ELK集群搭建/K8S集群日志采集

一、所需的基础环境

部署包
提取码：ddc2

1、JDK环境

对应版本对应支持的JDK环境也有所差异，但基本都支持JDK1.8
三台机器都需要执行

[root@node1 /]# mkdir /home/apps
[root@node1 /]# cd /home/apps
[root@node1 apps]# tar -zxf jdk-8u171-linux-x64.tar.gz
# 加入以下内容
[root@node1 apps]# vim /etc/profile.d/jdk.sh
#!/bin/bash
export JAVA_HOME=/home/apps/jdk1.8.0_171
export PATH=$JAVA_HOME/bin:$PATH:$JAVA_HOME/lib/tools.jar
export CLASSPATH=$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar
[root@node1 apps]# source /etc/profile

2、磁盘配置

parted /dev/vdb
	mklabel gpt
	mkpart primary 0% 100% 
	p
	toggle 1 lvm
yum -y install lvm2
pvcreate /dev/vdb1
vgcreate -s 16M vgone /dev/vdb1
lvcreate -L 4096G -n lvnew vgone
mkfs.xfs /dev/vgone/lvnew
# 创建并挂载存放数据的目录
mkdir -p /data
mount /dev/vgone/lvnew /data
echo "/dev/vgone/lvnew                          /jxpoc                  xfs     defaults        0 0" >> /etc/fstab

3、修改操作系统限制

# 修改操作系统单个进程可以打开的最大文件数
[root@node1 apps]#  vim /etc/security/limits.conf
*               soft    nofile          65535
*               hard    nofile          65535
*               hard    nproc           4096
*               soft    nproc           4096
[root@node1 apps]# vi /etc/sysctl.conf
vm.max_map_count=262144
[root@node1 apps]#  sysctl -p

二、部署ES

1、创建路径

# 网络下载ES部署包
[root@master ~]#  wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-7.12.0-linux-x86_64.tar.gz
# 创建用户
[root@master ~]# adduser es_user
[root@master ~]# passwd es_user
[root@master ~]# mkdir /usr/local/es/

# 192.168.75.121
[root@master ~]# tar -zxf elasticsearch-7.12.0-linux-x86_64.tar.gz
[root@master ~]# mv elasticsearch-7.12.0 /usr/local/es/elasticsearch-7.12.0-node-1
[root@master ~]# chmod u+x  /usr/local/es/elasticsearch-7.12.0-node-1/bin
[root@master ~]# chown -R es_user /usr/local/es/
[root@master ~]# mkdir -p /data/es/es-node-1/
[root@master ~]# chown -R es_user /data/es/es-node-1/
[root@master ~]# mkdir -p /data/log/es/es-node-1/
[root@master ~]# chown -R es_user /data/log/es/es-node-1/

# 192.168.75.120
[root@master ~]# tar -zxf elasticsearch-7.12.0-linux-x86_64.tar.gz
[root@master ~]# mv elasticsearch-7.12.0 /usr/local/es/elasticsearch-7.12.0-node-2
[root@master ~]# chmod u+x  /usr/local/es/elasticsearch-7.12.0-node-2/bin
[root@master ~]# chown -R es_user /usr/local/es/
[root@master ~]# mkdir -p /data/es/es-node-2/
[root@master ~]# chown -R es_user /data/es/es-node-2/
[root@master ~]# mkdir -p /data/log/es/es-node-2/
[root@master ~]# chown -R es_user /data/log/es/es-node-2/

# 192.168.65.80
[root@master ~]# tar -zxf elasticsearch-7.12.0-linux-x86_64.tar.gz
[root@master ~]# mv elasticsearch-7.12.0 /usr/local/es/elasticsearch-7.12.0-node-3
[root@master ~]# chmod u+x  /usr/local/es/elasticsearch-7.12.0-node-3/bin
[root@master ~]# chown -R es_user /usr/local/es/
[root@master ~]# mkdir -p /data/es/es-node-3/
[root@master ~]# chown -R es_user /data/es/es-node-3/
[root@master ~]# mkdir -p /data/log/es/es-node-3/
[root@master ~]# chown -R es_user /data/log/es/es-node-3/

2、修改配置

192.168.75.121-node-1

# 修改以下内容
[root@node1 es]# vim elasticsearch-7.12.0-node-1/config/elasticsearch.yml
# 集群名称
cluster.name: my-application
# 集群节点名称
node.name: node-1
# 因为这台机器要装docker环境，所以需要将IP写死
network.host: 192.168.75.121
# 默认ES节点端口9200，如果是伪集群（在一台服务器上搭建集群），需要修改。
http.port: 9200
path.data: /data/es/es-node-1/
path.logs: /data/log/es/es-node-1/
# 发现其他节点主机配置
discovery.seed_hosts: ["192.168.75.121:9300","192.168.75.120:9300","192.168.65.80:9300"]
# 哪些节点可以被选举为主节点配置
cluster.initial_master_nodes: ["node-1", "node-2","node-3"]
# 跨域访问
http.cors.enabled: true
http.cors.allow-origin: "*"

192.168.75.121-node-1

[root@node1 es]# vim elasticsearch-7.12.0-node-2/config/elasticsearch.yml
cluster.name: my-application

node.name: node-2

network.host: 0.0.0.0
http.port: 9200

path.data: /data/es/es-node-2/

path.logs: /data/log/es/es-node-2/

discovery.seed_hosts: ["192.168.75.121:9300","192.168.75.120:9300","192.168.65.80:9300"]

cluster.initial_master_nodes: ["node-1", "node-2","node-3"]
# 跨域访问
http.cors.enabled: true
http.cors.allow-origin: "*"

192.168.75.121-node-1

[root@node1 es]# vim elasticsearch-7.12.0-node-3/config/elasticsearch.yml
cluster.name: my-application

node.name: node-3

network.host: 0.0.0.0
http.port: 9200

path.data: /data/es/es-node-3/

path.logs: /data/log/es/es-node-3/

discovery.seed_hosts: ["192.168.75.121:9300","192.168.75.120:9300","192.168.65.80:9300"]

cluster.initial_master_nodes: ["node-1", "node-2","node-3"]
# 跨域访问
http.cors.enabled: true
http.cors.allow-origin: "*"

3、启动

三台机器都执行相同操作

[root@node1 es]# cd elasticsearch-7.12.0-node-1
[root@node1 elasticsearch-7.12.0-node-1]# su es_user
[es_user@node1 elasticsearch-7.12.0-node-1]$ ./bin/elasticsearch -d  # -d后台启动，先不要加这个参数，看看输出日志是否有报错

分别访问三台es

三、安装es-head

docker pull mobz/elasticsearch-head:5
docker run -itd --name=es-head -p 9100:9100 mobz/elasticsearch-head:5

三、安装kibana

# 下载
curl -L -O https://artifacts.elastic.co/downloads/kibana/kibana-7.12.0-linux-x86_64.tar.gz
tar xzvf kibana-7.12.0-linux-x86_64.tar.gz
cd kibana-7.12.0-linux-x86_64/

# 修改配置文件
vim config/kibana.yml
# Kibana 默认监听端口5601,如果需要改变就修改这个配置
server.port: 5601
# Kibana 部署服务器IP,如果是单网卡配置0.0.0.0即可,如果是多网卡需要配置IP
server.host: "0.0.0.0"
# 配置服务器的名称
server.name: my-kibana
# 配置ES的集群节点地址
elasticsearch.hosts: ["http://192.168.75.121:9200","http://192.168.75.120:9200","http://192.168.65.80:9200"]
# 创建一个kibana索引
# kibana.index: ".kibana"
logging.dest: /var/log/kibana/kibana.log
i18n.locale: "zh-CN"

# 运行
nohup ./bin/kibana --alow-root &

四、安装filebeat

# 下载部署包
curl -L -O https://artifacts.elastic.co/downloads/beats/filebeat/filebeat-7.12.0-linux-x86_64.tar.gz
# 解压 
tar xzvf filebeat-7.12.0-linux-x86_64.tar.gz -C /usr/local/
mv /usr/local/filebeat-7.12.0-linux-x86_64.tar.gz /usr/local/filebeat
cd /usr/local/filebeat

编辑一个实例，这里以ngin为例

vim test.yml
# 获取的日志类型，这里以log为例
filebeat.inputs:
- type: log
  enabled: true
  paths:
    - /tmp/access.log
# 输出到屏幕中
output.console:
  pretty: true
# 前台启动
./filebeat -c ./test.yml

# 开放新的模块，以nginx为例
[root@master filebeat]# ./filebeat -c ./filebeat.yml modules enable nginx
Enabled nginx
# 修改nginx模块内容
[root@master filebeat]# vim modules.d/nginx.yml
# 添加日志地址
var.paths: ["/home/apps/access.og", "/home/apps/error.log"]
# 启动-e表示扩展，会自动去寻找modules开启的模块
[root@master filebeat]# ./filebeat -c ./filebeat.yml -e

ELK采集K8S日志

容器特性给日志采集带来的难度：

• K8s弹性伸缩性:导致不能预先确定采集的目标
• 容器隔离性:容器的文件系统与宿主机是隔离，导致
  日志采集器读取日志文件受阻

应用程序日志记录体现方式分为两类：

• 标准输出:输出到控制台，使用kubectl logs可以看到
• 日志文件:写到容器的文件系统的文件

# 查看一个nginx的Pod日志
kubectl run  nginx --image=nginx
kubectl exec -it nginx -- bash
root@nginx:/# cd /var/log/nginx/
root@nginx:/var/log/nginx/# ls -l

可以发现nginx的日志是输出到标准输出和标准错误输出里面，通过kubectl logs 可以查看nginx的日志

这个是从node节点的docker存储路径更具docker-ID找的


而kubectl logs查看日志的流程如下
kubectl logs命令—> apiserver —> kubelet —> docker api —> <container-id>-json.log
因为docker是把所有的日志都存放在宿主机目录中的(/var/lib/docker/containers，目录因不唯一，可使用docker info查看)，所以可以采集/vat/lib/docker/containers//-json.log(这里的*就是容器ID)

以上看到的就是标准输出的日志文件，另一种就是存在于容器内部的输出文件
如何去采集这两种日志

在k8s集群中配置ELK

创建动态供给(PV)
这里只是配置的单点的，没有配置集群
配置es的yaml

# 创建ops命名空间
apiVersion: v1
kind: Namespace
metadata:
  name: ops
  
---
# 配置es
apiVersion: apps/v1
kind: Deployment
metadata:
  name: elasticsearch
  # 配置的命名空间为ops
  namespace: ops
  labels:
    k8s-app: elasticsearch
spec:
  replicas: 1
  selector:
    matchLabels:
      k8s-app: elasticsearch
  template:
    metadata:
      labels:
        k8s-app: elasticsearch
    spec:
      containers:
      - image: elasticsearch:7.9.2
        name: elasticsearch
        resources:
          limits:
            cpu: 2
            memory: 3Gi
          requests:
            cpu: 0.5 
            memory: 500Mi
        env:
          # 传入预定义的一个值，设置为单节点部署
          - name: "discovery.type"
            value: "single-node"
          - name: ES_JAVA_OPTS
            value: "-Xms512m -Xmx2g" 
        ports:
        - containerPort: 9200
          name: db
          protocol: TCP
        volumeMounts:
        - name: elasticsearch-data
          mountPath: /usr/share/elasticsearch/data
      volumes:
      - name: elasticsearch-data
        persistentVolumeClaim:
          claimName: es-pvc

---

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: es-pvc
  namespace: ops
spec:
  storageClassName: "managed-nfs-storage"
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 10Gi

---

apiVersion: v1
kind: Service
metadata:
  name: elasticsearch
  namespace: ops
spec:
  ports:
  - port: 9200
    protocol: TCP
    targetPort: 9200
  selector:
    k8s-app: elasticsearch

kibana,yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: kibana
  namespace: ops
  labels:
    k8s-app: kibana
spec:
  replicas: 1
  selector:
    matchLabels:
      k8s-app: kibana
  template:
    metadata:
      labels:
        k8s-app: kibana
    spec:
      containers:
      - name: kibana
        image: kibana:7.9.2
        resources:
          limits:
            cpu: 2
            memory: 2Gi
          requests:
            cpu: 0.5 
            memory: 500Mi
        env:
          - name: ELASTICSEARCH_HOSTS
            value: http://elasticsearch.ops:9200
          - name: I18N_LOCALE
            value: zh-CN
        ports:
        - containerPort: 5601
          name: ui
          protocol: TCP

---
apiVersion: v1
kind: Service
metadata:
  name: kibana
  namespace: ops
spec:
  type: NodePort
  ports:
  - port: 5601
    protocol: TCP
    targetPort: ui
    nodePort: 30601
  selector:
    k8s-app: kibana

filebeat.yaml
在正常的部署中，filebeat是可以配置该日志是那个项目，那个业务的标记，但是在k8s中无法获取，为了解决这个方法，需要先部署一个k8s官方的yaml，来实现filebeat对k8s服务的发现能力.
可以根据日志从k8s api里得知这条日志属于谁的，那个Pod的，那个命名空间

---
apiVersion: v1
kind: ConfigMap
metadata:
  name: filebeat-config
  namespace: ops
  labels:
    k8s-app: filebeat
data:
  filebeat.yml: |-
    filebeat.config:
      inputs:
        # Mounted `filebeat-inputs` configmap:
        path: ${path.config}/inputs.d/*.yml
        # Reload inputs configs as they change:
        reload.enabled: false
      modules:
        path: ${path.config}/modules.d/*.yml
        # Reload module configs as they change:
        reload.enabled: false

    output.elasticsearch:
      hosts: ['elasticsearch.ops:9200']
---
apiVersion: v1
kind: ConfigMap
metadata:
  name: filebeat-inputs
  namespace: ops
  labels:
    k8s-app: filebeat
data:
  kubernetes.yml: |-
    - type: docker
      containers.ids:
      - "*"
      processors:
        - add_kubernetes_metadata:
            in_cluster: true
---
apiVersion: apps/v1 
kind: DaemonSet
metadata:
  name: filebeat
  namespace: ops
  labels:
    k8s-app: filebeat
spec:
  selector:
    matchLabels:
      k8s-app: filebeat
  template:
    metadata:
      labels:
        k8s-app: filebeat
    spec:
      serviceAccountName: filebeat
      terminationGracePeriodSeconds: 30
      containers:
      - name: filebeat
        image: elastic/filebeat:7.9.2
        args: [
          "-c", "/etc/filebeat.yml",
          "-e",
        ]
        securityContext:
          runAsUser: 0
          # If using Red Hat OpenShift uncomment this:
          #privileged: true
        resources:
          limits:
            memory: 200Mi
          requests:
            cpu: 100m
            memory: 100Mi
        volumeMounts:
        - name: config
          mountPath: /etc/filebeat.yml
          readOnly: true
          subPath: filebeat.yml
        - name: inputs
          mountPath: /usr/share/filebeat/inputs.d
          readOnly: true
        - name: data
          mountPath: /usr/share/filebeat/data
        - name: varlibdockercontainers
          mountPath: /var/lib/docker/containers
          readOnly: true
      volumes:
      - name: config
        configMap:
          defaultMode: 0600
          name: filebeat-config
      - name: varlibdockercontainers
        hostPath:
          path: /var/lib/docker/containers
      - name: inputs
        configMap:
          defaultMode: 0600
          name: filebeat-inputs
      # data folder stores a registry of read status for all files, so we don't send everything again on a Filebeat pod restart
      - name: data
        hostPath:
          path: /var/lib/filebeat-data
          type: DirectoryOrCreate
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: filebeat
subjects:
- kind: ServiceAccount
  name: filebeat
  namespace: ops
roleRef:
  kind: ClusterRole
  name: filebeat
  apiGroup: rbac.authorization.k8s.io
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: filebeat
  labels:
    k8s-app: filebeat
rules:
- apiGroups: [""] # "" indicates the core API group
  resources:
  - namespaces
  - pods
  verbs:
  - get
  - watch
  - list
---
apiVersion: v1
kind: ServiceAccount
metadata:
  name: filebeat
  namespace: ops
  labels:
    k8s-app: filebeat

K8S-ELK集群部署
官方文档