分类目录归档:未分类

Jetbrain的安装包带JBR与不带JBR的区别

以GoLang2019.3.4为例。

带JBR是指该JBR使用的版本为jetbrain公司自家的运行时库。因为该运行时库不是OpenJDK的认证版本,故可能在某个平台上如Centos7.5的平台上,某些时候可能会运行故障。
故应该使用不带JBR标识别的版本,该版本才真正的与openJDK认证的高兼容性的版本。
在centos环境下,建议使用不带JBR的版本

以下是其安装包解压后的差异对比。

gitlab的部署方式

version: '3'

services:
  gitlab:
    image: gitlab/gitlab-ce:11.9.12-ce.0
    restart: always
    container_name: gitlab
    hostname: gitlab.pointsmart.cn
    privileged: true
    user: root
    ports:
      - "80:80"
      - "443:443"
      - "222:22"
    environment:
      GITLAB_OMNIBUS_CONFIG: |
        gitlab_rails['time_zone'] = 'Asia/Shanghai'
        gitlab_rails['gitlab_shell_ssh_port'] = 222
        gitlab_rails['gitlab_shell_git_timeout'] = 800
    volumes:
      - /etc/localtime:/etc/localtime:ro
      - /data/gitlab/config:/etc/gitlab
      - /data/gitlab/logs:/var/log/gitlab
      - /data/gitlab/data:/var/opt/gitlab

通过进程ID查找容器

for i in  `docker ps -q`;do docker top $i; done
 for i in $(docker ps -q); do docker top $i; done
或
for i in  `docker ps |grep Up|awk '{print $1}'`;do docker top $i; done

nginx的stream模块转发配置

upstream xaio443 {
  server 172.16.0.103:443;
  server 172.16.0.104:443;
}

upstream xaio80 {
  server 172.16.0.103:80;
  server 172.16.0.104:80;
}

log_format proxy '$proxy_protocol_addr $remote_addr [$time_local] '
    '$protocol $status $bytes_sent $bytes_received '
    '$session_time "$upstream_addr" '
    '"$upstream_bytes_sent" "$upstream_bytes_received" "$upstream_connect_time"';

server {
  listen 443;
  ssl_preread on;
  proxy_connect_timeout 1s;
  proxy_timeout 3s;
  proxy_pass xaio443;


  access_log /data/logs/access443.log proxy;
  error_log /data/logs/error443.log info;
}

server {
  listen 80;
  proxy_connect_timeout 1s;
  proxy_timeout 3s;
  proxy_pass xaio80;


  access_log /data/logs/access80.log proxy;
  error_log /data/logs/error80.log info;
}

sh和api服务冲突的解决方案脚本

#!/bin/bash
path_current=`pwd`
path_script=$(cd "$(dirname "$0")"; pwd)

cd ${path_script}
echo "$(date)" >> ./check.txt
podnamspace="-n test"
shpods=$(kubectl describe pods "sh-" ${podnamspace}|grep "^Name:"|awk '{print $2}')
echo ${shpods}
sh_tm_max=0
for sh in ${shpods[@]}
do
  mytime=$(kubectl describe pod ${sh} ${podnamspace}|grep "^Start Time:")&& mytime=$(echo ${mytime:11}) && mytime=$(date -d "$mytime" +%s)
  elapse=$(expr ${sh_tm_max} - ${mytime})
  if [ $? == 0 ] && [ ${elapse} -lt 0 ]; then
     sh_tm_max=${mytime}
  fi
done
echo "sh pods start time: $sh_tm_max"
apipods=$(kubectl describe pods "api-" ${podnamspace}|grep "^Name:"|awk '{print $2}')
echo ${apipods}
for api in ${apipods[@]}
do
  mytime=$(kubectl describe pod ${api} ${podnamspace}|grep "^Start Time:")&& mytime=$(echo ${mytime:11}) && mytime=$(date -d "$mytime" +%s)
  elapse=$(expr ${sh_tm_max} - ${mytime})
  howMinute=$(expr $elapse / 60)
  output="${api} pod start time: $mytime, ${sh_tm_max} - ${mytime}=[${elapse}s, ${howMinute}m]"
  echo "$output" &&  echo "$output" >> ./check.txt
  if [ $? == 0 ] && [ $howMinute -ge 0 ]; then
     echo "delete this pod $api"
     kubectl delete pod $api ${podnamspace} && echo "success to delete $api" >> ./check.txt
  fi
done

kvm虚拟机常用操作

1、修改虚拟机配置如CPU核数或内存等

virsh edit aiodev

2、启动虚拟机

virsh start aiodev

3、停止虚拟机

virsh stop aiodev

4、显示所有运行的虚拟机

virsh list

5、快照创建

virsh snapshot-create-as aiodev ospure

6、显示快照列表

virsh snapshot-list aiodev

7、快照回滚

virsh snapshot-revert aiodev ospure

K8s启动命令

apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: ccc.cccc.cn 
  labels:
    app: web-api
spec:
  replicas: 3
  template:
    metadata:
      labels:
        app: api
    spec:
      restartPolicy: "Always"
      containers:
      - name: api-cccc-cn
        image: api:39
        imagePullPolicy: "IfNotPresent"
        command: ["/bin/sh","-c"," /usr/local/bin/composer dump-autoload ; crontab -e -u www-data ; crond ; /usr/bin/supervisord "]
        resources:
          limits:
            cpu: "1500m"
            memory: "1000Mi"
          requests:
            cpu: "200m"
            memory: "200Mi"
        ports:
        - containerPort: 80
          name: http
          protocol: TCP
        - containerPort: 443
          name: https
          protocol: TCP
        - containerPort: 9913
          name: nginx-vts
          protocol: TCP
        - containerPort: 9190
          name: php-fpm-export
          protocol: TCP
        - containerPort: 42323
          #hostPort: 42323
          name: php
          protocol: TCP
        - containerPort: 8080
          name: tcp-8080
          protocol: TCP
        readinessProbe:
          failureThreshold: 3
          periodSeconds: 5
          successThreshold: 1
          tcpSocket:
            port: 443
        volumeMounts:
        - name: nginx-configmap
          mountPath: /etc/nginx/conf.d
        - name: nginx-configmap-consul
          mountPath: /etc/nginx/sites-enabled
        - name: nginxcert-configmap
          mountPath: /var/www/cert
        - name: api-nginx-log
        #- name: nfs-pvc
          mountPath: /var/log/nginx/
        - name: nfs-php-logs
          mountPath: /var/www/html/k8s/storage/logs/
        - name: api-php-uploads
          mountPath: /var/www/html/k8s/public/uploads
        - name: api-php-vendor
          mountPath: /var/www/html/k8s/vendor/
      imagePullSecrets:
      - name: registry-secret-new
      volumes:
      - name: nginx-configmap
        configMap:
          name: nginxconf-api
      - name: nginx-configmap-consul
        configMap:
          name: nginxconf-consul.cccc.cn
      - name: nginxcert-configmap
        configMap:
          name: nginxcert-cccc.cn
      - name: api-nginx-log
      #- name: nfs-pvc
        hostPath:
      # persistentVolumeClaim:
          path: /data/logs/api-nginx
         # claimName: product-claim-api
      - name: nfs-php-logs
        persistentVolumeClaim:
          claimName: product-claim-api-php-logs
      - name: api-php-uploads
        persistentVolumeClaim:
          claimName: product-claim-api-upload
      - name: api-php-vendor
        persistentVolumeClaim:
          claimName: product-claim-api-vendor

GIT误删库如何恢复

转:
误操作删除服务器上git仓库的处理方法

说明:

git 服务器上仓库代码不小心被删除了,但是通过git clone下来的code存在

另外git 服务器上没有建立任何branch信息;

1.在之前clone过服务器代码的客户端机器上,找到clone下来的代码目录,

执行git remote -vv
查询原来git 仓库所在服务器的IP地址以及路径;

2.根据步骤1上查询的结果,在服务器上找到对应路径,创建git仓库;

git init –bare

3.在客户端之前clone 代码所在目录,强制push客户端的代码至服务器

git push origin master –force

这样原来clone下来的代码以及change log等信息,就会被强制clone到服务器上;

说明:此处必须加–force选项,否则会因服务器上仓库信息与客户端仓库信息不匹配导致push失败
——————————————————
免密模式中的.ssh目录必须是700权限,和authorized_keys必须是600权限