27.9 调用go-ansible执行playbook拷贝json文件重载采集器

福大大架构师每日一题2024-10-30 14:14

本节重点介绍 :

  • go-ansible执行playbook
  • 编写分发重载的playbook
  • 编译执行
    • 测试停掉一个节点
    • 测试停掉的节点再回来

go-ansible执行playbook

  • 新增 goansiblerun/run.go
go 复制代码 
package goansiblerun

import (
	"context"
	"github.com/apenella/go-ansible/pkg/execute"
	"github.com/apenella/go-ansible/pkg/stdoutcallback/results"
	"github.com/go-kit/log"
	"github.com/go-kit/log/level"
	"time"

	"github.com/apenella/go-ansible/pkg/options"
	"github.com/apenella/go-ansible/pkg/playbook"
)

func AnsiRunPlay(logger log.Logger, srvName string, remoteHost string, extraVars map[string]interface{}, ansiYamlPath string) {
	ansiblePlaybookConnectionOptions := &options.AnsibleConnectionOptions{
		Connection: "smart",
	}
	ctx, cancel := context.WithTimeout(context.Background(), time.Duration(10)*time.Second)
	defer cancel()
	ansiblePlaybookOptions := &playbook.AnsiblePlaybookOptions{
		Inventory: remoteHost + ",",
		ExtraVars: extraVars,
	}

	lplaybook := &playbook.AnsiblePlaybookCmd{
		Playbooks:         []string{ansiYamlPath},
		ConnectionOptions: ansiblePlaybookConnectionOptions,
		Options:           ansiblePlaybookOptions,
		Exec: execute.NewDefaultExecute(
			execute.WithTransformers(
				results.Prepend("Go-ansible example"),
			),
		),
		//StdoutCallback: "json",
	}

	err := lplaybook.Run(ctx)
	if err != nil {
		level.Error(logger).Log("msg", "create_Watch_by_watch_config_error", "srv_name", srvName, "host", remoteHost, "error", err)

	}
}

解读一下

  • 使用 https://github.com/apenella/go-ansible
  • ansiYamlPath代表要执行那个playbook
  • extraVars代表 playbook中的外部参数
  • Inventory代表 执行的host
  • 每个执行 设置10秒的超时时间
go 复制代码 
ctx, cancel := context.WithTimeout(context.Background(), time.Duration(10)*time.Second)

在Dispatch分发的时候调用ansible playbook

  • 位置 service/shard_service.go
  • 将配置中的 src_sd_file_name，dest_sd_file_name，yaml_path等参数传入playbook
go 复制代码 
	for node, ts := range nodeMap {
		// 拼接一个json文件的名字
		// 服务名_节点ip_索引_分片总数_target总数.json
		jsonFileName := fmt.Sprintf("%s_%s_%d_%d_%d.json",
			this.SrvName,
			node,
			index,
			allNum,
			len(ts),

		)
		// 写json文件
		writeJsonFile(jsonFileName, ts)

		extraVars := make(map[string]interface{})
		extraVars["src_sd_file_name"] = jsonFileName
		extraVars["dest_sd_file_name"] = this.DestSdFileName
		extraVars["service_port"] = this.Port
		level.Info(this.logger).Log(
			"msg", "goansiblerun.run",

			"this.SrvName", this.SrvName,
			"jsonFileName", jsonFileName,
			"node", node,
			"index", index,
			"all", allNum,
			"targetNum", len(ts),

		)
		go goansiblerun.AnsiRunPlay(this.logger, this.SrvName, node, extraVars, this.YamlPath)
		index++
	}

编写分发重载的playbook

  • yaml名字 copy_file_and_reload_prome.yaml
  • 先将本地的json文件copy到目标机器上
  • 目标目录为 /opt/app/prometheus/sd
  • 然后给prometheus采集器发送reload命令
yaml 复制代码 
- name:  copy_file_and_reload
  hosts: all
  user: root
  gather_facts:  false
  vars:
      target_path: /opt/app/prometheus/sd
  tasks:
      - name: copy target file
        copy:
          src: '{{ item.src }}'
          dest: '{{ item.dest }}'
          owner: root
          group: root
          mode: 0644
          force: true
        with_items:
          - { src: './{{ src_sd_file_name }}', dest: '{{ target_path }}/{{ dest_sd_file_name }}' }

      - name: reload_service
        shell: /usr/bin/curl -X POST http://localhost:{{ service_port }}/-/reload &

prometheus上的配置

  • 将blackbox_http改造为管控的
yaml 复制代码 
  - job_name: 'blackbox-http-shard'
    # metrics的path 注意不都是/metrics
    metrics_path: /probe
    # 传入的参数
    params:
      module: [http_2xx]  # Look for a HTTP 200 response.
    file_sd_configs:
      - files:
          - /opt/app/prometheus/sd/file_sd_by_prome_shared.json
        refresh_interval: 2m
    relabel_configs:
      - source_labels: [__address__]
        target_label: __param_target
      - source_labels: [__param_target]
        target_label: instance
      - target_label: __address__
        replacement: 172.20.70.205:9115  # The blackbox exporter's real hostname:port.

编译执行

正常执行两个节点均分

shell 复制代码 
level=info ts=2021-08-26T19:08:09.770+08:00 caller=shard_service.go:103 msg="RunRefreshServiceNode start...."
level=info ts=2021-08-26T19:08:09.771+08:00 caller=shard_service.go:198 msg=RunDispatch.start name=scrape_prometheus_node_exporter
ts=2021-08-26T19:08:09.771+08:00 caller=log.go:124 level=info msg="RunRefreshServiceNode start...."
<nil>
level=info ts=2021-08-26T19:08:09.772+08:00 caller=shard_service.go:180 msg=goansiblerun.run this.SrvName=scrape_prometheus_node_exporter jsonFileName=scrape_prometheus_node_exporter_172.20.70.215_1_2_2.json node=172.20.70.215 index=1 all=2 targetNum=2
<nil>
level=info ts=2021-08-26T19:08:09.773+08:00 caller=shard_service.go:180 msg=goansiblerun.run this.SrvName=scrape_prometheus_node_exporter jsonFileName=scrape_prometheus_node_exporter_172.20.70.205_2_2_3.json node=172.20.70.205 index=2 all=2 targetNum=3
ts=2021-08-26T19:08:09.774+08:00 caller=log.go:124 level=info msg=service_node_change_by_healthy_check srv_name=scrape_prometheus_node_exporter num=2 detail="172.20.70.205 172.20.70.215"
level=info ts=2021-08-26T19:08:09.774+08:00 caller=shard_service.go:119 msg=RunReshardHashRing_node_same nodes=172.20.70.205,172.20.70.215
prome-shard ── 
prome-shard ── PLAY [copy_file_and_reload] *******************************************************************************
prome-shard ── 
prome-shard ── TASK [copy target file] ***********************************************************************************
prome-shard ── 
prome-shard ── PLAY [copy_file_and_reload] *******************************************************************************
prome-shard ── 
prome-shard ── TASK [copy target file] ***********************************************************************************
prome-shard ── ok: [172.20.70.205] => (item={u'dest': u'/opt/app/prometheus/sd/file_sd_by_prome_shared.json', u'src': u'./scrape_prometheus_node_exporter_172.20.70.205_2_2_3.json'})
prome-shard ── 
prome-shard ── TASK [reload_service] *************************************************************************************
prome-shard ── ok: [172.20.70.215] => (item={u'dest': u'/opt/app/prometheus/sd/file_sd_by_prome_shared.json', u'src': u'./scrape_prometheus_node_exporter_172.20.70.215_1_2_2.json'})
prome-shard ── 
prome-shard ── TASK [reload_service] *************************************************************************************
[WARNING]: Consider using the get_url or uri module rather than running 'curl'.  If you need to use
command because get_url or uri is insufficient you can add 'warn: false' to this command task or set
'command_warnings=False' in ansible.cfg to get rid of this message.
prome-shard ── changed: [172.20.70.215]
prome-shard ── 
prome-shard ── PLAY RECAP ************************************************************************************************
prome-shard ── 172.20.70.215              : ok=2    changed=1    unreachable=0    failed=0    skipped=0    rescued=0    ignored=0   
prome-shard ── 
[WARNING]: Consider using the get_url or uri module rather than running 'curl'.  If you need to use
command because get_url or uri is insufficient you can add 'warn: false' to this command task or set
'command_warnings=False' in ansible.cfg to get rid of this message.
prome-shard ── changed: [172.20.70.205]
prome-shard ── 
prome-shard ── PLAY RECAP ************************************************************************************************
prome-shard ── 172.20.70.205              : ok=2    changed=1    unreachable=0    failed=0    skipped=0    rescued=0    ignored=0   
prome-shard ──

停掉其中一个节点，全部分配给存活的节点

shell 复制代码 
ts=2021-08-26T19:09:56.422+08:00 caller=log.go:124 level=info msg=service_node_change_by_healthy_check srv_name=scrape_prometheus_node_exporter num=1 detail=172.20.70.215
level=info ts=2021-08-26T19:09:56.423+08:00 caller=shard_service.go:114 msg=RunReshardHashRing_node_update_reshard old_num=2 new_num=1 oldnodes=172.20.70.205,172.20.70.215 newnodes=172.20.70.215
<nil>
level=info ts=2021-08-26T19:09:56.424+08:00 caller=shard_service.go:180 msg=goansiblerun.run this.SrvName=scrape_prometheus_node_exporter jsonFileName=scrape_prometheus_node_exporter_172.20.70.215_1_1_5.json node=172.20.70.215 index=1 all=1 targetNum=5
prome-shard ── 
prome-shard ── PLAY [copy_file_and_reload] *******************************************************************************
prome-shard ── 
prome-shard ── TASK [copy target file] ***********************************************************************************
prome-shard ── changed: [172.20.70.215] => (item={u'dest': u'/opt/app/prometheus/sd/file_sd_by_prome_shared.json', u'src': u'./scrape_prometheus_node_exporter_172.20.70.215_1_1_5.json'})
prome-shard ── 
prome-shard ── TASK [reload_service] *************************************************************************************
[WARNING]: Consider using the get_url or uri module rather than running 'curl'.  If you need to use
command because get_url or uri is insufficient you can add 'warn: false' to this command task or set
prome-shard ── changed: [172.20.70.215]
'command_warnings=False' in ansible.cfg to get rid of this message.
prome-shard ── 
prome-shard ── PLAY RECAP ************************************************************************************************
prome-shard ── 172.20.70.215              : ok=2    changed=2    unreachable=0    failed=0    skipped=0    rescued=0    ignored=0

再启动节点，又再均分

shell 复制代码 
ts=2021-08-26T19:11:06.439+08:00 caller=log.go:124 level=info msg=service_node_change_by_healthy_check srv_name=scrape_prometheus_node_exporter num=2 detail="172.20.70.205 172.20.70.215"
level=info ts=2021-08-26T19:11:06.440+08:00 caller=shard_service.go:114 msg=RunReshardHashRing_node_update_reshard old_num=1 new_num=2 oldnodes=172.20.70.215 newnodes=172.20.70.205,172.20.70.215
<nil>
level=info ts=2021-08-26T19:11:06.441+08:00 caller=shard_service.go:180 msg=goansiblerun.run this.SrvName=scrape_prometheus_node_exporter jsonFileName=scrape_prometheus_node_exporter_172.20.70.215_1_2_2.json node=172.20.70.215 index=1 all=2 targetNum=2
<nil>
level=info ts=2021-08-26T19:11:06.441+08:00 caller=shard_service.go:180 msg=goansiblerun.run this.SrvName=scrape_prometheus_node_exporter jsonFileName=scrape_prometheus_node_exporter_172.20.70.205_2_2_3.json node=172.20.70.205 index=2 all=2 targetNum=3
prome-shard ── 
prome-shard ── PLAY [copy_file_and_reload] *******************************************************************************
prome-shard ── 
prome-shard ── TASK [copy target file] ***********************************************************************************
prome-shard ── 
prome-shard ── PLAY [copy_file_and_reload] *******************************************************************************
prome-shard ── 
prome-shard ── TASK [copy target file] ***********************************************************************************
prome-shard ── changed: [172.20.70.215] => (item={u'dest': u'/opt/app/prometheus/sd/file_sd_by_prome_shared.json', u'src': u'./scrape_prometheus_node_exporter_172.20.70.215_1_2_2.json'})
prome-shard ── 
prome-shard ── TASK [reload_service] *************************************************************************************
prome-shard ── changed: [172.20.70.205] => (item={u'dest': u'/opt/app/prometheus/sd/file_sd_by_prome_shared.json', u'src': u'./scrape_prometheus_node_exporter_172.20.70.205_2_2_3.json'})
prome-shard ──

效果图

  • target页面
  • 感知到节点变化的shard日志
  • ansible的日志
  • consul的服务截图

回顾一下架构图

本节重点总结 :

  • go-ansible执行playbook
  • 编写分发重载的playbook
  • 编译执行
    • 测试停掉一个节点
    • 测试停掉的节点再回来
相关推荐
hummhumm1 小时前
第 22 章 - Go语言 测试与基准测试
java·大数据·开发语言·前端·python·golang·log4j
hummhumm1 小时前
第 28 章 - Go语言 Web 开发入门
java·开发语言·前端·python·sql·golang·前端框架
YMWM_2 小时前
第一章 Go语言简介
开发语言·后端·golang
hummhumm3 小时前
第 25 章 - Golang 项目结构
java·开发语言·前端·后端·python·elasticsearch·golang
好奇的菜鸟3 小时前
Go语言中的引用类型：指针与传递机制
开发语言·后端·golang
一条晒干的咸魚3 小时前
【Web前端】创建我的第一个 Web 表单
服务器·前端·javascript·json·对象·表单
Alive～o.03 小时前
Go语言进阶&依赖管理
开发语言·后端·golang
学Linux的语莫4 小时前
Ansible Playbook剧本用法
linux·服务器·云计算·ansible
Nightwish55 小时前
ansible操作随记（一）
ansible
ifanatic8 小时前
[面试]-golang基础面试题总结
面试·职场和发展·golang
热门推荐
01(欧拉)openEuler系统添加网卡文件配置流程、(欧拉)openEuler系统手动配置ipv6地址流程、(欧拉)openEuler系统网络管理说明02PyTorch机器学习实现液态神经网络03【HarmonyOS】HUAWEI DevEco Studio 下载地址汇总04玄机平台应急响应—webshell查杀05Coze扣子平台完整体验和实践（附国内和国际版对比）06Ubuntu 20.04使用Livox mid 360 测试 FAST_LIO07怎样让音频速度变慢？请跟随以下方法进行操作08Unity中PICO实现 隔空取物 和 接触抓取物体09RAG 实践- Ollama+RagFlow 部署本地知识库10【目标跟踪】相机运动补偿