1.搭建環(huán)境:
系統(tǒng)版本:
Linux centos-linux.shared 3.10.0-123.el7.x86_64 #1 SMP Mon Jun 30 12:09:22 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux
python版本:
Python 3.5.1
1.1.搭建python3環(huán)境:
本人在嘗試過后選擇集成環(huán)境Anaconda
1.1.1.編譯
# 下載依賴
yum install -y ncurses-devel openssl openssl-devel zlib-devel gcc make glibc-devel libffi-devel glibc-static glibc-utils sqlite-devel readline-devel tk-devel gdbm-devel db4-devel libpcap-devel xz-deve
# 下載python版本
wget https://www.python.org/ftp/python/3.5.1/Python-3.5.1.tgz
# 或者使用國(guó)內(nèi)源
wget http://mirrors.sohu.com/python/3.5.1/Python-3.5.1.tgz
mv Python-3.5.1.tgz /usr/local/src;cd /usr/local/src
# 解壓
tar -zxf Python-3.5.1.tgz;cd Python-3.5.1
# 編譯安裝
./configure --prefix=/usr/local/python3.5 --enable-shared
make && make install
# 建立軟鏈接
ln -s /usr/local/python3.5/bin/python3 /usr/bin/python3
echo "/usr/local/python3.5/lib" > /etc/ld.so.conf.d/python3.5.conf
ldconfig
# 驗(yàn)證python3
python3
# Python 3.5.1 (default, Oct 9 2016, 11:44:24)
# [GCC 4.8.5 20150623 (Red Hat 4.8.5-4)] on linux
# Type "help", "copyright", "credits" or "license" for more information.
# >>>
# pip
/usr/local/python3.5/bin/pip3 install --upgrade pip
ln -s /usr/local/python3.5/bin/pip /usr/bin/pip
# 本人在安裝時(shí)出現(xiàn)問題 將pip重裝
wget https://bootstrap.pypa.io/get-pip.py --no-check-certificate
python get-pip.py
1.1.2.集成環(huán)境anaconda
# 集成環(huán)境anaconda(推薦)
wget https://repo.continuum.io/archive/Anaconda3-4.2.0-Linux-x86_64.sh
# 直接安裝即可
./Anaconda3-4.2.0-Linux-x86_64.sh
# 若出錯(cuò)裂七,可能是解壓失敗
yum install bzip2
1.2.安裝mariaDB
# 安裝
yum -y install mariadb mariadb-server
# 啟動(dòng)
systemctl start mariadb
# 設(shè)置為開機(jī)啟動(dòng)
systemctl enable mariadb
# 配置密碼 默認(rèn)為空
mysql_secure_installation
# 登錄
mysql -u root -p
# 創(chuàng)建一個(gè)用戶 自己設(shè)定賬戶密碼
CREATE USER 'user_name'@'localhost' IDENTIFIED BY 'user_pass';
GRANT ALL PRIVILEGES ON *.* TO 'user_name'@'localhost' WITH GRANT OPTION;
CREATE USER 'user_name'@'%' IDENTIFIED BY 'user_pass';
GRANT ALL PRIVILEGES ON *.* TO 'user_name'@'%' WITH GRANT OPTION;
1.3.安裝pyspider
本人使用Anaconda
# 搭建虛擬環(huán)境sbird python版本3.*
conda create -n sbird python=3*
# 進(jìn)入環(huán)境
source activate sbird
# 安裝pyspider
pip install pyspider
# 報(bào)錯(cuò)
# it does not exist. The exported locale is "en_US.UTF-8" but it is not supported
# 執(zhí)行 可寫入.bashrc
export LC_ALL=en_US.utf-8
export LANG=en_US.utf-8
#ImportError: pycurl: libcurl link-time version (7.29.0) is older than compile-time version (7.49.0)
conda install pycurl
# 退出
source deactivate sbird
# 若在虛擬機(jī)內(nèi) 出現(xiàn)無(wú)法訪問localhost:5000 可關(guān)閉防火墻
systemctl stop firewalld.service
#########直接運(yùn)行源碼==============
mkdir git;cd git
# 下載
git clone https://github.com/binux/pyspider.git
# 安裝
/root/anaconda3/envs/sbird/bin/python /root/git/pyspider/run.py
其他方法
# 搭建虛擬環(huán)境
pip install virtualenv
mkdir python;cd python
# 創(chuàng)建虛擬環(huán)境pyenv3
virtualenv -p /usr/bin/python3 pyenv3
# 進(jìn)入虛擬環(huán)境 激活環(huán)境
cd pyenv3/
source ./bin/activate
pip install pyspider
# 若pycurl報(bào)錯(cuò)
yum install libcurl-devel
# 繼續(xù)
pip install pyspider
# 關(guān)閉
deactivate
本人推薦用anaconda方式安裝
若pyspider運(yùn)行過程中出現(xiàn)錯(cuò)誤,參考anaconda
安裝部分,至此织阅,訪問localhost:5000
可看到頁(yè)面。
1.4.安裝Supervisor
# 安裝
yum install supervisor -y
# 若無(wú)法檢索 則添加阿里的epel源
vim /etc/yum.repos.d/epel.repo
# 添加以下內(nèi)容
[epel]
name=Extra Packages for Enterprise Linux 7 - $basearch
baseurl=http://mirrors.aliyun.com/epel/7/$basearch
http://mirrors.aliyuncs.com/epel/7/$basearch
failovermethod=priority
enabled=1
gpgcheck=0
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7
[epel-debuginfo]
name=Extra Packages for Enterprise Linux 7 - $basearch - Debug
baseurl=http://mirrors.aliyun.com/epel/7/$basearch/debug
http://mirrors.aliyuncs.com/epel/7/$basearch/debug
failovermethod=priority
enabled=0
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7
gpgcheck=0
[epel-source]
name=Extra Packages for Enterprise Linux 7 - $basearch - Source
baseurl=http://mirrors.aliyun.com/epel/7/SRPMS
http://mirrors.aliyuncs.com/epel/7/SRPMS
failovermethod=priority
enabled=0
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7
gpgcheck=0
# 安裝
yum install supervisor -y
# 測(cè)試是否安裝成功
echo_supervisord_conf
1.4.1.Supervisor用法
supervisord #supervisor的服務(wù)器端部分 啟動(dòng)
supervisorctl #啟動(dòng)supervisor的命令行窗口
# 假設(shè)創(chuàng)建進(jìn)程pyspider01
vim /etc/supervisord.d/pyspider01.ini
# 寫入以下內(nèi)容
[program:pyspider01]
command = /root/anaconda3/envs/sbird/bin/python /root/git/pyspider/run.py
directory = /root/git/pyspider
user = root
process_name = %(program_name)s
autostart = true
autorestart = true
startsecs = 3
redirect_stderr = true
stdout_logfile_maxbytes = 500MB
stdout_logfile_backups = 10
stdout_logfile = /pyspider/supervisor/pyspider01.log
# 重載
supervisorctl reload
# 啟動(dòng)
supervisorctl start pyspider01
# 也可這樣啟動(dòng)
supervisord -c /etc/supervisord.conf
# 查看狀態(tài)
supervisorctl status
# output
pyspider01 RUNNING pid 4026, uptime 0:02:40
# 關(guān)閉
supervisorctl shutdown
1.5.安裝redis
# 消息隊(duì)列采用redis
mkdir download;cd download
wget http://download.redis.io/releases/redis-3.2.4.tar.gz
tar xzf redis-3.2.4.tar.gz
cd redis-3.2.4
make
# 或者直接yum安裝
yum -y install redis
# 啟動(dòng)
systemctl start redis.service
# 重啟
systemctl restart redis.service
# 停止
systemctl stop redis.service
# 查看狀態(tài)
systemctl status redis.service
# 更改文件/etc/redis.conf
vim /etc/redis.conf
# 更改內(nèi)容
daemonize no 改為 daemonize yes
bind 127.0.0.1 改為 bind 10.211.55.22(當(dāng)前服務(wù)器ip)
# 重啟redis
systemctl restart redis.service
1.6.關(guān)于自啟動(dòng)
# Supervisor添加到自啟動(dòng)服務(wù)
systemctl enable supervisord.service
# redis添加到自啟動(dòng)服務(wù)
systemctl enable redis.service
# 關(guān)閉防火墻自啟動(dòng)
systemctl disable firewalld.service
至此嗅钻,pyspider單個(gè)服務(wù)器運(yùn)行環(huán)境搭建且部署完畢灼擂,啟動(dòng)localhost:5000
進(jìn)入web界面。
也可編寫腳本運(yùn)行峻贮,在/pyspider/supervisor/pyspider01.log
查看運(yùn)行狀態(tài)。
2.分布式部署
剛才配置的服務(wù)器碉纺,將其命名為centos01
,按照這樣的配置舱呻,再分別部署兩臺(tái)centos02、centos03
茬高。
如下:
服務(wù)器名稱 | ip | 說(shuō)明 |
---|---|---|
centos01 | 10.211.55.22 | redis,mariaDB, scheduler |
centos02 | 10.211.55.23 | fetcher, processor, result_worker,phantomjs |
centos03 | 10.211.55.24 | fetcher, processor,,result_worker,webui |
2.1.centos01
進(jìn)入服務(wù)器centos01
慨亲,經(jīng)過第一步,基本環(huán)境已經(jīng)搭好蛉签,首先編輯配置文件/pyspider/config.json
{
"taskdb": "mysql+taskdb://user_name:user_pass@10.211.55.22:3306/taskdb",
"projectdb": "mysql+projectdb://user_name:user_pass@10.211.55.22:3306/projectdb",
"resultdb": "mysql+resultdb://user_name:user_pass@10.211.55.22:3306/resultdb",
"message_queue": "redis://10.211.55.22:6379/db",
"logging-config": "/pyspider/logging.conf",
"phantomjs-proxy":"10.211.55.23:25555",
"webui": {
"username": "",
"password": "",
"need-auth": false,
"host":"10.211.55.24",
"port":"5000",
"scheduler-rpc":"http:// 10.211.55.22:5002",
"fetcher-rpc":"http://10.211.55.23:5001"
},
"fetcher": {
"xmlrpc":true,
"xmlrpc-host": "0.0.0.0",
"xmlrpc-port": "5001"
},
"scheduler": {
"xmlrpc":true,
"xmlrpc-host": "0.0.0.0",
"xmlrpc-port": "5002"
}
}
嘗試運(yùn)行下:
/root/anaconda3/envs/sbird/bin/python /root/git/pyspider/run.py -c /pyspider/config.json scheduler
# 報(bào)錯(cuò)
ImportError: No module named 'mysql'
# 下載 mysql-connector-python
cd ~/git/
git clone https://github.com/mysql/mysql-connector-python.git
# 安裝
source activate sbird
cd mysql-connector-python
python setup.py install
# 安裝redis
pip install redis
source deactivate
# 運(yùn)行
/root/anaconda3/envs/sbird/bin/python /root/git/pyspider/run.py -c /pyspider/config.json scheduler
# 輸出 ok
[I 161010 15:57:25 scheduler:644] scheduler starting...
[I 161010 15:57:25 scheduler:779] scheduler.xmlrpc listening on 0.0.0.0:5002
[I 161010 15:57:25 scheduler:583] in 5m: new:0,success:0,retry:0,failed:0
運(yùn)行成功后淮野,可直接更改/etc/supervisord.d/pyspider01.ini
如下:
[program:pyspider01]
command = /root/anaconda3/envs/sbird/bin/python /root/git/pyspider/run.py -c /pyspider/config.json scheduler
directory = /root/git/pyspider
user = root
process_name = %(program_name)s
autostart = true
autorestart = true
startsecs = 3
redirect_stderr = true
stdout_logfile_maxbytes = 500MB
stdout_logfile_backups = 10
stdout_logfile = /pyspider/supervisor/pyspider01.log
# 重載
supervisorctl reload
# 查看狀態(tài)
supervisorctl status
centos01
部署完畢。
2.2.centos02
在centos02
中,需要運(yùn)行result_worker、processor潭袱、phantomjs慎式、fetcher
分別建立文件:
/etc/supervisord.d/result_worker.ini
[program:result_worker]
command = /root/anaconda3/envs/sbird/bin/python /root/git/pyspider/run.py -c /pyspider/config.json result_worker
directory = /root/git/pyspider
user = root
process_name = %(program_name)s
autostart = true
autorestart = true
startsecs = 3
redirect_stderr = true
stdout_logfile_maxbytes = 500MB
stdout_logfile_backups = 10
stdout_logfile = /pyspider/supervisor/result_worker.log
/etc/supervisord.d/processor.ini
[program:processor]
command = /root/anaconda3/envs/sbird/bin/python /root/git/pyspider/run.py -c /pyspider/config.json processor
directory = /root/git/pyspider
user = root
process_name = %(program_name)s
autostart = true
autorestart = true
startsecs = 3
redirect_stderr = true
stdout_logfile_maxbytes = 500MB
stdout_logfile_backups = 10
stdout_logfile = /pyspider/supervisor/processor.log
/etc/supervisord.d/phantomjs.ini
[program:phantomjs]
command = /pyspider/phantomjs --config=/pyspider/pjsconfig.json /pyspider/phantomjs_fetcher.js 25555
directory = /root/git/pyspider
user = root
process_name = %(program_name)s
autostart = true
autorestart = true
startsecs = 3
redirect_stderr = true
stdout_logfile_maxbytes = 500MB
stdout_logfile_backups = 10
stdout_logfile = /pyspider/supervisor/phantomjs.log
/etc/supervisord.d/fetcher.ini
[program:fetcher]
command = /root/anaconda3/envs/sbird/bin/python /root/git/pyspider/run.py -c /pyspider/config.json fetcher
directory = /root/git/pyspider
user = root
process_name = %(program_name)s
autostart = true
autorestart = true
startsecs = 3
redirect_stderr = true
stdout_logfile_maxbytes = 500MB
stdout_logfile_backups = 10
stdout_logfile = /pyspider/supervisor/fetcher.log
在pyspider
目錄中建立pjsconfig.json
{
/*--ignore-ssl-errors=true */
"ignoreSslErrors": true,
/*--ssl-protocol=true */
"sslprotocol": "any",
/* Same as: --output-encoding=utf8 */
"outputEncoding": "utf8",
/* persistent Cookies. */
/*cookiesfile="e:/phontjscookies.txt",*/
cookiesfile="pyspider/phontjscookies.txt",
/* load image */
autoLoadImages = false
}
下載phantomjs至/pyspider/
文件夾幕屹,將git/pyspider/pyspider/fetcher/phantomjs_fetcher.js
復(fù)制到phantomjs_fetcher.js
# 重載
supervisorctl reload
# 查看狀態(tài)
supervisorctl status
# output
fetcher RUNNING pid 3446, uptime 0:00:07
phantomjs RUNNING pid 3448, uptime 0:00:07
processor RUNNING pid 3447, uptime 0:00:07
result_worker RUNNING pid 3445, uptime 0:00:07
centos02
部署完畢。
2.3.centos03
部署這三個(gè)進(jìn)程fetcher, processor, result_worker
和centos02
一樣,本服務(wù)器主要是在前面的基礎(chǔ)上加上webui
建立文件:
/etc/supervisord.d/webui.ini
[program:webui]
command = /root/anaconda3/envs/sbird/bin/python /root/git/pyspider/run.py -c /pyspider/config.json webui
directory = /root/git/pyspider
user = root
process_name = %(program_name)s
autostart = true
autorestart = true
startsecs = 3
redirect_stderr = true
stdout_logfile_maxbytes = 500MB
stdout_logfile_backups = 10
stdout_logfile = /pyspider/supervisor/webui.log
# 重載
supervisorctl reload
# 查看狀態(tài)
supervisorctl status
# output
fetcher RUNNING pid 2724, uptime 0:00:07
processor RUNNING pid 2725, uptime 0:00:07
result_worker RUNNING pid 2723, uptime 0:00:07
webui RUNNING pid 2726, uptime 0:00:07
3.總結(jié)
訪問 http://10.211.55.24:5000 即可拟淮,盡情爬取吧很泊。