一、概述
线上部署的k8s已经扛过了双11的洗礼，期间先是通过对网络和监控的优化顺利度过了双11并且表现良好。先简单介绍一下我们kubernetes的使用方式：
物理机系统：Ubuntu-16.04（kernel 升级到4.17）
kuberneets-version：1.13.2
网络组件：calico（采用的是BGP模式+bgp reflector）
kube-proxy：使用的是ipvs模式
监控：prometheus+grafana
日志： fluentd + ES
metrics： metrics-server
HPA：cpu + memory
告警：钉钉
CI/CD： gitlab-ci/gitlab-runner
应用管理工具：helm、chartmuseum（不建议直接使用helm，helm charts可读性很差，学习成本较高）
由于k8s、物理环境共存，需要打通通网络提供访问：kube-gateway
有的地方涉及到公司内部的东西不方便写出来，但是绝大部分在我之前的博客都有介绍，有兴趣的可以参考一下。
自己的反思：
开始的时候，k8s集群在线上跑了一段时间，但是我发现我对集群内部的变化没有办法把控的很清楚，比如某个pod被重新调度了、某个node节点上的imagegc失败了、某个hpa被触发了等等，而这些都是可以通过events拿到的，但是events并不是永久存储的，它包含了集群各种资源的状态变化，所以我们可以通过收集分析events来了解整个集群内部的变化，经过一番探索找到一个开源的eventrouter来收集events事件，经过一些改造使其符合我们的业务场景，更名为eventrouter-kafka（https://github.com/cuishuaigit/eventrouter-kafka）直接将修改配置直传kafka，而不是需要各种配置，感觉原版的配置有些繁琐不是很好用，而我们的日志也是走kafka队列的，减轻ES的写压力。现在的events收集流程：
eventrouter---->kafka---->logstash(过滤、解析)----->ES------elastalert---->钉钉
经过添加上面的收集events使k8s集群又完善了一步。
二、简述流程
1、部署eventrouter
eventrouter是使用golang写的，可以根据自己的需求二次开发，部署很简单，参考：https://github.com/cuishuaigit/eventrouter-kafka。这里就不细述了。
2、kafka集群
参考：https://github.com/cuishuaigit/k8s-kafka
3、logstash
现在相应版本的logstash，下载地址：https://www.elastic.co/guide/en/logstash/6.5/installing-logstash.html
然后进行配置，这里贴一下我的测试配置：
input{
kafka{
bootstrap_servers => ["kafka-0.kafka-svc.kafka.svc.cluster.local:9092,kafka-1.kafka-svc.kafka.svc.cluster.local:9092,kafka-2.kafka-svc.kafka.svc.cluster.local:9092"]
client_id => "eventrouter-prod"
#auto_offset_reset => "latest"
group_id => "eventrouter"
consumer_threads =>
#decorate_events => true
id => "eventrouter"
topics => ["eventrouter"]
}
}
filter {
if [message] =~ 'DNSConfigForming' {
drop { }
}
json {
source => "message"
}
mutate {
remove_field => [ "message","old_event" ]
}
}
output{
elasticsearch {
hosts => "10.4.9.28:9200"
index => "eventrouter-%{+YYYY-MM-dd}"
}
}
4、ES
version: ''
services:
elasticsearch:
image: docker.elastic.co/elasticsearch/elasticsearch:6.5.
container_name: elasticsearch
environment:
- cluster.name=docker-cluster
- bootstrap.memory_lock=true
- "ES_JAVA_OPTS=-Xms4096m -Xmx4096m" ulimits:
memlock:
soft: -
hard: -
volumes:
- /data/es1:/usr/share/elasticsearch/data
- /data/backups:/usr/share/elasticsearch/backups
- /data/longterm_backups:/usr/share/elasticsearch/longterm_backups
- ./config/jvm.options:/usr/share/elasticsearch/config/jvm.options
ports:
- "9200:9200"
networks:
- esnet
# elasticsearch2:
# image: docker.elastic.co/elasticsearch/elasticsearch:6.5.
# container_name: elasticsearch2
# environment:
# - cluster.name=docker-cluster
# - bootstrap.memory_lock=true
# - "ES_JAVA_OPTS=-Xms512m -Xmx512m"
# - "discovery.zen.ping.unicast.hosts=elasticsearch"
# ulimits:
# memlock:
# soft: -
# hard: -
# volumes:
# - /data/es2:/usr/share/elasticsearch/data
# networks:
# - esnet
kibana:
image: docker.elastic.co/kibana/kibana:6.5.
container_name: kibana
environment:
SERVER_NAME: kibana
SERVER_HOST: "0.0.0.0"
ELASTICSEARCH_URL: http://elasticsearch:9200
XPACK_MONITORING_UI_CONATINER_ELASTICSEARCH_ENABLED: "true"
volumes:
- /data/plugin:/usr/share/kibana/plugin
- /tmp/:/etc/archives
ports:
- "5601:5601"
networks:
- esnet
depends_on:
- elasticsearch
networks:
esnet:
driver: bridge
cat config/jvm.properties
## JVM configuration ################################################################
## IMPORTANT: JVM heap size
################################################################
##
## You should always set the min and max JVM heap
## size to the same value. For example, to set
## the heap to GB, set:
##
## -Xms4g
## -Xmx4g
##
## See https://www.elastic.co/guide/en/elasticsearch/reference/current/heap-size.html
## for more information
##
################################################################ # Xms represents the initial size of total heap space
# Xmx represents the maximum size of total heap space -Xms2g
-Xmx2g ################################################################
## Expert settings
################################################################
##
## All settings below this section are considered
## expert settings. Don't tamper with them unless
## you understand what you are doing
##
################################################################ ## GC configuration
-XX:+UseConcMarkSweepGC
-XX:CMSInitiatingOccupancyFraction=
-XX:+UseCMSInitiatingOccupancyOnly ## G1GC Configuration
# NOTE: G1GC is only supported on JDK version or later.
# To use G1GC uncomment the lines below.
# -:-XX:-UseConcMarkSweepGC
# -:-XX:-UseCMSInitiatingOccupancyOnly
# -:-XX:+UseG1GC
# -:-XX:InitiatingHeapOccupancyPercent= ## optimizations # pre-touch memory pages used by the JVM during initialization
-XX:+AlwaysPreTouch ## basic # explicitly set the stack size
-Xss1m # set to headless, just in case
-Djava.awt.headless=true # ensure UTF- encoding by default (e.g. filenames)
-Dfile.encoding=UTF- # use our provided JNA always versus the system one
-Djna.nosys=true # turn off a JDK optimization that throws away stack traces for common
# exceptions because stack traces are important for debugging
-XX:-OmitStackTraceInFastThrow # flags to configure Netty
-Dio.netty.noUnsafe=true
-Dio.netty.noKeySetOptimization=true
-Dio.netty.recycler.maxCapacityPerThread= # log4j
-Dlog4j.shutdownHookEnabled=false
-Dlog4j2.disable.jmx=true -Djava.io.tmpdir=${ES_TMPDIR} ## heap dumps # generate a heap dump when an allocation from the Java heap fails
# heap dumps are created in the working directory of the JVM
-XX:+HeapDumpOnOutOfMemoryError # specify an alternative path for heap dumps; ensure the directory exists and
# has sufficient space
-XX:HeapDumpPath=data # specify an alternative path for JVM fatal error logs
-XX:ErrorFile=logs/hs_err_pid%p.log ## JDK GC logging :-XX:+PrintGCDetails
:-XX:+PrintGCDateStamps
:-XX:+PrintTenuringDistribution
:-XX:+PrintGCApplicationStoppedTime
:-Xloggc:logs/gc.log
:-XX:+UseGCLogFileRotation
:-XX:NumberOfGCLogFiles=
:-XX:GCLogFileSize=64m # JDK + GC logging
-:-Xlog:gc*,gc+age=trace,safepoint:file=logs/gc.log:utctime,pid,tags:filecount=,filesize=64m
# due to internationalization enhancements in JDK Elasticsearch need to set the provider to COMPAT otherwise
# time/date parsing will break in an incompatible way for some date patterns and locals
-:-Djava.locale.providers=COMPAT # temporary workaround for C2 bug with JDK on hardware with AVX-
-:-XX:UseAVX=
5、elastalert
部署参考https://github.com/Yelp/elastalert.git
使用：
mkdir /etc/elastalert
将clone的elastalert目录下面的config.yaml.example拷贝到上面创建的目录里面：
cpoy elastalert/config.yaml.example /etc/elastalert/config.yaml
只需要修改：
rules_folder、es_host、es_port，如果设置了用户密码，还需要修改。
创建rules
mkdir /etc/elastalert/rules
6、钉钉
创建机器人参考我其他的博客，获取token，下载钉钉plugin， https://github.com/xuyaoqiang/elastalert-dingtalk-plugin
将elastalert_modules拷贝到/etc/elastalert目录下面
cp -r elastalert-dingtalk-plugin/elastalert_modules /etc/elastalert/elastalert
rules example
# Alert when the rate of events exceeds a threshold # (Optional)
# Elasticsearch host
es_host: 10.2.9.28 # (Optional)
# Elasticsearch port
es_port: # (OptionaL) Connect with SSL to Elasticsearch
#use_ssl: True # (Optional) basic-auth username and password for Elasticsearch
#es_username: someusername
#es_password: somepassword # (Required)
# Rule name, must be unique
name: Other event frequency rule # (Required)
# Type of alert.
# the frequency rule type alerts when num_events events occur with timeframe time
type: frequency # (Required)
# Index to search, wildcard supported
index: eventrouter-* # (Required, frequency specific)
# Alert when this many documents matching the query occur within a timeframe
num_events: # (Required, frequency specific)
# num_events must occur within this amount of time to trigger an alert
timeframe:
#hours:
minutes:
# (Required)
# A list of Elasticsearch filters used for find events
# These filters are joined with AND and nested in a filtered query
# For more info: http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/query-dsl.html
filter:
#- term:
# some_field: "some_value"
- query:
query_string:
query: "event.type: Warning NOT event.involvedObject.kind: Node"
# (Required)
# The alert is use when a match is found #smtp_host: smtp.exmail.qq.com
#smtp_port:
#smtp_auth_file: /etc/elastalert/smtp_auth_file.yaml
#email_reply_to: ci@qq.com
#from_addr: ci@qq.com
realert:
minutes:
exponential_realert:
hours: alert:
#- "email"
- "elastalert_modules.dingtalk_alert.DingTalkAlerter"
dingtalk_webhook: "https://oapi.dingtalk.com/robot/send?access_token=47194e6904c6e3133a9080980984444c8e5d7745e1f76c12cefa99c8c8ac718dd88d4c"
dingtalk_msgtype: "text" alert_text_type: alert_text_only
alert_text: "
====elastalert message====\n EventTime>>: {}\n Event_involvedObject_name>>: {}\n Event_involvedObject_kind>>: {}\n Event_involvedObject_namespace>>: {}\n Message>>: {}\n Event_reason>>: {}\n verb>>: {}
" alert_text_args:
- "@timestamp"
- event.involvedObject.name
- event.source.component
- event.involvedObject.namespace
- event.message
- event.reason
- verb
# (required, email specific)
# a list of email addresses to send alerts to
#email:
#- "ci@qq.com"
自己定制的告警消息格式：
alert:
#- "email"
- "elastalert_modules.dingtalk_alert.DingTalkAlerter"
dingtalk_webhook: "https://oapi.dingtalk.com/robot/send?access_token=47194e6904c6e3133a9080980984444c8e5d7745e1f76c12cefa99c8c8ac718dd88d4c"
dingtalk_msgtype: "text" alert_text_type: alert_text_only
alert_text: "
====elastalert message====\n EventTime>>: {}\n Event_involvedObject_name>>: {}\n Event_involvedObject_kind>>: {}\n Event_involvedObject_namespace>>: {}\n Message>>: {}\n Event_reason>>: {}\n verb>>: {}
" alert_text_args:
- "@timestamp"
- event.involvedObject.name
- event.source.component
- event.involvedObject.namespace
- event.message
- event.reason
- verb
详细信息参考官网：https://elastalert.readthedocs.io/en/latest/recipes/writing_filters.html#writingfilters