zoukankan      html  css  js  c++  java
  • elasticsearch 分布式部署

    修改配置文件 /config/elasticsearch.yml
    我用两台机器,内网地址分别为230 和 231

    处理启动报错一:

    [2017-01-12T15:55:55,433][INFO ][o.e.b.BootstrapCheck     ] [SfD5sIh] bound or publishing to a non-loopback or non-link-local address, enforcing bootstrap checks
    ERROR: bootstrap checks failed
    max virtual memory areas vm.max_map_count [65530] is too low, increase to at least [262144]
    
    • 临时提高了vm.max_map_count的大小,此操作需要root权限:
      sudo  sysctl -w vm.max_map_count=262144
      sysctl -a|grep vm.max_map_count
    • 永久修改vm.max_map_count:

      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。
     

    处理启动报错二:

      max file descriptors [4096] for elasticsearch process likely too low, increase to at least [65536]

      修改/etc/security/limits.conf文件,添加或修改如下行:

      *        hard    nofile           65536
      *        soft    nofile           65536
     
     

    处理启动报错三:

      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
     
     

      补:启动异常:ERROR: bootstrap checks failed
      system call filters failed to install; check the logs and fix your configuration or disable system call filters at your own risk

      问题原因:因为Centos6不支持SecComp,而ES5.2.1默认bootstrap.system_call_filter为true进行检测,所以导致检测失败,失败后直接导致ES不能启动。详见 :https://github.com/elastic/elasticsearch/issues/22899

      解决方法:在elasticsearch.yml中配置bootstrap.system_call_filter为false,注意要在Memory下面:
      bootstrap.memory_lock: false
      bootstrap.system_call_filter: false
     

      近日,单台机器,8个节点的es集群,8个节点都正常started了,但是就是无法形成集群,后来看日志,日志中出现一堆的MasterNotDiscoveredException这种异常,完整日志如下:

    [2016-04-27 15:08:22,445][DEBUG][action.admin.cluster.health] [es.10.16.66.152.0] no known master node, scheduling a retry
    [2016-04-27 15:08:52,457][INFO ][rest.suppressed          ] /_cat/health Params: {h=node.total}
    MasterNotDiscoveredException[waited for [30s]]
            at org.elasticsearch.action.support.master.TransportMasterNodeAction$4.onTimeout(TransportMasterNodeAction.java:160)
            at org.elasticsearch.cluster.ClusterStateObserver$ObserverClusterStateListener.onTimeout(ClusterStateObserver.java:239)
            at org.elasticsearch.cluster.service.InternalClusterService$NotifyTimeout.run(InternalClusterService.java:630)
            at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
            at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
            at java.lang.Thread.run(Thread.java:745)
      查看了每一个节点的elasticsearch.yml这个文件之后,发现其中只有一个节点的node.master这个配置为true,而其中discovery.zen.minimum_master_nodes这个配置的值却是5,而有资格成为主节点的只有一个,所以永远也形成不了集群啊。
    结果就是要么把其他节点的node.master都设置为true,要么discovery.zen.minimum_master_nodes将这个配置改为1。
     
     
    错误的解决办法均是从网上扒的,看官就不要吐槽了 虽然官网都有相关说明,但有现成的就不愿意看英文文档,谁叫英文不好呢
  • 相关阅读:
    Cuckoo Hashing
    Microsoft Windows的消息循环
    Simplex, Full-Duplex and Half-Duplex Operation
    Linux 技巧:让进程在后台运行更可靠的几种方法
    https://learnku.com/docs/go-blog/qihoo/6532 。 heap size went up to 69G, with maximum garbage collection (GC)
    Go GC: Latency Problem Solved
    Sapphire: Copying GC Without Stopping the World
    The Go Blog Getting to Go: The Journey of Go's Garbage Collector
    xml CDATA
    Joseph cicyle's algorithm
  • 原文地址:https://www.cnblogs.com/hughtxp/p/6667844.html
Copyright © 2011-2022 走看看