启动 Elasticsearch 报错 bootstrap checks failed 解决

文章

林里克斯

Elasticsearch 是一个基于 Lucene 的搜索服务器。它提供了一个分布式多用户能力的全文搜索引擎，基于 RESTful web 接口。Elasticsearch 是用 Java 语言开发的，并作为 Apache 许可条款下的开放源码发布，是一种流行的企业级搜索引擎。

实验平台：CentOS 7.7.1908

Elasticsearch Version: 7.5.1

Java Version： OpenJDK 13.0.1

1.报错信息

ERROR: [5] bootstrap checks failed
[1]: max file descriptors [4096] for elasticsearch process is too low, increase to at least [65535]
[2]: max virtual memory areas vm.max_map_count [65530] is too low, increase to at least [262144]
[3]: max number of threads [2048] for user [tongtech] is too low, increase to at least [4096]
[4]: the default discovery settings are unsuitable for production use; at least one of [discovery.seed_hosts, discovery.seed_providers, cluster.initial_master_nodes] must be configured
[5]:memory locking requested for elasticsearch process but memory is not locked
[6]: initial heap size [2147483648] not equal to maximum heap size [32210157568]; this can cause resize pauses and prevents mlockall from locking the entire heap

2.解决报错

报错 [1]

[1]: max file descriptors [4096] for elasticsearch process is too low, increase to at least [65535]
#最大文件描述符 [4096] 较小，调整至最低 [65535]

编辑 /etc/security/limits.conf 填加以下内容

$ vim /etc/security/limits.conf

* soft nofile 65536
* hard nofile 65536
#此文件修改后需要重开窗口，如果只直接在物理机上操作需重启物理机。

报错[2]

[2]: max virtual memory areas vm.max_map_count [65530] is too low, increase to at least [262144]
#最大虚拟内存 [65530] 较小，调整至最低 [262144]

编辑 /etc/sysctl.conf 添加以下内容：

$ vim /etc/sysctl.conf
vm.max_map_count=262144

保存后，需执行一下内容生效：

$ sysctl -p

报错 [3]

[3]: max number of threads [2048] for user [tongtech] is too low, increase to at least [4096]
#最大线程数 [2048] 较小，调整至最低 [4096]

编辑 /etc/security/limits.d/20-nproc.conf

$ vim /etc/security/limits.d/20-nproc.conf  （前面数字可能不一致）

 *          soft    nproc     2048
#将 2048 改为 4096 或更大

报错[4]

[4]: the default discovery settings are unsuitable for production use; at least one of [discovery.seed_hosts, discovery.seed_providers, cluster.initial_master_nodes] must be configured
# elasticsearch.yml 不适用于生产环境

编辑配置文件 elasticsearch.yml

$ vim /data/elasticsearch/config/elasticsearch.yml

discovery.seed_hosts: ["192.168.50.151","192.168.50.152","192.168.50.150"]
cluster.initial_master_nodes: ["192.168.50.151","192.168.50.152","192.168.50.150"]

报错[5]

[5]:memory locking requested for elasticsearch process but memory is not locked

elasticsearch 官网建议生产环境需要设置 bootstrap.memory_lock: true ，也可以使用以下配置解决报错。

$ vim /etc/security/limits.conf
user soft memlock unlimited
user hard memlock unlimited
#user 为启动 elasticsearch 的用户

报错[6]

elasticsearch jvm 内存分配不够导致，调整一下值重启即可

$ vim /data/elasticsearch/config/jvm.options

-Xms16g
-Xmx16g

elasticsearch 集群启动出现 failed to send join request to master

具体报错如下

[node-2] failed to send join request to master [{node-1}{10.10.10.131}{10.10.10.130:9300}], reason [RemoteTransportException[[node-1][10.10.10.131:9300][internal:discovery/data/]]; nested: IllegalArgumentException[can't add node {node-2}{10.10.10.130}{10.10.10.131:9301}, found existing node {node-1}{10.10.10.130}{10.10.11.131:9300} with the same id but is a different node instance]; ]

具体原因是因为我把 node-1 误删了，将另一台机器上的 node-2 拷贝过来改了配置文件启动

解决这个问题将 /data/elasticsearch/data/ 下的数据删除掉后重新启动就可以了。再次警示删除操作一定要谨慎！谨慎！谨慎！

Over~