jenkins发布程序触发shell调用python脚本刷新akamai cdn api

刷新cdn的流程：
jenkins获取git中的代码，触发脚本推送到生产环境中(即cdn的源站) --> 触发脚本获取git工作目录的更新列表，将更新列表拼凑成带域名信息的url，写入到目录中 --> 触发python脚本读取目录中的url发送给akamai的api进行cdn的刷新

参考文档创建client api，此次我们的账号没有创建client api的权限，需要管理员处理
文档地址：https://developer.akamai.com/api/getting-started#beforeyoubegin

创建和api交互的client api后会得到类似如下信息

client_secret = "pass"
host = "host.purge.akamaiapis.net"
access_token = "token"
client_token = "client_token"

api参考地址：

https://github.com/akamai/api-kickstart

jenkins配合shell脚本触发python自动刷新akamai cdn

项目背景：
设计中心开启了统一资源管理系统neveragain.chinaosft.com，即公司后续新开发的站点引用的image,css,js等公共资源统一从该系统获取

需求：
当代码合并到master的指定目录后需要自动发布代码及图片等资源到生产环境并且能即时刷新CDN

转化为运维需求：
git中提交代码并且合并master --> 使用jenkins发布代码到 neveragain.chinaosft.com 所在服务器 --> 获取更新的代码并且刷新CND

具体实现过程
1.配置jenkins让jenkins能拉取代码到jenkins服务器，因配置较为简单，此处略

2.配置发布的脚本：

脚本的主要逻辑：发布指定代码到生产环境服务器 --> 同时获取代码中dist目录更新的文件，将文件拼凑成CDN的api可以识别的URL --> 使用python脚本读取需要更新的URL列表并且触发AKAMAI CDN API刷新资源

jenkins中的shell脚本

[root@jenkins:/usr/local/worksh/jeninks_task]# cat neveragain_chinasoft_com.sh
#!/bin/bash
#############################################
# 通过jenkins发布任务 neveragain.chinasoft.com 发布  注意/data/www/vhosts/neveragain.chinasoft.com/httpdocs/dist/ 发布到线上对应的是2019目录

cart_iplist="1.1.1.1"

function neveragain_chinasoft_eus_rsync()
{
for ip in $cart_iplist
do
        echo "-- start pub --- 预发布到外网 ${ip} ----------------------------------------"
    /usr/local/bin/rsync -vazP --bwlimit= --exclude='.git/' --exclude='.gitignore/' --password-file=/data/www/.rsync/rsyncd.pass /data/www/vhosts/neveragain.chinasoft.com/httpdocs/dist/ apache@$ip::apache/data/www/vhosts/neveragain.chinasoft.com/httpdocs//
    if [[ $? ==  || $? ==  ]];then
            rsync_edit=
    else
            rsync_edit=
            echo "`date` rsync发布失败! -> editUrls.txt"
            exit
    fi

    echo -e "-- end pub ${ip} ----------------------------------------------------------\n\n"
done
}

# 执行同步
neveragain_chinasoft_eus_rsync

# 读取git的更新列表，发送请求调用python脚本刷新akamai CDN
function update_cdn
{
    # 工作目录
    WORKSPACE="/data/jenkins_home/workspace/DesignCenter.neveragain.chinasoft.com/"
    cd $WORKSPACE

    # 获取git变更列表
    changefiles=$(git diff --name-only HEAD~ HEAD)
    #echo $changefiles
    # 先把文件路径写死，作为测试使用
    #changefiles="dist/assets/image/box/drfone-mac.png dist/assets/image/box/drfone-win.png dist/assets/image/box/dvdcreator-mac.png dist/assets/image/box/dvdcreator-win.png"

    #
    now_time="`date +%Y%m%d%H%M%S`"
    # 将更新的文件列表写入日志文件中
    for newfile in $changefiles;
    do
        start_str=${newfile::}
        #echo $start_str
        # 如果变更的文件是 dist 目录下的文件就触发该文件刷新CDN
        if [ $start_str == 'dist' ];then
            need_file=${newfile:}
            #echo $need_file
            need_url="https://neveragain.chinasoft.com/2019/$need_file"
            #echo $need_url
            echo "${need_url}" >> "/usr/local/worksh/jeninks_task/akamai_api/logs/${now_time}.log"
        fi
    done

# 调用Python脚本刷新cdn
/usr/local/worksh/jeninks_task/akamai_api_venv/bin/python /usr/local/worksh/jeninks_task/akamai_api/akamai_api.py $now_time
if [ $? !=  ];then
    echo "刷新CDN失败"
    exit
else
    echo "刷新CDN成功"
fi

}
# 刷新cdn
update_cdn

# python脚本

# 刷新cdn的python脚本结构
[root@jenkins:/usr/local/worksh/jeninks_task/akamai_api]# tree
.
├── akamai_api.py
├── lib
│   ├── http_calls.py
│   ├── __init__.py
├── logs
│   ├── 20190814164047.log
│   └── 20190814172256.log
├── log.txt
├── README.md
└── requirement.txt

# cat /usr/local/worksh/jeninks_task/akamai_api/logs/20190814172256.log
https://neveragain.chinasoft.com/2019/assets/icon/brand/finder.svg
https://neveragain.chinasoft.com/2019/assets/icon/logo/edraw-horizontal-white.png

# 主程序
[root@jenkins:/usr/local/worksh/jeninks_task]# cat /usr/local/worksh/jeninks_task/akamai_api/akamai_api.py
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @File  : akamai_api.py
# @Desc  : 读取指定文件中url的路径内容，刷新Akamai CDN的边缘节点数据

import requests, json,time,os,sys
from lib.http_calls import EdgeGridHttpCaller
from akamai.edgegrid import EdgeGridAuth
import logging

class Akamai_API():
    def __init__(self,api_host,api_access_token,api_client_token,api_client_secret,verbose=False,debug=False,action="delete",network="production"):
        self.host = api_host
        self.access_token = api_access_token
        self.client_token = api_client_token
        self.client_secret = api_client_secret

        self.verbose = verbose
        self.debug = debug

        #API的清除动作:delete invalidate
        self.action =action
        self.network =network

        self.session = requests.Session()

    def __auth(self):
        self.session.auth = EdgeGridAuth(
            client_token=self.client_token,
            client_secret=self.client_secret,
            access_token=self.access_token
        )
        return self.session

    def postPurgeRequest(self,refush_url_list):
        self.__auth()
        baseurl = '%s://%s/' % ('https', self.host)
        httpCaller = EdgeGridHttpCaller(self.session, self.debug, self.verbose, baseurl)
        purge_obj = {
            # "objects": [
            #     "https://bc.akamaiapibootcamp.com/index.html"
            # ]
            "objects": refush_url_list,
        }
        # print("Adding %s request to queue - %s" % (self.action, json.dumps(purge_obj)))
        purge_post_result = httpCaller.postResult('/ccu/v3/%s/url/%s' % (self.action, self.network), json.dumps(purge_obj))
        return purge_post_result

def ReadFile(filename):
    """
    读取文件内容中的url路径
    每行一条url路径
    """
    l = []
    error_l = []
    with open(filename) as f:
        for url in  f.readlines():
            url_str = str(url).strip("\n")
            if str(url).startswith("https://neveragain.chinasoft.com"):
                l.append(url_str)
            else:
                error_l.append(url_str)
    if error_l:
        raise Exception("The format of the file path is incorrect. %s"%('\n'.join(error_l)))
    return l

if __name__ == "__main__":
    #API信息
    API_HOST = "host.purge.akamaiapis.net"
    API_ACCESS_TOKEN = "token"
    API_CLIENT_TOKEN = "client_token"
    API_CLIENT_SECRET = "api_client_secret="

    api = Akamai_API(api_host=API_HOST,api_access_token=API_ACCESS_TOKEN,api_client_token=API_CLIENT_TOKEN,api_client_secret=API_CLIENT_SECRET)

    #接收url文件名称
    if len(sys.argv) != 2:
        raise Exception("Not enough parameters for %s"%sys.argv[0])
    prefix_url_filename = sys.argv[1]

    # 定义日志级别
    baseDir = os.path.dirname(os.path.abspath(__file__))
    logfile = os.path.join(baseDir,"log.txt")
    logging.basicConfig(level=logging.INFO,
                        filename=logfile,
                        filemode='a',
                        format='%(asctime)s - %(filename)s - %(levelname)s: %(message)s')
    logger = logging.getLogger(__name__)

    #读取url的文件内容
    filename = os.path.join(baseDir,os.path.join("logs","%s.log"%prefix_url_filename))
    if not os.path.isfile(filename):
        raise Exception("Not exists file %s" %filename)
    url_list = ReadFile(filename)

    #每次POST提交url的条数
    MAX_REQUEST_SIZE = 800
    while url_list:
        batch = []
        batch_size = 0

        while url_list and batch_size < MAX_REQUEST_SIZE:
            next_url = url_list.pop()
            batch.append(next_url)
            batch_size += 1
        if batch:
            response = api.postPurgeRequest(batch)
            if response["httpStatus"] != 201:
                # 将本次请求url返回到总url列表中，以便稍后在试
                url_list.extend(batch)
                #速率限制
                if response["httpStatus"] == 507:
                    details = response.json().get('detail', '<response did not contain "detail">')
                    print('Will retry request in 1s seconds due to API rate-limit: %s,Try again now.'%details)
                    logger.info('Will retry request in 1s seconds due to API rate-limit: %s,Try again now.'%details)
                    time.sleep(1)
                # 针对速率限制以外的错误  抛出
                if response["httpStatus"] != 507:
                    details = response.json().get('detail', '<response did not contain "detail">')
                    print("{status:Failed,detail:%s}"%details)
                    logger.info("{status:Failed,detail:%s}"%details)
                    response.raise_for_status()
            else:
                logger.info("{status:Success,supportId:%s,purgeId:%s,queue:%s}"%(response["supportId"],response["purgeId"],json.dumps(batch)))

# 依赖包：

[root@jenkins:/usr/local/worksh/jeninks_task/akamai_api]# cat requirement.txt
asn1crypto==0.24.0
certifi==2019.6.16
cffi==1.12.3
chardet==3.0.4
configparser==3.7.4
cryptography==2.7
edgegrid-python==1.1.1
idna==2.8
ndg-httpsclient==0.5.1
pyasn1==0.4.6
pycparser==2.19
pyOpenSSL==19.0.0
requests==2.22.0
six==1.12.0
urllib3==1.25.3

[root@jenkins:/usr/local/worksh/jeninks_task/akamai_api]# cat lib/http_calls.py
#!/usr/bin/env python

# Python edgegrid module
""" Copyright 2015 Akamai Technologies, Inc. All Rights Reserved.

 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.

 You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

 Unless required by applicable law or agreed to in writing, software
 distributed under the License is distributed on an "AS IS" BASIS,
 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License.
"""
import sys

if sys.version_info[0] >= 3:
    # python3
    from urllib import parse
else:
    # python2.7
    import urlparse as parse

import logging, json

logger = logging.getLogger(__name__)

class EdgeGridHttpCaller():
    def __init__(self, session, debug, verbose, baseurl):
        self.debug = debug
        self.verbose = verbose
        self.session = session
        self.baseurl = baseurl
        return None

    def urlJoin(self, url, path):
        return parse.urljoin(url, path)

    def getResult(self, endpoint, parameters=None):
        path = endpoint
        endpoint_result = self.session.get(parse.urljoin(self.baseurl, path), params=parameters)
        if self.verbose: print(">>>\n" + json.dumps(endpoint_result.json(), indent=2) + "\n<<<\n")
        status = endpoint_result.status_code
        if self.verbose: print("LOG: GET %s %s %s" % (endpoint, status, endpoint_result.headers["content-type"]))
        self.httpErrors(endpoint_result.status_code, path, endpoint_result.json())
        return endpoint_result.json()

    def httpErrors(self, status_code, endpoint, result):
        if not isinstance(result, list):
            details = result.get('detail') or result.get('details') or ""
        else:
            details = ""
        if status_code == 403:
            error_msg = "ERROR: Call to %s failed with a 403 result\n" % endpoint
            error_msg += "ERROR: This indicates a problem with authorization.\n"
            error_msg += "ERROR: Please ensure that the credentials you created for this script\n"
            error_msg += "ERROR: have the necessary permissions in the Luna portal.\n"
            error_msg += "ERROR: Problem details: %s\n" % details
            exit(error_msg)

        if status_code in [400, 401]:
            error_msg = "ERROR: Call to %s failed with a %s result\n" % (endpoint, status_code)
            error_msg += "ERROR: This indicates a problem with authentication or headers.\n"
            error_msg += "ERROR: Please ensure that the .edgerc file is formatted correctly.\n"
            error_msg += "ERROR: If you still have issues, please use gen_edgerc.py to generate the credentials\n"
            error_msg += "ERROR: Problem details: %s\n" % result
            exit(error_msg)

        if status_code in [404]:
            error_msg = "ERROR: Call to %s failed with a %s result\n" % (endpoint, status_code)
            error_msg += "ERROR: This means that the page does not exist as requested.\n"
            error_msg += "ERROR: Please ensure that the URL you're calling is correctly formatted\n"
            error_msg += "ERROR: or look at other examples to make sure yours matches.\n"
            error_msg += "ERROR: Problem details: %s\n" % details
            exit(error_msg)

        error_string = None
        if "errorString" in result:
            if result["errorString"]:
                error_string = result["errorString"]
        else:
            for key in result:
                if type(key) is not str or isinstance(result, dict) or not isinstance(result[key], dict):
                    continue
                if "errorString" in result[key] and type(result[key]["errorString"]) is str:
                    error_string = result[key]["errorString"]
        if error_string:
            error_msg = "ERROR: Call caused a server fault.\n"
            error_msg += "ERROR: Please check the problem details for more information:\n"
            error_msg += "ERROR: Problem details: %s\n" % error_string
            exit(error_msg)

    def postResult(self, endpoint, body, parameters=None):
        headers = {'content-type': 'application/json'}
        path = endpoint
        endpoint_result = self.session.post(parse.urljoin(self.baseurl, path), data=body, headers=headers,
                                            params=parameters)
        status = endpoint_result.status_code
        if self.verbose: print("LOG: POST %s %s %s" % (path, status, endpoint_result.headers["content-type"]))
        if status == 204:
            return {}
        self.httpErrors(endpoint_result.status_code, path, endpoint_result.json())

        if self.verbose: print(">>>\n" + json.dumps(endpoint_result.json(), indent=2) + "\n<<<\n")
        return endpoint_result.json()

    def postFiles(self, endpoint, file):
        path = endpoint
        endpoint_result = self.session.post(parse.urljoin(self.baseurl, path), files=file)
        status = endpoint_result.status_code
        if self.verbose: print("LOG: POST FILES %s %s %s" % (path, status, endpoint_result.headers["content-type"]))
        if status == 204:
            return {}
        self.httpErrors(endpoint_result.status_code, path, endpoint_result.json())

        if self.verbose: print(">>>\n" + json.dumps(endpoint_result.json(), indent=2) + "\n<<<\n")
        return endpoint_result.json()

    def putResult(self, endpoint, body, parameters=None):
        headers = {'content-type': 'application/json'}
        path = endpoint

        endpoint_result = self.session.put(parse.urljoin(self.baseurl, path), data=body, headers=headers,
                                           params=parameters)
        status = endpoint_result.status_code
        if self.verbose: print("LOG: PUT %s %s %s" % (endpoint, status, endpoint_result.headers["content-type"]))
        if status == 204:
            return {}
        if self.verbose: print(">>>\n" + json.dumps(endpoint_result.json(), indent=2) + "\n<<<\n")
        return endpoint_result.json()

    def deleteResult(self, endpoint):
        endpoint_result = self.session.delete(parse.urljoin(self.baseurl, endpoint))
        status = endpoint_result.status_code
        if self.verbose: print("LOG: DELETE %s %s %s" % (endpoint, status, endpoint_result.headers["content-type"]))
        if status == 204:
            return {}
        if self.verbose: print(">>>\n" + json.dumps(endpoint_result.json(), indent=2) + "\n<<<\n")
        return endpoint_result.json()

# 报错

warning: inexact rename detection was skipped due to too many files.
warning: you may want to set your diff.renameLimit variable to at least 3074 and retry the command.
Traceback (most recent call last):
  File "/usr/local/worksh/jeninks_task/akamai_api/akamai_api.py", line 109, in <module>
    response = api.postPurgeRequest(batch)
  File "/usr/local/worksh/jeninks_task/akamai_api/akamai_api.py", line 48, in postPurgeRequest
    purge_post_result = httpCaller.postResult('/ccu/v3/%s/url/%s' % (self.action, self.network), json.dumps(purge_obj))
  File "/usr/local/worksh/jeninks_task/akamai_api/lib/http_calls.py", line 112, in postResult
    self.httpErrors(endpoint_result.status_code, path, endpoint_result.json())
  File "/usr/local/worksh/jeninks_task/akamai_api_venv/lib/python3.6/site-packages/requests/models.py", line 897, in json
    return complexjson.loads(self.text, **kwargs)
  File "/usr/local/python3/lib/python3.6/json/__init__.py", line 354, in loads
    return _default_decoder.decode(s)
  File "/usr/local/python3/lib/python3.6/json/decoder.py", line 339, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/local/python3/lib/python3.6/json/decoder.py", line 357, in raw_decode
    raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)
刷新CDN失败

修改 MAX_REQUEST_SIZE = 200，每次批量提交800条，被cdn拒绝，修改为200后问题解决