I suffered from a lot of errors when deploying to AWS, so I decided to put it together in an article. I hope it helps beginners like me!
I myself deployed using Ruby on Rails, VPC / EC2 (Nginx, Unicorn) / RDS (PostgreSQL) / Route53 / ALB. Articles referred to when deploying https://qiita.com/naoki_mochizuki/items/f795fe3e661a3349a7ce https://qiita.com/Yuki_Nagaoka/items/1f0b814e52e603613556
You can check the process by using the ps command. Articles that referred to the ps command https://eng-entrance.com/linux-command-ps
$ ps -x #Command to check the currently running process
When the application you are deploying displays we're sorry, but something went wrong.. Check if there is a problem with the app
Server environment/var/www/rails/app name
$ cd log
$ tail -n 30 production.log
If there is a description of the error, there is a problem with the application. If there is no particular error, there is a high possibility that there is a problem with the web server settings and the deployment has not been successful.
・ Check if nginx is running
Server environment
$ sudo systemctl status nginx
ʻActive: active (running) If it is displayed as `, it is running.
-Restart command
Server environment
$ sudo service nginx restart
-Check if there is an error in the nginx file
Server environment
$ sudo nginx -t
This is a convenient command that tells you where the error is if there is an error. If there is no problem with the nginx file, it will be displayed like this.
nginx: the configuration file /etc/nginx/nginx.conf syntax is ok nginx: configuration file /etc/nginx/nginx.conf test is successful
A command to check Unicorn logs. I think a common error is ʻAlready running on PID: (process number)`, which occurs when there are unwanted processes left.
Server environment/App directory
$ cat log/unicorn.log -n
Unicorn process confirmation command
Server environment
$ ps -ef | grep unicorn | grep -v grep
Command to terminate Unicorn process When the error of ʻAlready running on PID: (process number)` occurs, I think that the error can be solved by deleting unnecessary processes with this command.
Server environment
$ kill (Process number)
Command to check if PostgreSQL is running
If it says Active: active (running), it is running.
Server environment
$ systemctl status postgresql.service
Start PostgreSQL
Server environment
$ sudo systemctl enable postgresql
$ sudo systemctl start postgresql
When I accessed the application deployed with Google Chrome, I forcibly accessed https (when https conversion was not implemented) and the application did not open. The solution was to delete the history data by checking Chrome history → Delete browsing history data → Cached images and files.
You can check the response with the following command. If the web server file is error-free and the curl command returns a solid response, but redirects to https, this may be the problem.
$ curl http://example.com #http://example.com is the domain of the app.
Also, when connecting via http
Must be config.force_ssl = false.
config/environment/production.rb
$ config.force_ssl = false
At first, I thought about making it https by editing the file on the web server, but I knew that it would be easier to make it https using the service called ALB on AWS, so I used ALB. If you are a beginner and have trouble converting to https, I personally recommend using ALB. You can use ALB to redirect http requests to https.
[Articles used as a reference for https conversion] https://aws.amazon.com/jp/premiumsupport/knowledge-center/elb-redirect-http-to-https-using-alb/ https://dev.classmethod.jp/articles/alb-redirects/ In addition to this, if you search for "ALB https redirect", you will find many useful articles!
Recommended Posts