanism
if [ "x$ES_CLASSPATH" != "x" ]; then
cat >&2 << EOF
Error: Don‘t modify the classpath with ES_CLASSPATH. Best is to add
additional elements via the plugin mechanism, or if code must really be
added to the main classpath, add jars to lib/ (unsupported).
EOF
exit 1
fi
ES_CLASSPATH="$ES_HOME/lib/elasticsearch-2.3.4.jar:$ES_HOME/lib/*"
if [ "x$ES_MIN_MEM" = "x" ]; then
ES_MIN_MEM=8g
fi
if [ "x$ES_MAX_MEM" = "x" ]; then
ES_MAX_MEM=8g
fi
if [ "x$ES_HEAP_SIZE" != "x" ]; then
ES_MIN_MEM=$ES_HEAP_SIZE
ES_MAX_MEM=$ES_HEAP_SIZE
fi
# min and max heap sizes should be set to the same value to avoid
# stop-the-world GC pauses during resize, and so that we can lock the
# heap in memory on startup to prevent any of it from being swapped
# out.
JAVA_OPTS="$JAVA_OPTS -Xms${ES_MIN_MEM}"
JAVA_OPTS="$JAVA_OPTS -Xmx${ES_MAX_MEM}"
# new generation
if [ "x$ES_HEAP_NEWSIZE" != "x" ]; then
JAVA_OPTS="$JAVA_OPTS -Xmn${ES_HEAP_NEWSIZE}"
fi
# max direct memory
if [ "x$ES_DIRECT_SIZE" != "x" ]; then
JAVA_OPTS="$JAVA_OPTS -XX:MaxDirectMemorySize=${ES_DIRECT_SIZE}"
fi
# set to headless, just in case
JAVA_OPTS="$JAVA_OPTS -Djava.awt.headless=true"
# Force the JVM to use IPv4 stack
if [ "x$ES_USE_IPV4" != "x" ]; then
JAVA_OPTS="$JAVA_OPTS -Djava.net.preferIPv4Stack=true"
fi
# Add gc options. ES_GC_OPTS is unsupported, for internal testing
if [ "x$ES_GC_OPTS" = "x" ]; then
ES_GC_OPTS="$ES_GC_OPTS -XX:+UseParNewGC"
ES_GC_OPTS="$ES_GC_OPTS -XX:+UseConcMarkSweepGC"
ES_GC_OPTS="$ES_GC_OPTS -XX:CMSInitiatingOccupancyFraction=75"
ES_GC_OPTS="$ES_GC_OPTS -XX:+UseCMSInitiatingOccupancyOnly"
fi
JAVA_OPTS="$JAVA_OPTS $ES_GC_OPTS"
# GC logging options
if [ -n "$ES_GC_LOG_FILE" ]; then
JAVA_OPTS="$JAVA_OPTS -XX:+PrintGCDetails"
JAVA_OPTS="$JAVA_OPTS -XX:+PrintGCTimeStamps"
JAVA_OPTS="$JAVA_OPTS -XX:+PrintGCDateStamps"
JAVA_OPTS="$JAVA_OPTS -XX:+PrintClassHistogram"
JAVA_OPTS="$JAVA_OPTS -XX:+PrintTenuringDistribution"
JAVA_OPTS="$JAVA_OPTS -XX:+PrintGCApplicationStoppedTime"
JAVA_OPTS="$JAVA_OPTS -Xloggc:$ES_GC_LOG_FILE"
# Ensure that the directory for the log file exists: the JVM will not create it.
mkdir -p "`dirname \"$ES_GC_LOG_FILE\"`"
fi
# Causes the JVM to dump its heap on OutOfMemory.
JAVA_OPTS="$JAVA_OPTS -XX:+HeapDumpOnOutOfMemoryError"
# The path to the heap dump location, note directory must exists and have enough
# space for a full heap dump.
#JAVA_OPTS="$JAVA_OPTS -XX:HeapDumpPath=$ES_HOME/logs/heapdump.hprof"
# Disables explicit GC
JAVA_OPTS="$JAVA_OPTS -XX:+DisableExplicitGC"
# Ensure UTF-8 encoding by default (e.g. filenames)
JAVA_OPTS="$JAVA_OPTS -Dfile.encoding=UTF-8"
# Use our provided JNA always versus the system one
JAVA_OPTS="$JAVA_OPTS -Djna.nosys=true"
bin/elasticsearch.in.sh
时间： 2024-08-11 03:28:11