x-pack搭建破解请参考之前的安装文档ELK+filebeat+x-pack平台搭建

1. 一些命令

• 查询所有用户

[elk@elk ~]$  curl -XGET -u elastic 'localhost:9200/_xpack/security/user?pretty'
配置http ssl认证后需要使用证书
[elk@elk ~]$  curl -XGET -u elastic 'https://localhost3:9200/_xpack/security/user?pretty' --key client.key --cert client.cer --cacert client-ca.cer -k -v

• 查询所有roles

[elk@elk ~]$  curl -XGET -u elastic 'localhost:9200/_xpack/security/role?pretty'

• 创建用户[chengen]

[elk@elk ~]$ curl -X POST -u elastic "localhost:9200/_xpack/security/user/chengen" -H 'Content-Type: application/json' -d'
{
  "password" : "your passwd",
  "roles" : [ "admin"],
  "full_name" : "chengen",
  "email" : "xxxxxx@qq.com",
  "metadata" : {
    "intelligence" : 7
  }
}
'

画外音：加上“-u elastic”是因为只有elastic用户有管理用户权限，另外，请求参数后面可以带上?pretty，这样返回的格式会好看一点儿

• 修改密码

curl -X POST -u elastic "localhost:9200/_xpack/security/user/test/_password" -H 'Content-Type: application/json' -d'
{
  "password" : "your passwd"
}
'

修改username为test的密码，注意test是username不是fullname

• 禁用/启动/删除用户

curl -X PUT -u elastic "localhost:9200/_xpack/security/user/test/_disable"  
curl -X PUT -u elastic "localhost:9200/_xpack/security/user/test/_enable"
curl -X DELETE -u elastic "localhost:9200/_xpack/security/user/test"

创建beats_admin1的roles，该用户组对filebeat有all权限，对.kibana有manage，read，index权限

curl -X POST -u elastic 'localhost:9200/_xpack/security/role/beats_admin1' -H 'Content-Type: application/json' -d '{
  "indices" : [
    {
      "names" : [ "filebeat*" ],
      "privileges" : [ "all" ]
    },
    {
      "names" : [ ".kibana*" ],
      "privileges" : [ "manage", "read", "index" ]
    }
  ]
}'

安全API:https://www.elastic.co/guide/en/elasticsearch/reference/current/security-api.html

2. ElasticSearchHead

当你再次打开浏览器ElasticSearchHead插件的时候，会提示你输入密码

image.png

3. kibana再次打开kibana的时候会提示你输入密码

image.png

image.png

image.png

启用了X-Pack安全性之后，如果你加载一个Kibana指示板，该指示板访问你没有权限查看的索引中的数据，那么你将得到一个索引不存在的错误。X-Pack安全性目前还没有提供一种方法来控制哪些用户可以加载哪些仪表板。

4.基于角色的权限控制

该功能的入口在 Management -> Users/Roles。Users 可以方便的管理用户并且对其赋予角色，角色和权限挂钩。Roles 可以方便的管理角色，对其进行赋权。Role 是 Permission 的集合，Permission 是 Privilege 的集合，下面来说说权限：

集群权限(Cluster Privilege)；
Run As Privileges：可以使得新建角色拥有所选用户集的权限；
索引权限(Index Privilege)：
Indices：指定在哪些索引上赋权；
Privileges：指定赋予哪些权限；
Granted Documents Query(可选)：指定在哪些 Query 上赋权；
Granted Fields(可选)：指定在哪些 fields 上赋权；

只将单个索引的权限放开给某些用户

创建role

image.png

授予role的elasticsearch权限

image.png

image.png

5.x-pack权限介绍

内置账号

Security-Roles权限

集群权限和索引权限请参考

https://www.elastic.co/guide/en/elastic-stack-overview/6.7/security-privileges.html#privileges-list-cluster

6. x-pack功能开关

修改配置后需要重启三个服务

![image.png](https://upload-images.jianshu.io/upload_images/13799508-78a2203d17a6c8af.png?imageMogr2/auto-orient/strip%7CimageView2/2/w/1240)

 nohup ./bin/logstash -f config/logstash.conf >/dev/null 2>&1 &
./bin/elasticsearch -d
./bin/kibana --verbose > kibana.log 2>&1 &
./filebeat -c filebeat_secure.yml -e -v

7. 对es集群，kibana，logsatsh，filebeat配置SSL，TLS和HTTPS安全

注意：

1.创建es证书证书一定要加--dns 和--ip否则后期通信会报错，如7.5记录的问题。
2.如果是集群请将es1中生成的证书elastic-stack-ca.p12拷贝到es2，es3并重新依次生成带有dns和ip的elastic-certificates.p12证书
3.如果集群安装x-pack是一定要配置证书加密的，否则集群无法正常启动
4.可以参照我下面的步骤进行安全证书操作，也可以参考

https://www.elastic.co/cn/blog/configuring-ssl-tls-and-https-to-secure-elasticsearch-kibana-beats-and-logstash#run-filebeat

配置证书，因为配置完才发现了这篇文档，还没来得及尝试，感觉文章写的配置证书方法比较简单，有兴趣同学可以尝试一下。

7.1 es集群安全配置

• elasticsearch-certuti创建证书

bin/elasticsearch-certutil ca    //创建了elastic-stack-ca.p12
bin/elasticsearch-certutil cert --ca config/certs/elastic-stack-ca.p12 --dns 192.168.100.203 --ip 192.168.100.203     //创建了elastic-certificates.p12

config目录下创建certs文件夹，放入证书elastic-stack-ca.p12和elastic-certificates.p12

• 传输通信配置(第一步--集群节点间的认证)

xpack.security.transport.ssl.enabled: true
xpack.security.transport.ssl.verification_mode: certificate
xpack.security.transport.ssl.keystore.path: certs/elastic-certificates.p12
xpack.security.transport.ssl.truststore.path: certs/elastic-certificates.p12

• 内置用户定义密码(第二步，需要在配置HTTP ssl之前配置，因为设置密码的命令通过不安全的http与集群通信)

bin/elasticsearch-setup-passwords interactive

• HTTP SSL加密配置(第三步--配置es集群的https)

对于http通信,Elasticsearch节点仅用作服务器，在elasticsearch.yml文件中指定如下

xpack.security.http.ssl.enabled: true
xpack.security.http.ssl.keystore.path: certs/elastic-certificates.p12
xpack.security.http.ssl.truststore.path: certs/elastic-certificates.p12
xpack.security.http.ssl.client_authentication: optional

综上，在elasticsearch.yml配置中定义以下内容如下：

xpack.security.enabled: true

xpack.security.transport.ssl.enabled: true
xpack.security.transport.ssl.verification_mode: certificate
xpack.security.transport.ssl.keystore.path: certs/elastic-certificates.p12
xpack.security.transport.ssl.truststore.path: certs/elastic-certificates.p12

xpack.security.http.ssl.enabled: true
xpack.security.http.ssl.keystore.path: certs/elastic-certificates.p12
xpack.security.http.ssl.truststore.path: certs/elastic-certificates.p12
xpack.security.http.ssl.client_authentication: optional

更改完毕后需要重启es集群此时访问集群需要https认证。此时elasticsearch head需要配置https://192.168.100.203:9200/

• 报错：
  如果在elasticsearch.yml中配置了 xpack.security.authc.realms.pki1.type: pki
  启动kibana后，会出现登录kibana地址时只能使用内置账户登录，如elastic，kibana等，后创建的自定义账户test等不可以登录。
  解决：取消 xpack.security.authc.realms.pki1.type: pki 就恢复正常

参考: 在Elasticsearch中启用安全性，TLS / SSL和PKI身份验证的分步指南

7.2 kibana.yml安全配置

• 配置kibana到es的证书通信(生成elastic-stack-ca.pem)

cd /elk/elasticsearch-6.7.0
openssl pkcs12 -in config/certs/elastic-stack-ca.p12 -clcerts -nokeys -chain -out elastic-stack-ca.pem
config目录下创建certs并将elastic-stack-ca.pem放入

• 配置kibana https访问
  创建kibana ssl证书（生成kibana.key和kibana.crt，在certs中新建ssll放入）

openssl req -subj '/CN=192.168.100.203/' -x509 -days $((100 * 365)) -batch -nodes -newkey rsa:2048 -keyout kibana.key -out kibana.crt

综上，在kibana.yml中配置文件如下:

#注意es地址为https
elasticsearch.hosts: ["https://192.168.100.203:9200"]

xpack.security.enabled: true
elasticsearch.username: "kibana"
elasticsearch.password: "your passwd"

#配置kibana https登录
server.ssl.enabled: true
server.ssl.certificate: /elk/kibana-6.7.0/config/certs/ssl/kibana.crt
server.ssl.key: /elk/kibana-6.7.0/config/certs/ssl/kibana.key

#kibana与es信任通信
elasticsearch.ssl.certificateAuthorities: config/certs/old/elastic-stack-ca.pem
elasticsearch.ssl.verificationMode: certificate

参考：
ELK的安全加固
kiban与logstash with x-pack
https://www.elastic.co/guide/en/kibana/6.3/configuring-tls.html

7.3 logstash安全配置

• 配置logstash与es集群通信
  将7.2中生成的elastic-stack-ca.pem放入到/elk/logstash-6.7.0/config/certs中
  在logstash中配置

xpack.monitoring.enabled: true
xpack.monitoring.elasticsearch.username: elastic
xpack.monitoring.elasticsearch.password: your passwd
xpack.monitoring.elasticsearch.hosts: ["https://192.168.100.203:9200","https://192.168.100.202:9200","https://192.168.100.201:9200"]

xpack.monitoring.elasticsearch.ssl.certificate_authority: "/elk/logstash-6.7.0/config/certs/elastic-stack-ca.pem"

xpack.monitoring.elasticsearch.ssl.verification_mode: certificate
xpack.monitoring.elasticsearch.sniffing: false
xpack.monitoring.collection.interval: 60s
xpack.monitoring.collection.pipeline.details.enabled: true

• 配置logstash与filebeat之间的通信
  生成zip文件，包含instance.crt，instance.key

bin/elasticsearch-certutil cert --pem -ca config/certs/elastic-stack-ca.p12 --dns 192.168.100.203 --ip 192.168.100.203    ##生成zip文件，包含instance.crt，instance.key

转换成filebeat可用格式,生成instance.pkcs8.key

openssl pkcs8 -in instance.key -topk8 -nocrypt -out instance.pkcs8.key

logstash-sample.conf配置

input {
  beats {
    port => 5044
    ssl => true
    ssl_key => "/elk/logstash-6.7.0/config/certs/instance/instance.pkcs8.key"
    ssl_certificate => "/elk/logstash-6.7.0/config/certs/instance/instance.crt"
  }
}

output {
  elasticsearch {
    user => "elastic"
    password => "your passwd"
    ssl => true
    ssl_certificate_verification => true
    cacert => "/elk/logstash-6.7.0/config/certs/elastic-stack-ca.pem"
    hosts => ["https://192.168.100.203:9200/","https://192.168.100.202:9200/","https://192.168.100.201:9200/"]
    index => "testlog"

  }
}

output将数据加密传输到es集群，input对filebeat传输的数据进行加密
如果logstash可以正常启动，说明配置正确

7.4 filebeat安全配置

将logsatsh中的ca证书：elastic-stack-ca.pem传输到beat所在的机器上，如：/opt/filebeat/certs/
修改filebeat.yml文件的output

output.logstash:
  # The Logstash hosts
    hosts: ["192.168.100.203:5044"]
    ssl.certificate_authorities: ["/opt/filebeat/certs/elastic-stack-ca.pem"]

重启filebeat，如果数据可以在kibana上正常显示，便是配置正常

7.5 问题处理

1 “Host name '192.168.100.203' does not match the certificate subject provided by the pee”
报错信息

[2019-06-24T13:57:08,032][ERROR][logstash.licensechecker.licensereader] Unable to retrieve license information from license server {:message=>"Host name '192.168.100.203' does not match the certificate subject provided by the peer (CN=instance)"}

解决：如果创建证书不添加dns和ip，es和logstash通过TLS通信时会此问题
PS 证书问题让我很抓狂，配置es集群以及logstash，kibana，filebest加密用了两天@@@，配置如下：

bin/elasticsearch-certutil cert --ca config/certs/elastic-stack-ca.p12 --dns 192.168.100.203 --ip 192.168.100.203

2 也是证书问题，和1类似
因为elk搭建在内网需要将外网（IP:1.2.3.4）的日志发送到elk，使用了frp内网穿透进行转发，外网机器在通过外网IP+端口访问时报错，如下：

2019-07-01T16:29:17.253+0800  ERROR pipeline/output.go:100  Failed to connect to backoff(async(tcp://1.2.3.4:5044)): x509: certificate is valid for 192.168.100.203, not 118.25.198.208
2019-07-01T16:29:17.253+0800  INFO  pipeline/output.go:93 Attempting to reconnect to backoff(async(tcp://1.2.3.4:5044)) with 1 reconnect attempt(s)

需要在7.3 logstash安全配置，配置logstash与filebeat之间的通信生成zip文件时，dns和IP加如外网IP

bin/elasticsearch-certutil cert --pem -ca config/certs/elastic-stack-ca.p12 --dns 192.168.100.203,1.2.3.4 --ip 192.168.100.203,1.2.3.4    ##生成zip文件，包含instance.crt，instance.key

接下来的步骤一致，只需要更改logstash中的证书instance.pkcs8.key和instance.crt，代码如下，filebeat的ca证书不需要更改，数据就可以正常传输过来

input {
  beats {
    port => 5044
    ssl => true
    ssl_key => "/elk/logstash-6.7.0/config/certs/instance/instance.pkcs8.key"
    ssl_certificate => "/elk/logstash-6.7.0/config/certs/instance/instance.crt"
  }
}

参考：https://www.elastic.co/guide/en/elasticsearch/reference/6.7/configuring-tls.html#node-certificates
https://discuss.elastic.co/t/certificates-and-keys-for-kibana-and-logstash-with-x-pack/150390
关注证书的制作方法
关注证书的制作方法

3 “Could not index event to Elasticsearch”
报错信息

][WARN ][logstash.outputs.elasticsearch] Could not index event to Elasticsearch. {:status=>404, :action=>["index", {:_id=>nil, :_index=>"logstash-2017.07.05", :_type=>"syslog", :_routing=>nil}, 2017-07-05T03:50:10.577Z 10.91.142.103 <179>Jul 5 09:20:10 10.91.126.1 TMNX: 45006214 Base PORT-MINOR-etherAlarmSet-2017 [Port 5/2/4]: Alarm Remote Fault Set], :response=>{"index"=>{"_index"=>"logstash-2017.07.05", "_type"=>"syslog", "_id"=>nil, "status"=>404, "error"=>{"type"=>"index_not_found_exception", "reason"=>"no such index and [action.auto_create_index] ([.security,.monitoring*,.watches,.triggered_watches,.watcher-history*,.ml*]) doesn't match", "index_uuid"=>"na", "index"=>"logstash-2017.07.05"}}}}

解决：
去掉elasticsearch.yml中的限制即可

action.auto_create_index: ".security*,.monitoring*,.watches,.triggered_watches,.watcher-history*,.ml*"

参考：https://discuss.elastic.co/t/index-404-error-in-logstash/91830

4. es搭建集群时，需要集群正常，否则会报错如下

 [wp-node1] not enough master nodes discovered during pinging (found [[Candidate{node={wp-node1}{MxvGfK7lR6iIwww_frh_BOzcg}{yRKFsh2PT3aEzjI6IwwwG7n1Q}{192.168.100.203}{192.168.100.203:9300}{ml.machine_memory=16603000832, xpack.installed=true, ml.max_open_jobs=20, 

 ][INFO ][o.e.d.z.ZenDiscovery     ] [wp-node3] failed to send join request to master [{wp-node1}{MxvGfK7lR6iI_frh_BOzcg}{7Bc5VHF9QzCs-G2ERoAYaQ}{192.168.100.203}{192.168.100.203:9300}{ml.machine_memory=16603000832, 

[wp-node2] failed to send join request to master [{wp-node1}{MxvGfK7lR6iI_frh_BOzcg}{M_VIMWHFSPmZT9n2D2_fZQ}{192.168.100.203}{192.168.100.203:9300}{ml.machine_memory=16603000832, ml.max_open_jobs=20, xpack.installed=true, ml.enabled=true}], reason [RemoteTransportException[[wp-node1][192.168.100.203:9300][internal:discovery/zen/join]]; nested: IllegalArgumentException[can't add node {wp-node2}{MxvGfK7lR6iI_frh_BOzcg}{gO2EuagySwq9d5AfDTK6qg}{192.168.1.202}{192.168.1.202:9300}{ml.machine_memory=12411887616, ml.max_open_jobs=20, xpack.installed=true, ml.enabled=true}, found existing node } with the same id but is a different node instance]; ]

5 此时curl访问es集群需要使用证书访问如

curl https://192.168.100.203:9200/_xpack/security/_authenticate?pretty \
--key client.key --cert client.cer --cacert client-ca.cer -k -v

访问logstash

curl -v --cacert ca.crt https://192.168.100.203:5044

6.logtstash日志报错如下：

[2019-07-03T18:18:29,035][ERROR][logstash.outputs.elasticsearch] Attempted to send a bulk request to elasticsearch' but Elasticsearch appears to be unreachable or down! {:error_message=>"Elasticsearch Unreachable: [https://elastic:xxxxxx@192.168.100.203:9200/][Manticore::ConnectTimeout] Read timed out", :class=>"LogStash::Outputs::ElasticSearch::HttpClient::Pool::HostUnreachableError", :will_retry_in_seconds=>2}
[2019-07-03T18:18:29,035][ERROR][logstash.outputs.elasticsearch] Attempted to send a bulk request to elasticsearch' but Elasticsearch appears to be unreachable or down! {:error_message=>"Elasticsearch Unreachable: [https://elastic:xxxxxx@192.168.100.203:9200/][Manticore::ConnectTimeout] Read timed out", :class=>"LogStash::Outputs::ElasticSearch::HttpClient::Pool::HostUnreachableError", :will_retry_in_seconds=>2}
[2019-07-03T18:18:29,515][WARN ][logstash.outputs.elasticsearch] Restored connection to ES instance {:url=>"https://elastic:xxxxxx@192.168.100.203:9200/"}

原因：配置是对的，因为有些数据是可以发到elasticsearch上的，如果发的数据过多的话会发现这个问题

另：参考

http://www.51niux.com/?id=210
https://www.elastic.co/guide/en/beats/filebeat/current/configuring-ssl-logstash.html