How to use Docker to build a Django, Nginx, R, and Python deployment environment
This article introduces how to use Docker to build a Django, Nginx, R, and Python deployment environment. I would like to share it with you. The details are as follows: Basic environment: Ubuntu 16.10 docker 17.06.0-ce Compress your own project files 1. What needs to be noted here is that when compressing, you also need to write the Python packages you need in requirement.txt, so that when you set up the environment, you will know what packages you need and can install them together. Mine The content of the project’s requirement.txt is as follows. Django==1.10.5 rpy2==2.8.5 PyMySQL==0.7.9 numpy==1.13.0 pandas==0.20.2 scikit-learn==0.18.1 scipy==0.19.0 uwsgi 2. Change your settings.py to set the deployment environment for the Nginx server agent The changes are as follows STATIC_ROOT = os.path.join(BASE_DIR, ‘collect_static’) ALLOWED_HOST = [‘*’] Note: The deployment environment cannot be the same as the static files in the development environment, otherwise errors will occur. After modification, run the following command: python manage.py collectstatic Now you can safely compress files tar czvf FileName.tar DirName Building a web environment 1. Place the packaged project files into the web folder cp /home/haoyu/AAPlatform.tar . 2. Write Dockerfile This is mainly about building a basic environment #Basic image FROM ubuntu:16.10 FROM…
How to use Docker to build a Django, Nginx, R, and Python deployment environment
This article introduces how to use Docker to build a Django, Nginx, R, and Python deployment environment. I would like to share it with you. The details are as follows: Basic environment: Ubuntu 16.10 docker 17.06.0-ce Compress your own project files 1. What needs to be noted here is that when compressing, you also need to write the Python packages you need in requirement.txt, so that when you set up the environment, you will know what packages you need and can install them together. Mine The content of the project’s requirement.txt is as follows. Django==1.10.5 rpy2==2.8.5 PyMySQL==0.7.9 numpy==1.13.0 pandas==0.20.2 scikit-learn==0.18.1 scipy==0.19.0 uwsgi 2. Change your settings.py to set the deployment environment for the Nginx server agent The changes are as follows STATIC_ROOT = os.path.join(BASE_DIR, ‘collect_static’) ALLOWED_HOST = [‘*’] Note: The deployment environment cannot be the same as the static files in the development environment, otherwise errors will occur. After modification, run the following command: python manage.py collectstatic Now you can safely compress files tar czvf FileName.tar DirName Building a web environment 1. Place the packaged project files into the web folder cp /home/haoyu/AAPlatform.tar . 2. Write Dockerfile This is mainly about building a basic environment #Basic image FROM ubuntu:16.10 FROM…
Clear logs with Django, Nginx, Gunicorn and supervisor
I’m having issues with django login in my production environment. I’m using Nginx Gunicorn Supervisorctl to run my Django website. I’ve been trying everything for a few days with no results. My The idea is to have two log files, like the ones I have defined in my Django application. One to store all log records (messages.log) and another to store WARNING, ERROR and CRITICAL log records (errors.log) LOGGING = {‘version’: 1,’disable_existing_loggers’: True,’formatters’: { ‘ verbose’: { ‘format’: ‘ [%(asctime)s] [%(levelname)s] [%(name)s] %(message)s’ }, ‘simple’ : { ‘format’: ‘ %(levelname)s %(message)s’ },}, ‘handlers’: { ‘console’: { ‘level’: ‘DEBUG’, ‘class’: ‘logging.StreamHandler’, ‘formatter’: ‘verbose’ }, ‘file’: { ‘level’ : ‘DEBUG’, ‘class’: ‘logging.FileHandler’, ‘formatter’: ‘verbose’, ‘filename’: ‘/home/myuser/logs/messages.log’ }, ‘file_errors’: { ‘level’: ‘WARNING’, ‘class’: ‘logging.FileHandler’, ‘formatter’: ‘verbose’, ‘ filename’: ‘/home/myuser/logs/errors.log’ }, ‘mail_admins’: { ‘level’: ‘ERROR’, ‘class’: ‘django.utils .log.AdminEmailHandler’, ‘include_html’: True }},’loggers’: { ‘main’: { ‘handlers’: [‘console’ , ‘file’, ‘file_errors’, ‘mail_admins’], ‘level’: ‘DEBUG’ }, ‘caching’: { ‘handlers’: [‘console’, ‘ file’, ‘file_errors’], ‘level’: ‘DEBUG’ }} Using this logging configuration, I have To register logging in the django application, the command is as follows: logger = logging.getLogger(“main”)logger.info(“INFO message” )logger.critical(“CRITICAL message”) These commands register fine in messages.log and errors.log in my development environment. Now , it’s time to…
![javascript-Excuse me, what is the reason for this? [Network security, nginx, php, js related]](https://www.1024programmer.com/wp-content/themes/iux/thumb.php?src=/img/bVFhwJ?w=943&h=613&w=300&h=193)
javascript-Excuse me, what is the reason for this? [Network security, nginx, php, js related]
Solved, replaced jquery import with Qiniu Cloud static resource acceleration. Then this problem no longer occurs. It means there is a problem inside the js file. I also reminded myself not to use external js indiscriminately. Thank you to those who answered! After occasionally refreshing the page, normally it should all come from the same address. Why are there three things requested from external addresses? (The black lines above are all from the same address, and the yellow lines are from other addresses) Normally it should look like this: I am a front-end person, I don’t quite understand why? (Guess whether the router or server was hijacked?) 1. CDN is not used. 2.js references only jquery and layui. 3. It is just a login page and does not use js from external resources. 4. This is the page header: Reply content: Solved, replaced jquery import with Qiniu Cloud static resource acceleration. Then this problem no longer occurs. It means there is a problem inside the js file. I also reminded myself not to use external js indiscriminately. Thank you to those who answered! After occasionally refreshing the page, normally it should all come from the same address. Why are there…
Deploy java, mysql, redis, fastdfs, tracker, stroage, nginx, rabbitMq services on the centos7 system and build corresponding servers
centos system builds development environment and services Foreword The centos system must ensure that it can access the Internet 1. Install java-1.8.0 version yum install java- 1.8.0-openjdk* This command can successfully install java-1.8.0. Enter java -version to view the java version 2. Mysql57 is installed and supports remote connection Operation steps Command (the yellow part changes according to the actual environment) (1) Install wget yum -y install wget (2) Download the MySQL57 database installation program wget -i -c http://dev.mysql.com/get/ mysql57-community-release-el7-10.noarch.rpm (3 )Install database yum -y install mysql57-community-release-el7-10.noarch.rpm (4) Install database service yum -y install mysql-community-server (5) Start the database systemctl start mysqld.service (6) View the database initial password grep “temporary password” /var/log/mysqld.log, the original password of the database here is: *(bywl-2hDlP （ 7) Log in to the database with the initial password mysql -uroot -p (8) Reset database password alter user root@localhost identified by ‘password‘; If the password you set is too simple You need to change the password rules (9) Change password rules set global validate_password_policy=0; (10) Change password length limit set global validate_password_length=6;Changesuccessful! (11)Afterthechangesarecompleted,proceedtostep8 alteruserroot@localhostidentifiedby’password‘; （12）Setdatabaseremoteconnection GRANTALLPRIVILEGESON.TO’root’@’%’IDENTIFIEDBY’password‘WITHGRANTOPTION; (13)Turnoffthefirewall systemctlstopfirewalld.service (14)Permanentlyturnoffthefirewall systemctldisablefirewalld.service (15)ThatisRemotesoftwareconnectionisavailable Remotesoftwarenavicat,aftertheinstallationiscompleted,openitandyoucantesttheconnection 3. Install redis-4.0.8 (1) Run the following command in the /usr/local/ directory, or you can…
[Project] Deploy Java services Tomcat, Nginx, shell deployment script
Foreword The editor has just finished I have a task of deploying Java services. There are about three points that can be introduced: Tomcat service installation and modification configuration files, Nginx routing configuration, writing shell scripts to deploy services, and several related pitfalls encountered. Records Come down and share with everyone！ Technical points 1. Install Tomcat service， modify the configuration file 1. Install Tomcat service // Unzip the tomcat compressed packagetar -zxvf apache-tomcat-8.5.54.tar.gz//Rename the Tomcat service name：conveniently identify，multiple Tomcat services in one servermv apache-tomcat-8.5. 54 apache-tomcat-8.5.54_8040 2 .Modify the configuration file server.xml （1）Use the vim command to open the configuration file，Press i for editing vim apache-tomcat-8.5.54_8040/conf/server.xml （2）Modify the port number of the http request <Connector port=”8040″ protocol=”HTTP/1.1 “connectionTimeout=”20000″redirectPort& #61;”8443” URIEncoding= “UTF-8” /> 2. Nginx configuration，Add routing server{listen 80;server_name domain name;access_log /var/log/nginx/oapi-access.log main;root / home/appserver/apache-tomcat -8.5.54_8040/webapps/ROOT/;index index.html index.htm;location /oapi {charset utf-8;proxy_pass http://paas-oapi;proxy_set_header Host $ host;proxy_set_header X-Real-IP $remote_addr;proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;}} upstream paas-oapi{ #ip_hash;server 10.0.0 .66:8040;server 10.0.0.67:8040;} 3. Shell script #!/bin/bashtoday=`date ' ;+%Y%m%d'`mkdir -p /tmp/$today/new mkdir -p /tmp/$today/oldmkdir -p /tmp/$today/rbbaknewpath=/tmp/$today/new/oldpath=/tmp/$today /old/project=oapitomcat=/home/appserver/apache-tomcat-8.5.54_8040cd $newpathecho “downloading new war file…”wget http ://jk.hecom.cn/jobs-download/workspace/project-Install/paas/paas-oapi/paas-oapi-gateway/target/paas-oapi-gateway-1.0.war -O ${project}. warnewwar=`ls /tmp/$today/new/ ${project}.war 2>/dev/null | head -n 1` ;if [[ -f “$newwar” ]];thenecho “new war file downloaded OK!!!!!!”fiecho “deploying new war file…”backwar=&# 96;ls…
Deploy java, mysql, redis, fastdfs, tracker, stroage, nginx, rabbitMq services on the centos7 system and build corresponding servers
centos system builds development environment and services Foreword The centos system must ensure that it can access the Internet 1. Install java-1.8.0 version yum install java- 1.8.0-openjdk* This command can successfully install java-1.8.0. Enter java -version to view the java version 2. Mysql57 is installed and supports remote connection Operation steps Command (the yellow part changes according to the actual environment) (1) Install wget yum -y install wget (2) Download the MySQL57 database installation program wget -i -c http://dev.mysql.com/get/ mysql57-community-release-el7-10.noarch.rpm (3 )Install database yum -y install mysql57-community-release-el7-10.noarch.rpm (4) Install database service yum -y install mysql-community-server (5) Start the database systemctl start mysqld.service (6) View the database initial password grep “temporary password” /var/log/mysqld.log, the original password of the database here is: *(bywl-2hDlP （ 7) Log in to the database with the initial password mysql -uroot -p (8) Reset database password alter user root@localhost identified by ‘password‘; If the password you set is too simple You need to change the password rules (9) Change password rules set global validate_password_policy=0; (10) Change password length limit set global validate_password_length=6;Changesuccessful! (11)Afterthechangesarecompleted,proceedtostep8 alteruserroot@localhostidentifiedby’password‘; （12）Setdatabaseremoteconnection GRANTALLPRIVILEGESON.TO’root’@’%’IDENTIFIEDBY’password‘WITHGRANTOPTION; (13)Turnoffthefirewall systemctlstopfirewalld.service (14)Permanentlyturnoffthefirewall systemctldisablefirewalld.service (15)ThatisRemotesoftwareconnectionisavailable Remotesoftwarenavicat,aftertheinstallationiscompleted,openitandyoucantesttheconnection 3. Install redis-4.0.8 (1) Run the following command in the /usr/local/ directory, or you can…
Nginx, PHP5.5 and Ercon in OpenShift
It seems that cloud deployment environments are all the rage these years – from AWS to Google App Engine, Windows Azure to DigitalOcean, and more popping up every month. In all of these environments, it’s important to be able to deploy the software you need. In this tutorial, adapted from the step-by-step trial and error procedure of duythien, and with his full blessing, we will discuss the installation steps of Phaclon in another environment: OpenShift. The process is grueling – OpenShift doesn’t make it easy for us to deploy custom environments – but the rewards are worth it. Start Just one account with OpenShift. They offer a free tier and we will take advantage. As is typically the case with cloud environments, OpenShift has its own tools too. Follow the RHC Tools installation procedure. I chose “spphalcon” as the namespace for the final rhc setup step. At https://openshift.redhat.com/app/console/applications select “Create your first application now” and then select “Do it yourself” – the option should be at the bottom of the screen. Give the application a name – At this point, you will find that the namespace you previously defined in the High Court Rules settings is reflected in your application’s…
nginx,php log splitting
Author: zhanhailiang Date: 2014-01-06 By default nginx will only generate one access.log and one error.log, and they will accumulate every day. The log file will become very large. If you need to do some log analysis, whether you use script analysis or download the log locally Analysis is not convenient. There is also a similar need for php logs. So splitting access logs and error logs every day is helpful for analyzing the logs. #!/bin/sh # Scheduled splits every day. You can choose to run the corresponding version during off-peak periods. ngPid=’/usr/local/nginx/logs/nginx.pid’ phpFpmPid=’/usr/local/php/var/run/php-fpm.pid’ ngLogs=’/usr/local/nginx/logs’ # access.log error.log phpLogs=’/usr/local/php/log’ # errors.log slow.log saveLog=’/home/service/logs’ top=$saveLog/$(date +%Y)/$(date +%m) mkdir -p $top/nginx if [ -f $ngLogs/access.log ]; then mv $ngLogs/access.log $top/nginx/access.$(date +%Y%m%d).log fi if [ -f $ngLogs/error.log ]; then mv $ngLogs/error.log $top/nginx/error.$(date +%Y%m%d).log fi echo `cat $ngPid` kill -USR1 `cat $ngPid` # Reread the log files Reopen the log files mkdir -p $top/php if [ -f $phpLogs/errors.log ]; then mv $phpLogs/errors.log $top/php/errors.$(date +%Y%m%d).log fi if [ -f $phpLogs/php-fpm.log ]; then mv $phpLogs/php-fpm.log $top/php/php-fpm.$(date +%Y%m%d).log fi echo `cat $phpFpmPid` kill -USR1 `cat $phpFpmPid` # Restart the php log Reopen the log files chmod -R 777 $top
nginx, php-fpm default configuration and performance – TCPsocket or unixdomainsock
I saw a blog a few days ago, which mentioned that the server where PHP is located frequently creates TCP short connections under large concurrency conditions, and the 2MSL time of the server where it is located is too long, resulting in no ports available, and the system Unable to create TCP socket and a lot of errors reported. The solution given by the blogger later is to reduce the 2MSL time, clear the TCP connection in TIME_WAIT status as soon as possible, and recycle the port. I saw a blog a few days ago that mentioned that the server where PHP is located frequently creates TCP short connections under large concurrency conditions, and the 2MSL time of the server where it is located is too long, resulting in no ports available, and the system cannot create TCP sockets. And a lot of errors are reported. The solution given by the blogger later is to reduce the 2MSL time, clear the TCP connection in TIME_WAIT status as soon as possible, and recycle the port. At the same time, the reason for not using long connections is written at the end of the article, but is this really the best solution?…
