zoukankan      html  css  js  c++  java
  • elasticsearch5.5.2安装

     elasticsearch5.x安装中一些问题的解决办法
    • 最近在学习elk,由于编译安装使用5.2.1版本的elasticsearch,所以遇到了很多问题,下面是一些问题及解决办法。

    1、修改访问elasticsearch的IP及端口

    [seven@localhost config]$ vim /usr/java/elasticsearch/config/elasticsearch.yml

    找到如下代码段,并取消network.host及http.port所在行的注释,修改IP及端口

    # ---------------------------------- Network -----------------------------------
    #
    # Set the bind address to a specific IP (IPv4 or IPv6):
    #
    network.host: 192.168.0.155
    #
    # Set a custom port for HTTP:
    #
    http.port: 9200
    #
    # For more information, see the documentation at:
    # <http://www.elastic.co/guide/en/elasticsearch/reference/current/modules-network.html>


    2、max file descriptors [4096] for elasticsearch process likely too low, increase to at least [65536]
    [2016-06-28 14:55:49,770][INFO ][node                     ] [Goldbug] stopping ...
    [2016-06-28 14:55:49,875][INFO ][node                     ] [Goldbug] stopped
    [2016-06-28 14:55:49,875][INFO ][node                     ] [Goldbug] closing ...
    [2016-06-28 14:55:49,887][INFO ][node                     ] [Goldbug] closed

    这个问题折腾了我一下午,最后还是找到了解决方案,同样回到config/elasticsearch.yml文件,找到如下配置,开放discovery.zen.ping.unicast.hosts及discovery.zen.minimum_master_nodes

    # --------------------------------- Discovery ----------------------------------
    #
    # Pass an initial list of hosts to perform discovery when new node is started:
    # The default list of hosts is ["127.0.0.1", "[::1]"]
    #
    discovery.zen.ping.unicast.hosts: ["192.168.0.155"]
    #
    # Prevent the "split brain" by configuring the majority of nodes (total number of nodes / 2 + 1):
    #
    discovery.zen.minimum_master_nodes: 3
    #
    # For more information, see the documentation at:
    # <http://www.elastic.co/guide/en/elasticsearch/reference/current/modules-discovery.html>

    然后修改max file descriptors [4096] for elasticsearch process likely too low, increase to at least [65536]这个错误(切换到root操作)



    3.root超级用户不能正常启动

    由于elasticsearch2.0版本以后不能使用root来启动,所以需要创建一个普通用户来启动。
    
    [root@bogon ~]# groupadd elasticsearch
    [root@bogon ~]# useradd elasticsearch -g elasticsearch
    [root@bogon ~]# chown -R elasticsearch.elasticsearch /usr/local/elk/elasticsearch-5.2.1  
    [root@bogon ~]# su - elasticsearch
    [elasticsearch@bogon ~]$ cd /usr/local/elk/elasticsearch-5.2.1/bin/
    [elasticsearch@bogon bin]$ ./elasticsearch

    由普通用户来启动,则可以正常启动服务。

    4.ERROR: bootstrap checks failed

    max file descriptors [4096] for elasticsearch process likely too low, increase to at least [65536]
    max number of threads [1024] for user [lishang] likely too low, increase to at least [2048]
    
    解决:切换到root用户,编辑limits.conf 添加类似如下内容
    
    vi /etc/security/limits.conf 
    
    添加如下内容:
    
    * soft nofile 65536
    
    * hard nofile 131072
    
    * soft nproc 2048
    
    * hard nproc 4096

    使用如下命令查看是否修改成功

    [seven@localhost ~]$ ulimit -Hn
    65536

    5.max number of threads [1024] for user [lish] likely too low, increase to at least [2048]

    解决:切换到root用户,进入limits.d目录下修改配置文件。
    
    vi /etc/security/limits.d/90-nproc.conf 
    
    修改如下内容:
    
    * soft nproc 1024
    
    #修改为
    
    * soft nproc 2048
    

    6.max virtual memory areas vm.max_map_count [65530] likely too low, increase to at least [262144]

    解决:切换到root用户修改配置sysctl.conf
    
    vi /etc/sysctl.conf 
    
    添加下面配置:
    
    vm.max_map_count=655360
    
    并执行命令:
    
    sysctl -p
    
    然后,重新启动elasticsearch,即可启动成功。
    
     
     
     

    Ubuntu 14.04中Elasticsearch集群配置

    前言:本文可用于elasticsearch集群搭建参考。细分为elasticsearch.yml配置和系统配置

    达到的目的:各台机器配置成功,可以达到如下效果:

    • elasticsearch节点组成集群,对外来看俨如一台机器
    • elasticsearch节点的内存得到优化调整

    一. elasticsearch.yml配置

    network.host: "172.16.0.4"
    #这里可以是内网ip、机器名
    bootstrap.mlockall: true
    discovery.zen.ping.multicast.enabled: false
    #禁止掉多播
    discovery.zen.minimum_master_nodes: 2
    #(number of master-eligible nodes / 2) + 1 即 :(有master潜力的节点/2 )+1
    discovery.zen.ping.unicast.hosts: ["172.16.0.4", "172.16.0.6"]
    #这里配置的IP为master潜力节点的内网IP地址
    path:
    data: ["G:/elasticsearch/Data","X:/"]
    #es擎住的数据的存放地址。linux虚拟机中需要挂载磁盘作为存放目录。
    
    #cluster.routing.allocation.enable : "all"
    #indices.fielddata.cache.size : "75%"
    indices.recovery.max_bytes_per_sec: 100mb
    #这个配不之配置不影响集群运行,可以不写
    
    node.master: true
    #设置为true,表示为master潜力节点
    node.data: false
    #表示为非数据节点
    node.name: ${COMPUTERNAME}
    #设置节点名称

    二. 系统文件配置

    es_heap_size=$(free -m |grep Mem | awk '{if ($2/2 >31744) print 31744;else print $2/2;}')
    # 设置es的heap_size。一般取内存的一半,但是不大于32G
    sudo printf "
    ES_HEAP_SIZE=%sm
    " $es_heap_size >> /etc/default/elasticsearch
    sudo printf "MAX_LOCKED_MEMORY=unlimited
    " >> /etc/default/elasticsearch
    # 如果在elasticsearch.yml设置了bootstrap.mlockall: true,此处应该设置为unlimited。并且需要设置上述的es_heap_size
    sudo echo "elasticsearch - nofile 65536" >> /etc/security/limits.conf
    #设置elasticsearch的最大文件打开数位65536
    sudo echo "elasticsearch - memlock unlimited" >> /etc/security/limits.conf
    
    #sudo service elasticsearch restart
    #sudo update-rc.d elasticsearch defaults 95 10
    #查看是否设置成功

    检查内存是否配置成功

    http://yourip:9200/_cat/nodes?v&h=host,heap.current,heap.percent,heap.max,ram.max,disk.avail,node.role,m

    大概得到如下

    NOTE:上述采用的是sysvinit方式。如果你采用的是system方式,系统文件配置有部分需要修改。具体修改我有地方不理解,所以不做介绍

    具体查看:

    Elsticsearch running as service on Linux

    关于sysvinit和system的介绍可以查看:

    sysvinit and system

     
     
     
     

    ElasticSearch的工作机制

    ElasticSearch,和Solr一样,是底层基于Apache Lucene,且具备高可靠性的企业级搜索引擎。

    ElasticSearch中的一些概念其实和关系型数据库都有对应关系,比如数据库在ES中被称为索引,表在ES中被称作Type。

    具体对应关系见下表。

    ElasticSearch的工作机制

    ElasticSearch中的Replica是副本的意思,创建副本的好处有两个,1,可以分流部分查询请求,2,如果集群中的某个分片丢失了,就可以使用这个副本将数据全部找回来,因为这个原因,副本分片和源分片不会放在同一节点上。 ES中每一个索引都可以被分成多个分片,但不一定每个分片都有副本,但是一旦创建了副本,就会有主分片的说法(作为复制源的分片),分片和副本的数量可以在索引创建的时候指定。下图是副本和分片的示意图,分片和它的副本不会在同一个节点上。

    ElasticSearch的工作机制

    在索引创建之后,你可以在任何时候动态地改变副本的数量,但是你事后不能改变分片的数量。  默认情况下,Elasticsearch中的每个索引被分片5个主分片和1套副本,这意味着,如果你的集群中至少有两个节点,你的索引将会有5个主分片和另外5个副本,这样的话每个索引总共就有10个分片。

    当ES的一个节点启动后,它会通过广播方式找到集群中的其他节点,并且建立连接。

    在集群中,其中的某个节点会被选取作为主节点,这个主节点负责管理集群状态。这个主节点对于用户来说是透明的,用户不需要知道哪个节点是主节点。任何操作都可以发送到任何节点。必要的时候,任何节点可以并行的发送子查询到其他节点,并且将得到的响应合并后发送给用户,这些操作都不需要访问主节点。

    主节点读取集群信息,在读取过程中,它会检测分片的情况,哪些分片是主分片,并且是可用的,在这一步之后,所有的分片已经准备好了,而副本还没有。下一步的操作就是找到那些已经被复制过的分片,将他们作为副本。如果一切顺利,那么ES启动成功了,所有的分片和副本都已经准备好了。

    在ES工作的时候,主节点会监控所有的节点是否正常,默认配置为:节点每隔1s主节点会发送1次心跳,超时时间为30s,测试次数为3次,超过3次,则认为该节点同主节点已经脱离了。如果某一个节点出现问题,ES认为这个节点损坏,该节点会从集群中删除,并且ES会重新平衡整个集群。

    ES通过Query DSL (基于json的查询语言)来查询数据,在ES内部,每次查询分成2个步骤,分散和聚合,分散是指查询所有相关的分片,聚合是指把所有分片上的查询结果合并,排序,处理然后在返回给客户端。

    ElasticSearch 有4中方式来构建数据库,最简单的方法是使用index API,将一个Document发送到特定的index,一般通过curl tools实现。第二第三种方法是通过bulk API和UDP bulk API。两者的区别仅在于连接方式。第四种方式是通过一个插件-river。river运行在ElasticSearch上,并且可以从外部数据库导入数据到ES中。需要注意的是,数据构建仅在分片上进行,而不能在副本上进行。

     
     
     
     
     
     

    配置文件

    *以下列表中的参数可支持自动化配置,其余未列出来皆用默认配置(如有不妥,请及时纠偏,尤其是 配置节点类型一列)

    配置参数功能简介配置节点类型自动化配置建议配置所属模块
    cluster.name
    集群名称
    • data
    • master
    cluster
    node.name 节点名称
    • data
    • master

    node

    node.master

    是否是master
    • data
    • master

    node.data

    是否是data
    • data
    • master

    index.number_of_shards

    索引分片数
    • data
    • master







    index

    index.number_of_replicas

    索引备份数
    • data
    • master

    index.refresh_interval

    refresh时间
    • data
    • master

    index.merge.scheduler.max_thread_count

    merge线程数
    • data
    • master
    Χ

    index.unassigned.node_left.delayed_timeout

    一个node脱离集群后多长时间之外才开始进行一系列的备份操作
    • data
    • master

    index.search.slowlog.threshold.query.warn

    query慢日志时间设置
    • data

    index.search.slowlog.threshold.fetch.warn

    fetch慢日志时间设置
    • data

    index.indexing.slowlog.threshold.index.warn

    index慢日志时间设置
    • data

    monitor.jvm.gc.old.warn

    gc时间设置



    • data
    • master




    monitor

    monitor.jvm.gc.old.info

    • data
    • master

    monitor.jvm.gc.young.warn

    • data
    • master

    monitor.jvm.gc.young.info

    • data
    • master

    script.inline

    是否支持script表达式搜索

    • data
    Χ

    script

    script.indexed

    • data
    Χ

    path.logs

    log日志路径
    • data
    • master
    Χ

    path

    path.data

    存储数据路径
    • data
    • master
    Χ

    network.host

    对外发布本机ip
    • data
    • master
    Χ Χ

    network


    transport.tcp.port

    通信端口
    • data
    • master
    Χ

    http.port

    http端口
    • data
    • master
    Χ

    discovery.zen.ping.multicast.enabled

    是否开启相同集群名称则组成集群
    • data
    • master
    Χ Χ

    discovery


    discovery.zen.ping.unicast.hosts

    单播机器列表
    • data
    • master
    Χ

    discovery.zen.minimum_master_nodes

    组成master集群的最小节点数
    • master
    Χ

    gateway.recover_after_data_nodes

    full restart 参数设置






    • data
    Χ

    gateway






    gateway.expected_data_nodes

    • data
    Χ

    gateway.expected_master_nodes

    • master
    Χ

    gateway.recover_after_master_nodes

    • master
    Χ

    gateway.expected_nodes

    • data
    • master
    Χ

     gateway.recover_after_nodes

    • data
    • master
    Χ

    gateway.recover_after_time

    • data
    • master
    Χ

    action.disable_delete_all_indices

    是否允许全部删除
    • data
    • master
    Χ Χ

    action

    action.destructive_requires_name

    是否允许正则表达式删除
    • data
    • master
    Χ Χ

    shield.enabled

    是否支持shield
    • data
    • master
    Χ Χ shield

    插件

    • head
     
     
     
     
     

    ElasticSearch 配置文件译文解析

    [日期:2015-02-28] 来源:Linux社区  作者:an74520 [字体:  ]

    ##################### ElasticSearch 配置示例 #####################


    # This file contains an overview of various configuration settings,
    # targeted at operations staff. Application developers should
    # consult the guide at <http://elasticsearch.org/guide>.
    # 这个文件包含了各种配置的概览,旨在配置与运行操作相关的东西。
    # 应用程序开发人员应该咨询<http://elasticsearch.org/guide>
    #
    # The installation procedure is covered at
    # <http://elasticsearch.org/guide/reference/setup/installation.html>.
    # 安装过程在这里有<http://elasticsearch.org/guide/reference/setup/installation.html>.
    #
    #
    # ElasticSearch comes with reasonable defaults for most settings,
    # so you can try it out without bothering with configuration.
    # ElasticSearch 已经提供了大部分设置,都是合理的默认配置。
    # 所以你不必进行烦人的配置就可以尝试一下。
    #
    # Most of the time, these defaults are just fine for running a production
    # cluster. If you're fine-tuning your cluster, or wondering about the
    # effect of certain configuration option, please _do ask_ on the
    # mailing list or IRC channel [http://elasticsearch.org/community].
    # 大多数时候,这些默认的配置就足以运行一个生产集群了。
    # 如果你想优化你的集群,或者对一个特定的配置选项的作用好奇,你可以访问邮件列表
    # 或者IRC频道[http://elasticsearch.org/community].
    #


    # Any element in the configuration can be replaced with environment variables
    # by placing them in ${...} notation. For example:
    # 配置中的任何一个元素都可以被环境变量取代,这些环境变量使用${...}符号占位
    # 例如:
    # node.rack: ${RACK_ENV_VAR}


    # See <http://elasticsearch.org/guide/reference/setup/configuration.html>
    # for information on supported formats and syntax for the configuration file.
    # 查看<http://elasticsearch.org/guide/reference/setup/configuration.html>了解更多
    # 的可支持的格式和配置文件的语法。


    ################################### 集群 ###################################


    # Cluster name identifies your cluster for auto-discovery. If you're running
    # multiple clusters on the same network, make sure you're using unique names.
    # 集群名称标识了你的集群,自动探查会用到它。
    # 如果你在同一个网络中运行多个集群,那就要确保你的集群名称是独一无二的。
    #
    # cluster.name: elasticsearch


    #################################### 节点 #####################################


    # Node names are generated dynamically on startup, so you're relieved
    # from configuring them manually. You can tie this node to a specific name:
    # 节点名称会在启动的时候自动生成,所以你可以不用手动配置。你也可以给节点指定一个
    # 特定的名称
    #
    # node.name: "Franz Kafka"


    # Every node can be configured to allow or deny being eligible as the master,
    # and to allow or deny to store the data.
    # 每一个节点是否允许被选举成为主节点,是否允许存储数据,都是可以配置的
    #
    #
    # Allow this node to be eligible as a master node (enabled by default):
    # 允许这个节点被选举为一个主节点(默认为允许)
    #
    #
    # node.master: true
    #
    # Allow this node to store data (enabled by default):
    # 允许这个节点存储数据(默认为允许)
    #
    # node.data: true


    # You can exploit these settings to design advanced cluster topologies.
    # 你可以利用这些设置设计高级的集群拓扑
    #
    # 1. You want this node to never become a master node, only to hold data.
    #    This will be the "workhorse" of your cluster.
    # 1. 你不想让这个节点成为一个主节点,只想用来存储数据。
    #    这个节点会成为你的集群的“负载器”
    #
    # node.master: false
    # node.data: true
    #
    # 2. You want this node to only serve as a master: to not store any data and
    #    to have free resources. This will be the "coordinator" of your cluster.
    # 2. 你想让这个节点成为一个主节点,并且不用来存储任何数据,并且拥有空闲资源。
    #    这个节点会成为你集群中的“协调器”
    #
    # node.master: true
    # node.data: false
    #
    # 3. You want this node to be neither master nor data node, but
    #    to act as a "search load balancer" (fetching data from nodes,
    #    aggregating results, etc.)
    # 4. 你既不想让这个节点变成主节点也不想让其变成数据节点,只想让其成为一个“搜索负载均衡器”
    #    (从节点中获取数据,聚合结果,等等)
    #
    # node.master: false
    # node.data: false


    # Use the Cluster Health API [http://localhost:9200/_cluster/health], the
    # Node Info API [http://localhost:9200/_cluster/nodes] or GUI tools
    # such as <http://github.com/lukas-vlcek/bigdesk> and
    # <http://mobz.github.com/elasticsearch-head> to inspect the cluster state.
    # 使用集群体检API[http://localhost:9200/_cluster/health] ,
    # 节点信息API[http://localhost:9200/_cluster/nodes] 或者GUI工具例如:
    # <http://github.com/lukas-vlcek/bigdesk>和<http://mobz.github.com/elasticsearch-head>
    # 可以查看集群状态
    #


    # A node can have generic attributes associated with it, which can later be used
    # for customized shard allocation filtering, or allocation awareness. An attribute
    # is a simple key value pair, similar to node.key: value, here is an example:
    # 一个节点可以附带一些普通的属性,这些属性可以在后面的自定义分片分配过滤或者allocation awareness中使用。
    # 一个属性就是一个简单的键值对,类似于node.key: value, 这里有一个例子:
    #
    # node.rack: rack314


    # By default, multiple nodes are allowed to start from the same installation location
    # to disable it, set the following:
    # 默认的,多个节点允许从同一个安装位置启动。若想禁止这个特性,按照下面所示配置:
    # node.max_local_storage_nodes: 1
    #


    #################################### 索引 ####################################


    # You can set a number of options (such as shard/replica options, mapping
    # or analyzer definitions, translog settings, ...) for indices globally,
    # in this file.
    # 你可以在这个文件中为所有的索引设置一系列的全局操作(例如 分片/副本 操作,mapping(映射)
    # 或者分词器定义,translog配置,...)

    #
    # Note, that it makes more sense to configure index settings specifically for
    # a certain index, either when creating it or by using the index templates API.
    # 提示,针对一个特定的索引进行配置更合理,不论是在创建索引还是使用索引模板API的时候。
    #
    #
    # See <http://elasticsearch.org/guide/reference/index-modules/> and
    # <http://elasticsearch.org/guide/reference/api/admin-indices-create-index.html>
    # for more information.
    # 详情见<http://elasticsearch.org/guide/reference/index-modules/>和
    # <http://elasticsearch.org/guide/reference/api/admin-indices-create-index.html>


    # Set the number of shards (splits) of an index (5 by default):
    # 设置一个索引的分片数量(默认为5)
    #
    # index.number_of_shards: 5


    # Set the number of replicas (additional copies) of an index (1 by default):
    # 设置一个索引的副本数量(默认为1)
    #
    # index.number_of_replicas: 1


    # Note, that for development on a local machine, with small indices, it usually
    # makes sense to "disable" the distributed features:
    # 注意,为了使用小的索引在本地机器上开发,禁用分布式特性是合理的做法。
    #
    #
    # index.number_of_shards: 1
    # index.number_of_replicas: 0


    # These settings directly affect the performance of index and search operations
    # in your cluster. Assuming you have enough machines to hold shards and
    # replicas, the rule of thumb is:
    # 这些设置会直接影响索引和查询操作的在集群中的性能。假如你有足够的机器来放分片和副本,
    # 最佳实践是:
    #
    # 1. Having more *shards* enhances the _indexing_ performance and allows to
    #    _distribute_ a big index across machines.
    # 1. 索引分片分的多一些,可以提高索引的性能,并且把一个大的索引分布到机器中去。
    # 2. Having more *replicas* enhances the _search_ performance and improves the
    #    cluster _availability_.
    # 2. 副本分片分的多一些,可以提高搜索的性能,并且提高集群的可用性。
    #
    # The "number_of_shards" is a one-time setting for an index.
    # "number_of_shards"对一个索引来说只能配置一次
    #
    # The "number_of_replicas" can be increased or decreased anytime,
    # by using the Index Update Settings API.
    # "number_of_replicas"在任何时候都可以增加或减少,通过Index Update Settings(索引更新配置)API可以做到这一点。
    #
    #
    # ElasticSearch takes care about load balancing, relocating, gathering the
    # results from nodes, etc. Experiment with different settings to fine-tune
    # your setup.
    # ElasticSearch 会维护load balancin(负载均衡),relocating(重定位),合并来自各个节点的结果等等。
    # 你可以实验不同的配置来进行优化。
    #


    # Use the Index Status API (<http://localhost:9200/A/_status>) to inspect
    # the index status.
    # 使用Index Status(索引状态)API (<http://localhost:9200/A/_status>)查看索引状态


    #################################### Paths(路径) ####################################


    # Path to directory containing configuration (this file and logging.yml):
    # 包含配置(这个文件和logging.yml)的目录的路径
    #
    # path.conf: /path/to/conf


    # Path to directory where to store index data allocated for this node.
    # 存储这个节点的索引数据的目录的路径

    # path.data: /path/to/data
    #
    # Can optionally include more than one location, causing data to be striped across
    # the locations (a la RAID 0) on a file level, favouring locations with most free
    # space on creation. For example:
    # 可以随意的包含不止一个位置,这样数据会在文件层跨越多个位置(a la RAID 0),创建时会
    # 优先选择大的剩余空间的位置
    #
    # path.data: /path/to/data1,/path/to/data2


    # Path to temporary files:
    # 临时文件的路径
    #
    # path.work: /path/to/work


    # Path to log files:
    # 日志文件的路径
    #
    # path.logs: /path/to/logs


    # Path to where plugins are installed:
    # 插件安装路径
    #
    # path.plugins: /path/to/plugins


    #################################### 插件 ###################################


    # If a plugin listed here is not installed for current node, the node will not start.
    # 如果当前结点没有安装下面列出的插件,结点不会启动
    #
    # plugin.mandatory: mapper-attachments,lang-groovy


    ################################### 内存 ####################################


    # ElasticSearch performs poorly when JVM starts swapping: you should ensure that
    # it _never_ swaps.
    # 当JVM开始swapping(换页)时ElasticSearch性能会低下,你应该保证它不会换页
    #
    #
    # Set this property to true to lock the memory:
    # 设置这个属性为true来锁定内存
    #
    # bootstrap.mlockall: true


    # Make sure that the ES_MIN_MEM and ES_MAX_MEM environment variables are set
    # to the same value, and that the machine has enough memory to allocate
    # for ElasticSearch, leaving enough memory for the operating system itself.
    # 确保ES_MIN_MEM和ES_MAX_MEM环境变量设置成了同一个值,确保机器有足够的内存来分配
    # 给ElasticSearch,并且保留足够的内存给操作系统
    #
    #
    # You should also make sure that the ElasticSearch process is allowed to lock
    # the memory, eg. by using `ulimit -l unlimited`.
    # 你应该确保ElasticSearch的进程可以锁定内存,例如:使用`ulimit -l unlimited`
    #


    ############################## Network(网络) 和 HTTP ###############################


    # ElasticSearch, by default, binds itself to the 0.0.0.0 address, and listens
    # on port [9200-9300] for HTTP traffic and on port [9300-9400] for node-to-node
    # communication. (the range means that if the port is busy, it will automatically
    # try the next port).
    # 默认的ElasticSearch把自己和0.0.0.0地址绑定,HTTP传输的监听端口在[9200-9300],节点之间
    # 通信的端口在[9300-9400]。(范围的意思是说如果一个端口已经被占用,它将会自动尝试下一个端口)
    #
    #


    # Set the bind address specifically (IPv4 or IPv6):
    # 设置一个特定的绑定地址(IPv4 or IPv6):
    #
    # network.bind_host: 192.168.0.1


    # Set the address other nodes will use to communicate with this node. If not
    # set, it is automatically derived. It must point to an actual IP address.
    # 设置其他节点用来与这个节点通信的地址。如果没有设定,会自动获取。
    # 必须是一个真实的IP地址。
    #
    # network.publish_host: 192.168.0.1


    # Set both 'bind_host' and 'publish_host':
    # 'bind_host'和'publish_host'都设置
    #
    # network.host: 192.168.0.1


    # Set a custom port for the node to node communication (9300 by default):
    # 为节点之间的通信设置一个自定义端口(默认为9300)
    #
    # transport.tcp.port: 9300


    # Enable compression for all communication between nodes (disabled by default):
    # 为所有的节点间的通信启用压缩(默认为禁用)
    #
    # transport.tcp.compress: true


    # Set a custom port to listen for HTTP traffic:
    # 设置一个监听HTTP传输的自定义端口
    #
    # http.port: 9200


    # Set a custom allowed content length:
    # 设置一个自定义的允许的内容长度
    #
    # http.max_content_length: 100mb


    # Disable HTTP completely:
    # 完全禁用HTTP
    #
    # http.enabled: false


    ################################### Gateway ###################################


    # The gateway allows for persisting the cluster state between full cluster
    # restarts. Every change to the state (such as adding an index) will be stored
    # in the gateway, and when the cluster starts up for the first time,
    # it will read its state from the gateway.
    # Gateway支持持久化集群状态。状态的每一个改变(例如添加一个索引)将会被存储在gateway,
    # 当集群第一次启动时,它会从gateway中读取它的状态。
    #


    # There are several types of gateway implementations. For more information,
    # see <http://elasticsearch.org/guide/reference/modules/gateway>.
    # 还有多种类型的gateway实现。详情见<http://elasticsearch.org/guide/reference/modules/gateway>


    # The default gateway type is the "local" gateway (recommended):
    # 默认的gateway类型是 "local" gateway(推荐)
    #
    # gateway.type: local


    # Settings below control how and when to start the initial recovery process on
    # a full cluster restart (to reuse as much local data as possible when using shared
    # gateway).
    # 下面的配置控制怎样以及何时启动一整个集群重启的初始化恢复过程
    # (当使用shard gateway时,是为了尽可能的重用local data(本地数据))
    #


    # Allow recovery process after N nodes in a cluster are up:
    # 一个集群中的N个节点启动后,才允许进行恢复处理
    #
    # gateway.recover_after_nodes: 1


    # Set the timeout to initiate the recovery process, once the N nodes
    # from previous setting are up (accepts time value):
    # 设置初始化恢复过程的超时时间,超时时间从上一个配置中配置的N个节点启动后算起
    #
    # gateway.recover_after_time: 5m


    # Set how many nodes are expected in this cluster. Once these N nodes
    # are up (and recover_after_nodes is met), begin recovery process immediately
    # (without waiting for recover_after_time to expire):
    # 设置这个集群中期望有多少个节点。一旦这N个节点启动(并且recover_after_nodes也符合),
    # 立即开始恢复过程(不等待recover_after_time超时)
    #
    # gateway.expected_nodes: 2


    ############################# Recovery Throttling (节点恢复限流阀) #############################


    # These settings allow to control the process of shards allocation between
    # nodes during initial recovery, replica allocation, rebalancing,
    # or when adding and removing nodes.
    # 这些配置允许在初始化恢复,副本分配,再平衡,或者添加和删除节点时控制节点间的分片分配
    #


    # Set the number of concurrent recoveries happening on a node:
    # 设置一个节点的并行恢复数
    #
    # 1. During the initial recovery 
    # 1. 初始化恢复期间
    #
    # cluster.routing.allocation.node_initial_primaries_recoveries: 4
    #
    # 2. During adding/removing nodes, rebalancing, etc 
    # 2. 添加/删除节点,再平衡等期间
    #
    # cluster.routing.allocation.node_concurrent_recoveries: 2


    # Set to throttle throughput when recovering (eg. 100mb, by default unlimited):
    # 设置恢复时的吞吐量(例如,100mb,默认没有上限)
    #
    # indices.recovery.max_size_per_sec: 0


    # Set to limit the number of open concurrent streams when
    # recovering a shard from a peer:
    # 设置当一个分片从对等点恢复时能够打开的并发流的上限
    #


    # indices.recovery.concurrent_streams: 5


    ################################## Discovery(探查) ##################################


    # Discovery infrastructure ensures nodes can be found within a cluster
    # and master node is elected. Multicast discovery is the default.
    # 探查机制能够保障一个集群中的节点能被找到,并且主节点能够被选举出来。
    # 默认的方式为多播。


    # Set to ensure a node sees N other master eligible nodes to be considered
    # operational within the cluster. Set this option to a higher value (2-4)
    # for large clusters (>3 nodes):
    # 这个选项用来设置一个节点可以看到其他N个在集群中具有可操性的并且具有当选主节点资格的节点
    # 对于大的集群(大于3个节点),这个选项应该设置成一个高一点的值(2-4)
    #
    # discovery.zen.minimum_master_nodes: 1


    # Set the time to wait for ping responses from other nodes when discovering.
    # Set this option to a higher value on a slow or congested network
    # to minimize discovery failures:
    # 设置在探查过程中从其他节点返回ping的回应的等待时间
    # 在一个低速或者拥堵的网络环境中这个选项应该设置的大一些,这样可以降低探查失败的可能性。
    #
    # discovery.zen.ping.timeout: 3s


    # See <http://elasticsearch.org/guide/reference/modules/discovery/zen.html>
    # for more information.
    # 详情见<http://elasticsearch.org/guide/reference/modules/discovery/zen.html>


    # Unicast discovery allows to explicitly control which nodes will be used
    # to discover the cluster. It can be used when multicast is not present,
    # or to restrict the cluster communication-wise.
    # 利用单播探查,我们可以显示的指定哪些节点在探查集群过程中会被用到。
    # 当多播不可用,或者需要约束集群的通信时可以使用单播探查。
    #
    # 1. Disable multicast discovery (enabled by default):
    # 1. 禁用多播探查(默认可用)
    #
    # discovery.zen.ping.multicast.enabled: false
    #
    # 2. Configure an initial list of master nodes in the cluster
    #    to perform discovery when new nodes (master or data) are started:
    # 2. 这是一个集群中的主节点的初始列表,当节点(主节点或者数据节点)启动时使用这个列表进行探查
    #
    #
    # discovery.zen.ping.unicast.hosts: ["host1", "host2:port", "host3[portX-portY]"]


    # EC2 discovery allows to use AWS EC2 API in order to perform discovery.
    # 为了执行探查EC2探查允许使用AWS EC2 API
    #
    # You have to install the cloud-aws plugin for enabling the EC2 discovery.
    # 想要启用EC2探查功能,你必须安装cloud-aws插件
    #
    # See <http://elasticsearch.org/guide/reference/modules/discovery/ec2.html>
    # for more information.
    # 详情见<http://elasticsearch.org/guide/reference/modules/discovery/ec2.html>
    #
    #
    # See <http://elasticsearch.org/tutorials/2011/08/22/elasticsearch-on-ec2.html>
    # for a step-by-step tutorial.
    # 详情见<http://elasticsearch.org/tutorials/2011/08/22/elasticsearch-on-ec2.html>


    ################################## Slow Log(慢日志) ##################################


    # Shard level query and fetch threshold logging.
    #


    #index.search.slowlog.threshold.query.warn: 10s
    #index.search.slowlog.threshold.query.info: 5s
    #index.search.slowlog.threshold.query.debug: 2s
    #index.search.slowlog.threshold.query.trace: 500ms


    #index.search.slowlog.threshold.fetch.warn: 1s
    #index.search.slowlog.threshold.fetch.info: 800ms
    #index.search.slowlog.threshold.fetch.debug: 500ms
    #index.search.slowlog.threshold.fetch.trace: 200ms


    #index.indexing.slowlog.threshold.index.warn: 10s
    #index.indexing.slowlog.threshold.index.info: 5s
    #index.indexing.slowlog.threshold.index.debug: 2s
    #index.indexing.slowlog.threshold.index.trace: 500ms


    ################################## GC Logging ################################


    #monitor.jvm.gc.ParNew.warn: 1000ms
    #monitor.jvm.gc.ParNew.info: 700ms
    #monitor.jvm.gc.ParNew.debug: 400ms


    #monitor.jvm.gc.ConcurrentMarkSweep.warn: 10s
    #monitor.jvm.gc.ConcurrentMarkSweep.info: 5s
    #monitor.jvm.gc.ConcurrentMarkSweep.debug: 2s

  • 相关阅读:
    结构化编程:确定迭代循环控制for
    为什么就业难,因为信息化的威力
    结构化编程:单支条件选择控制
    C/C++ 标准输入输出重定向
    C语言中用scanf连续输入两个字符类型的问题
    void value not ignored as it ought to be
    不要做浮躁的嵌入式系统工程师
    INI配置文件的格式
    Linux 下串口编程入门
    时钟周期及秒(s) 毫秒(ms) 微秒(μs) 纳秒(ns) 皮秒(ps)之间转换
  • 原文地址:https://www.cnblogs.com/timssd/p/7406921.html
Copyright © 2011-2022 走看看