第一部分 概括
ELK是集分布式数据存储、可视化查询和日志解析于一体的日志分析平台。ELK=elasticsearch+Logstash+kibana,三者各司其职,相互配合,共同完成日志的数据处理工作。ELK各组件的主要功能如下:
- elasticsearch,数据存储以及全文检索;
- logstash,日志加工、“搬运工”;
- kibana:数据可视化展示和运维管理。
我们在搭建平台时,还借助了filebeat插件。Filebeat是本地文件的日志数据采集器,可监控日志目录或特定日志文件(tail file),并可将数据转发给Elasticsearch或Logstatsh等。
本案例的实践,主要通过ELK收集、管理、检索mysql实例的慢查询日志和错误日志。
简单的数据流程图如下:
第二部分 elasticsearch
2.1 ES特点和优势
- 分布式实时文件存储,可将每一个字段存入索引,使其可以被检索到。
- 实时分析的分布式搜索引擎。分布式:索引分拆成多个分片,每个分片可有零个或多个副本;负载再平衡和路由在大多数情况下自动完成。
- 可以扩展到上百台服务器,处理PB级别的结构化或非结构化数据。也可以运行在单台PC上。
- 支持插件机制,分词插件、同步插件、Hadoop插件、可视化插件等。
2.2 ES主要概念
ES数据库 | MySQL数据库 |
Index | Database |
Tpye[在7.0之后type为固定值_doc] | Table |
Document | Row |
Field | Column |
Mapping | Schema |
Everything is indexed | Index |
Query DSL[Descriptor structure language] | SQL |
GET http://... | Select * from table … |
PUT http://... | Update table set … |
- 关系型数据库中的数据库(DataBase),等价于ES中的索引(Index);
- 一个关系型数据库有N张表(Table),等价于1个索引Index下面有N多类型(Type);
- 一个数据库表(Table)下的数据由多行(ROW)多列(column,属性)组成,等价于1个Type由多个文档(Document)和多Field组成;
- 在关系型数据库里,schema定义了表、每个表的字段,还有表和字段之间的关系。 与之对应的,在ES中:Mapping定义索引下的Type的字段处理规则,即索引如何建立、索引类型、是否保存原始索引JSON文档、是否压缩原始JSON文档、是否需要分词处理、如何进行分词处理等;
- 关系型数据库中的增insert、删delete、改update、查search操作等价于ES中的增PUT/POST、删Delete、改_update、查GET.
2.3 执行权限问题
报错提示
[usernimei@testes01 bin]$ Exception in thread "main" org.elasticsearch.bootstrap.BootstrapException: java.nio.file.AccessDeniedException: /data/elasticsearch/elasticsearch-7.4./config/elasticsearch.keystore
Likely root cause: java.nio.file.AccessDeniedException: /data/elasticsearch/elasticsearch-7.4./config/elasticsearch.keystore
at java.base/sun.nio.fs.UnixException.translateToIOException(UnixException.java:)
at java.base/sun.nio.fs.UnixException.rethrowAsIOException(UnixException.java:)
at java.base/sun.nio.fs.UnixException.rethrowAsIOException(UnixException.java:)
at java.base/sun.nio.fs.UnixFileSystemProvider.newByteChannel(UnixFileSystemProvider.java:)
at java.base/java.nio.file.Files.newByteChannel(Files.java:)
at java.base/java.nio.file.Files.newByteChannel(Files.java:)
at org.apache.lucene.store.SimpleFSDirectory.openInput(SimpleFSDirectory.java:)
at org.elasticsearch.common.settings.KeyStoreWrapper.load(KeyStoreWrapper.java:)
at org.elasticsearch.bootstrap.Bootstrap.loadSecureSettings(Bootstrap.java:)
at org.elasticsearch.bootstrap.Bootstrap.init(Bootstrap.java:)
at org.elasticsearch.bootstrap.Elasticsearch.init(Elasticsearch.java:)
at org.elasticsearch.bootstrap.Elasticsearch.execute(Elasticsearch.java:)
at org.elasticsearch.cli.EnvironmentAwareCommand.execute(EnvironmentAwareCommand.java:)
at org.elasticsearch.cli.Command.mainWithoutErrorHandling(Command.java:)
at org.elasticsearch.cli.Command.main(Command.java:)
at org.elasticsearch.bootstrap.Elasticsearch.main(Elasticsearch.java:)
at org.elasticsearch.bootstrap.Elasticsearch.main(Elasticsearch.java:)
Refer to the log for complete error details
问题分析
第一次误用了root账号启动,此时路径下的elasticsearch.keystore 权限属于了root
-rw-rw---- root root Mar : elasticsearch.keystore
解决方案--切换到root用户修改文件elasticsearch.keystore权限
调整到es用户下,即
chown -R es用户:es用户组 elasticsearch.keystore
问题2.4 maximum shards open 问题
根据官方解释,从Elasticsearch v7.0.0 开始,集群中的每个节点默认限制 1000 个shard,如果你的es集群有3个数据节点,那么最多 3000 shards。这里我们是只有一台es。所以只有1000。
[--11T11::,][WARN ][logstash.outputs.elasticsearch][main] Marking url as dead. Last error: [LogStash::Outputs::ElasticSearch::HttpClient::Pool::HostUnreachableError] Elasticsearch Unreachable: [http://qqelastic:[email protected]:55944/][Manticore::SocketTimeout] Read timed out {:url=>http://qqelastic:[email protected]:55944/, :error_message=>"Elasticsearch Unreachable: [http://qqelastic:[email protected]:55944/][Manticore::SocketTimeout] Read timed out", :error_class=>"LogStash::Outputs::ElasticSearch::HttpClient::Pool::HostUnreachableError"}
[--11T11::,][ERROR][logstash.outputs.elasticsearch][main] Attempted to send a bulk request to elasticsearch' but Elasticsearch appears to be unreachable or down! {:error_message=>"Elasticsearch Unreachable: [http://qqelastic:[email protected]:55944/][Manticore::SocketTimeout] Read timed out", :class=>"LogStash::Outputs::ElasticSearch::HttpClient::Pool::HostUnreachableError", :will_retry_in_seconds=>2}
[--11T11::,][WARN ][logstash.outputs.elasticsearch][main] Restored connection to ES instance {:url=>"http://qqelastic:[email protected]:55944/"}
[--11T11::,][WARN ][logstash.outputs.elasticsearch][main] Could not index event to Elasticsearch. {:status=>, :action=>["index", {:_id=>nil, :_index=>"mysql-error-testqq-2019.05.11", :routing=>nil, :_type=>"_doc"}, #<LogStash::Event:0x65416fce>], :response=>{"index"=>{"_index"=>"mysql-error-qqweixin-2020.05.11", "_type"=>"_doc", "_id"=>nil, "status"=>, "error"=>{"type"=>"validation_exception", "reason"=>"Validation Failed: 1: this action would add [2] total shards, but this cluster currently has [1000]/[1000] maximum shards open;"}}}}
可以用Kibana来设置
主要命令:
PUT /_cluster/settings
{
"transient": {
"cluster": {
"max_shards_per_node":
}
}
}
操作截图如下:
注意事项:
建议设置后重启下lostash服务
第三部分 Filebeat
问题3.1 不读取log文件中的数据
--23T19::41.772+ INFO [monitoring] log/log.go: Non-zero metrics in the last 30s
{"monitoring": {"metrics": {"beat":{"cpu":{"system":{"ticks":,"time":{"ms":}},"total":{"ticks":,"time":{"ms":},"value":},"user":{"ticks":,"time":{"ms":}}},"handles":{"limit":{"hard":,"soft":},"open":},"info":{"ephemeral_id":"a4c61321-ad02-2c64-9624-49fe4356a4e9","uptime":{"ms":}},"memstats":{"gc_next":,"memory_alloc":,"memory_total":},"runtime":{"goroutines":}},"filebeat":{"harvester":{"open_files":,"running":}},"libbeat":{"config":{"module":{"running":}},"pipeline":{"clients":,"events":{"active":}}},"registrar":{"states":{"current":}},"system":{"load":{"":,"":0.05,"":0.01,"norm":{"":,"":0.0125,"":0.0025}}}}}}
修改 filebeat.yml 的配置参数
问题3.2 多个服务进程
--27T20::22.985+ ERROR logstash/async.go: Failed to publish events caused by: write tcp [::]:->[::]:: write: connection reset by peer
--27T20::23.985+ INFO [monitoring] log/log.go: Non-zero metrics in the last 30s {"monitoring": {"metrics": {"beat":{"cpu":{"system":{"ticks":,"time":{"ms":}},"total":{"ticks":,"time":{"ms":},"value":},"user":{"ticks":,"time":{"ms":}}},"handles":{"limit":{"hard":,"soft":},"open":},"info":{"ephemeral_id":"a02ed909-a7a0-49ee-aff9-5fdab26ecf70","uptime":{"ms":}},"memstats":{"gc_next":,"memory_alloc":,"memory_total":,"rss":},"runtime":{"goroutines":}},"filebeat":{"events":{"active":,"added":},"harvester":{"open_files":,"running":}},"libbeat":{"config":{"module":{"running":}},"output":{"events":{"batches":,"failed":,"total":},"write":{"errors":}},"pipeline":{"clients":,"events":{"active":,"published":,"total":}}},"registrar":{"states":{"current":}},"system":{"load":{"":0.05,"":0.11,"":0.06,"norm":{"":0.0063,"":0.0138,"":0.0075}}}}}}
--27T20::24.575+ ERROR pipeline/output.go: Failed to publish events: write tcp [::]:->[::]:: write: connection reset by peer
原因是同时有多个logstash进程在运行,关闭重启
问题3.3 将Filebeat 配置成服务进行管理
filebeat 服务所在路径:
/etc/systemd/system
编辑filebeat.service文件
[Unit]
Description=filebeat.service
[Service]
User=root
ExecStart=/data/filebeat/filebeat-7.4.-linux-x86_64/filebeat -e -c /data/filebeat/filebeat-7.4.-linux-x86_64/filebeat.yml
[Install]
WantedBy=multi-user.target
管理服务的相关命令
systemctl start filebeat #启动filebeat服务
systemctl enable filebeat #设置开机自启动
systemctl disable filebeat #停止开机自启动
systemctl status filebeat #查看服务当前状态
systemctl restart filebeat #重新启动服务
systemctl list-units --type=service #查看所有已启动的服务
问题3.4 Filebeat 服务启动报错
注意错误
Exiting: error loading config file: yaml: line : did not find expected key
主要问题是:filebeat.yml 文件中的格式有破坏,应特别注意修改和新增的地方,对照前后文,验证格式是否有变化。
问题 3.5 Linux 版本过低,无法以systemctl管理filebeat服务
此时我们可以以service来管理,在目录init.d下创建一个filebeat.service文件。主要脚本如下:
#!/bin/bash
agent="/data/filebeat/filebeat-7.4.2-linux-x86_64/filebeat"
args="-e -c /data/filebeat/filebeat-7.4.2-linux-x86_64/filebeat.yml"
start() {
pid=`ps -ef |grep /data/filebeat/filebeat-7.4.-linux-x86_64/filebeat |grep -v grep |awk '{print $2}'`
if [ ! "$pid" ];then
echo "Starting filebeat: "
nohup $agent $args >/dev/null >& &
if [ $? == '' ];then
echo "start filebeat ok"
else
echo "start filebeat failed"
fi
else
echo "filebeat is still running!"
exit
fi
}
stop() {
echo -n $"Stopping filebeat: "
pid=`ps -ef |grep /data/filebeat/filebeat-7.4.-linux-x86_64/filebeat |grep -v grep |awk '{print $2}'`
if [ ! "$pid" ];then
echo "filebeat is not running"
else
kill $pid
echo "stop filebeat ok"
fi
}
restart()
{
stop
start
}
status(){
pid=`ps -ef |grep /data/filebeat/filebeat-7.4.-linux-x86_64/filebeat |grep -v grep |awk '{print $2}'`
if [ ! "$pid" ];then
echo "filebeat is not running"
else
echo "filebeat is running"
fi
}
case "$1" in
start)
start
;;
stop)
stop
;;
restart)
restart
;;
status)
status
;;
*)
echo $"Usage: $0 {start|stop|restart|status}"
exit
esac
注意事项
1.文件授予执行权限
chmod filebeat.service
2.设置开机自启动
chkconfig --add filebeat.service
上面的服务添加自启动时,会报错
解决方案 在 service file的开头添加以下 两行
即修改完善后的代码如下:
#!/bin/bash # chkconfig: 2345 10 80
# description: filebeat is a tool for colletct log data agent="/data/filebeat/filebeat-7.4.2-linux-x86_64/filebeat"
args="-e -c /data/filebeat/filebeat-7.4.2-linux-x86_64/filebeat.yml"
start() {
pid=`ps -ef |grep /data/filebeat/filebeat-7.4.-linux-x86_64/filebeat |grep -v grep |awk '{print $2}'`
if [ ! "$pid" ];then
echo "Starting filebeat: "
nohup $agent $args >/dev/null?>& &
if [ $? == '' ];then
echo "start filebeat ok"
else
echo "start filebeat failed"
fi
else
echo "filebeat is still running!"
exit
fi
}
stop() {
echo -n $"Stopping filebeat: "
pid=`ps -ef |grep /data/filebeat/filebeat-7.4.-linux-x86_64/filebeat |grep -v grep |awk '{print $2}'`
if [ ! "$pid" ];then
echo "filebeat is not running"
else
kill $pid
echo "stop filebeat ok"
fi
}
restart()
{
stop
start
}
status(){
pid=`ps -ef |grep /data/filebeat/filebeat-7.4.-linux-x86_64/filebeat |grep -v grep |awk '{print $2}'`
if [ ! "$pid" ];then
echo "filebeat is not running"
else
echo "filebeat is running"
fi
}
case "$1" in
start)
start
;;
stop)
stop
;;
restart)
restart
;;
status)
status
;;
*)
echo $"Usage: $0 {start|stop|restart|status}"
exit
esac
第四部分 Logstash
问题 4.1 服务化配置
logstash最常见的运行方式即命令行运行./bin/logstash -f logstash.conf启动,结束命令是ctrl+c。这种方式的优点在于运行方便,缺点是不便于管理,同时如果遇到服务器重启,则维护成本会更高一些,如果在生产环境运行logstash推荐使用服务的方式。以服务的方式启动logstash,同时借助systemctl的特性实现开机自启动。
(1)安装目录下的config中的startup.options需要修改
修改主要项:
1.服务默认启动用户和用户组为logstash;可以修改为root;
2. LS_HOME 参数设置为 logstash的安装目录;例如:/data/logstash/logstash-7.6.0
3. LS_SETTINGS_DIR参数配置为含有logstash.yml的目录;例如:/data/logstash/logstash-7.6.0/config
4. LS_OPTS 参数项,添加 logstash.conf 指定项(-f参数);例如:LS_OPTS="--path.settings ${LS_SETTINGS_DIR} -f /data/logstash/logstash-7.6.0/config/logstash.conf"
(2)以root身份执行logstash命令创建服务
创建服务的命令
安装目录/bin/system-install
执行创建命令后,在/etc/systemd/system/目录中生成了logstash.service 文件
(3)logstash 服务的管理
设置服务自启动:systemctl enable logstash
启动服务:systemctl start logstash
停止服务:systemctl stop logstash
重启服务:systemctl restart logstash
查看服务状态:systemctl status logstash
问题 4.2 安装logstash服务需先安装jdk
报错提示如下:
通过查看jave版本,验证是否已安装
上图说明没有安装。则将安装包下载(或上传)至本地,执行安装
执行安装命令如下:
yum localinstall jdk-8u211-linux-x64.rpm
安装OK,执行验证
问题 4.3 Linux 版本过低,安装 logstash 服务失效
问题提示
查看Linux系统版本
原因: centos 6.5 不支持 systemctl 管理服务
解决方案
方案验证
相关命令
.启动命令
initctl start logstash
.查看状态
initctl status logstash
注意事项:
注意以下生成服务的命令还是要执行的
./system-install
否则提示错误
initctl: Unknown job: logstash
问题 4.4 配置文件中定义的index name 命名需小写
"Invalid index name [mysql-error-Test-2019.05.13], must be lowercase", "index_uuid"=>"_na_", "index"=>"mysql-error-Test-2019.05.13"}}}}
May :: hzvm1996 logstash[]: [--13T13::,][ERROR][logstash.outputs.elasticsearch][main] Could not index event to Elasticsearch. {:status=>, :action=>["index", {:_id=>nil, :_index=>"mysql-slow-Test-2020.05.13", :routing=>nil, :_type=>"_doc"}, #<LogStash::Event:0x1f0aedbc>], :response=>{"index"=>{"_index"=>"mysql-slow-Test-2019.05.13", "_type"=>"_doc", "_id"=>nil, "status"=>, "error"=>{"type"=>"invalid_index_name_exception", "reason"=>"Invalid index name [mysql-slow-Test-2019.05.13], must be lowercase", "index_uuid"=>"_na_", "index"=>"mysql-slow-Test-2019.05.13"}}}}
May :: hzvm1996 logstash[]: [--13T13::,][ERROR][logstash.outputs.elasticsearch][main] Could not index event to Elasticsearch. {:status=>, :action=>["index", {:_id=>nil, :_index=>"mysql-error-Test-2020.05.13", :routing=>nil, :_type=>"_doc"}, #<LogStash::Event:0x4bdce1db>], :response=>{"index"=>{"_index"=>"mysql-error-Test-2019.05.13", "_type"=>"_doc", "_id"=>nil, "status"=>, "error"=>{"type"=>"invalid_index_name_exception", "reason"=>"Invalid index name [mysql-error-Test-2019.05.13], must be lowercase", "index_uuid"=>"_na_", "index"=>"mysql-error-Test-2019.05.13"}}}}
第五部分 kibana
问题5.1 开启密码认证
[root@testkibaba bin]# ./kibana-plugin install x-pack
Plugin installation was unsuccessful due to error "Kibana now contains X-Pack by default, there is no longer any need to install it as it is already present.
说明:新版本的Elasticsearch和Kibana都已经支持自带支持x-pack了,不需要进行显式安装。老版本的需要进行安装。
问题5.2 应用启动报错
[root@testkibana bin]# ./kibana
报错
Kibana should not be run as root. Use --allow-root to continue.
添加个专门的账号
useradd qqweixinkibaba --添加账号
chown -R qqweixinkibaba:hzdbakibaba kibana-7.4.-linux-x86_64 --为新增账号赋予文档目录的权限
su qqweixinkibaba ---切换账号,让后再启动
问题5.3 登入kibana报错
{"statusCode":,"error":"Forbidden","message":"Forbidden"}
报错原因是:用kibana账号登录kibana报错,改为elastic用户就行了
问题5.4 多租户实现的问题
一个公司会有多个业务线,也可能会有多个研发小组,那么如何实现收集到的数据只对相应的团队开放呢?即实现只能看到自家的数据。一种思路就是搭建多个ELK,一个业务线一个ELK,但这个方法会导致资源浪费和增加运维工作量;另一种思路就是通过多租户来实现。
实现时,应注意以下问题:
要在 elastic 账号下,转到指定的空间(space)下,再设置 index pattern 。
先创建role(注意与space关联),最后创建user。
参考资料
1.https://www.jianshu.com/p/0a5acf831409 《ELK应用之Filebeat》
2.http://www.voidcn.com/article/p-nlietamt-zh.html 《filebeat 启动脚本》
3.https://www.bilibili.com/video/av68523257/?redirectFrom=h5 《ElasticTalk #22 Kibana 多租户介绍与实战》
4.https://www.cnblogs.com/shengyang17/p/10597841.html 《ES集群》
5.https://www.jianshu.com/p/54cdddf89989 《Logstash配置以服务方式运行》
6.https://www.elastic.co/guide/en/logstash/current/running-logstash.html#running-logstash-upstart 《Running Logstash as a Service on Debian or RPM》